# Remediation vs Mitigation # Remediation vs mitigation in cybersecurity: a five-question tool that picks the response for one finding Canonical: https://remediationvsmitigation.com/ Remediation vs mitigation in cybersecurity comes down to what is left afterward: remediation eliminates the vulnerability, mitigation leaves it in place and lowers exposure or impact. The tool below takes five answers about one finding and returns remediate, mitigate now and fix later, or accept with a dated exception, with the record to keep. Read 19 Nov 2026. ## The decision tool 1. Does a vendor fix exist today? (yes, no, not sure) 2. Is the vulnerable component reachable from the internet? (yes, no, not sure) 3. Is the CVE on the CISA Known Exploited Vulnerabilities catalog? (yes, no, not sure) 4. Can the fix go in before its due date without breaking a critical system? (yes, no, not sure) 5. Is a control already in front of the flaw, and has someone tested it? (yes, no, not sure) ## The source behind each of the five questions The tool takes one finding at a time. It runs in the browser and sends nothing. On load, with no answers, each question reads cautiously, Rule 1 matches, and the tool shows Mitigate now, fix later. The questions above are the ones published guidance asks, in the order that settles most findings fastest; each is listed here with its source. 1. **Question 1, vendor fix.** NIST SP 800-40 Rev. 4 lists "A patch may not be available yet" as a reason a patch cannot go in immediately ([NIST SP 800-40 Rev. 4](https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf)). 2. **Question 2, reachability.** Exposure is one of four inputs to CISA's BOD 26-04 Table 1, with KEV status, whether an adversary can automate exploitation, and technical impact ([BOD 26-04](https://www.cisa.gov/news-events/directives/bod-26-04-prioritizing-security-updates-based-risk)). 3. **Question 3, KEV status.** BOD 26-04 states that "Known exploited vulnerabilities are a frequent attack vector for malicious cyber actors", and makes KEV listing one of its four inputs; the clock starts when CISA adds the entry. The feed held 1,786 entries in catalog version 2026.11.18, read 19 Nov 2026 ([BOD 26-04](https://www.cisa.gov/news-events/directives/bod-26-04-prioritizing-security-updates-based-risk); [KEV feed](https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json)). 4. **Question 4, fix in time.** SP 800-40 Rev. 4, Section 2.1, accepts waiting for an outage window, testing first, or updating interacting software ([NIST SP 800-40 Rev. 4](https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf)). 5. **Question 5, tested control.** SP 800-40 Rev. 4, Section 2.2, step 3c, says to "ensure they are functioning as intended" when additional controls are deployed ([NIST SP 800-40 Rev. 4](https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf)). The answers land on 6 of the 13 situations in the [Decision Table](https://remediationvsmitigation.com/decision-table/): no fix yet, a KEV item that can be fixed in time, a non-KEV item that can, a control already in front, an unreachable flaw outside the KEV catalog, and a patch that breaks a critical system. The other seven are reached from the Decision Table directly: end-of-life software, SaaS that cannot be patched, a scheduled decommission, a PCI requirement that cannot be met, an OT or medical device, a federal agency that cannot meet a KEV date, and a false positive. The tool's usual answer is wrong in three cases. It does not ask whether exploitation can be automated, which BOD 26-04 treats as an input, so a non-KEV flaw with a public automated exploit needs a faster response than the tool returns. It does not ask whether you are a federal civilian agency, for whom the KEV date is binding. It does not ask what the product is, so end-of-life, SaaS and OT cases need your own reading of the matching rows in the Decision Table, listed at the foot of the rules table below. The rules table says where each rule assumes an answer. The method page explains how the answers pick a row. ## One finding worked through Four entries were added to the KEV catalog on 18 Nov 2026: Synology DiskStation Manager, BerriAI LiteLLM, Vanderbilt REDCap and Apache Axis. CVE-2024-10441, in Synology DiskStation Manager, has a due date of 21 Nov 2026, three days later, with forensic triage marked Yes. CVE-2026-40217, in BerriAI LiteLLM, has a due date of 2 Dec 2026, 14 days later, with forensic triage marked No. The REDCap and Axis entries are also due 21 Nov 2026 ([KEV feed](https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json), catalog version 2026.11.18). The walk-through takes the Synology entry. The feed answers question 3; the other answers are assumptions that an asset owner replaces with facts. Walk-through for CVE-2024-10441. Only the KEV facts come from the feed, read 19 Nov 2026. | Question | Answer used | Basis | |---|---|---| | 1. Vendor fix exists? | Yes (assumed) | The entry's notes link Synology advisory Synology_SA_24_20; the owner confirms a fixed release exists | | 2. Reachable from the internet? | Yes (assumed) | The owner's architecture drawing | | 3. On the KEV catalog? | Yes | Added 18 Nov 2026, due 21 Nov 2026 | | 4. Fix in before the due date? | No (assumed) | 21 Nov 2026 is a Saturday; the next change window is assumed to be Tuesday 24 Nov 2026 | | 5. Tested control in front? | Yes (assumed) | A tested access restriction on the management interface | Rule 5 matches (fix yes, date missed, tested control yes), so the result is Mitigate now, fix later, on the row for a [control already in front](https://remediationvsmitigation.com/decision-table/compensating-control-in-front/). The control goes into the record with its test date, the hosts it covers, the approver, and the date it comes off, which is after the fix lands on 24 Nov 2026. Had the control been untested, rule 7 would return the row for a patch that breaks a critical system, with the same response. The entry's required action is to apply mitigations per vendor instructions "or discontinue use of the product if mitigations are unavailable". For a federal civilian agency the date stays 21 Nov 2026, and its KEV date row applies, not rule 5: CISA "does not issue waivers or exceptions for actions required in cyber directives" ([implementation guidance](https://www.cisa.gov/news-events/directives/bod-26-04-implementation-guidance-prioritizing-security-updates-based-risk)). Outside that scope the KEV date is guidance, and FedRAMP providers have a separate rule, quoted in the section on the primary sources below. The dates page sets this entry's clocks side by side. ## The three responses and the record to keep for each Built from the evidence lists in the Compensating Controls Register and the fields of the Risk Acceptance Form. Read 19 Nov 2026. | Response | What it means | Record to keep | |---|---|---| | Remediate | Eliminate the vulnerability through patching, decommissioning the system, or another action (BOD 26-04 wording). | Fixed version and date, the scan or test that no longer finds the flaw, the change ticket. | | Mitigate now, fix later | A control reduces exposure or impact while the flaw stays in place. | Control and mode (block or log-only), test date and tester, hosts and routes covered, approver, the date the control comes off and the date the fix lands. | | Accept with a dated exception | The residual risk stays under a signed decision with an end date. | Reason, residual risk against stated tolerance, controls in place, approver and date, expiry date, the event that reopens it. | The mitigate record is the one most often missing a date. NIST SP 800-40 Rev. 4, Section 3.5.3, says "Schedules should be set and enforced for both patch deployment and mitigation removal." The Compensating Controls Register lists the evidence to keep for each of 11 controls. The [Risk Acceptance Form](https://remediationvsmitigation.com/risk-acceptance-form/) holds the 15 fields of an exception, and the [exception register](https://remediationvsmitigation.com/exception-register/) tracks them to expiry. ## Remediation and mitigation in the primary sources NIST's glossary gives mitigation as "A decision, action, or practice intended to reduce the level of risk associated with one or more threat events, threat scenarios, or vulnerabilities" (SP 800-160 Vol. 2 Rev. 1) and as "The temporary reduction or lessening of the impact of a vulnerability or the likelihood of its exploitation" (SP 800-216) ([NIST glossary, mitigation](https://csrc.nist.gov/glossary/term/mitigation)). Mitigation lowers risk and leaves the flaw in place; remediation removes it. The [glossary](https://remediationvsmitigation.com/glossary/) has both wordings. CISA's BOD 26-04 revoked BOD 22-01 on 10 Jun 2026. It sets 3, 14 or 60 calendar days, or "fix on system upgrade", from exposure, KEV status, automation and technical impact, and defines remediation as the vulnerability being "eliminated through patching, decommissioning the system, or another action" ([BOD 26-04](https://www.cisa.gov/news-events/directives/bod-26-04-prioritizing-security-updates-based-risk)). FedRAMP's 2026 rules go further on vocabulary: "A fully mitigated vulnerability will still exist (with negligible risk) until it has been remediated." Providers should remediate KEV entries by the CISA due date even when fully mitigated, and the rules are required for Rev5 certification from 7 Dec 2026 ([FedRAMP VDR](https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-detection-and-response/)). The [difference guide](https://remediationvsmitigation.com/guides/difference-between-remediation-and-mitigation/) sets the definitions of each source side by side. ## What the tool cannot settle Some decisions belong to a named party. In PCI DSS, "QSAs are responsible to determine whether a compensating control is sufficient to meet the intent of a requirement" ([PCI SSC FAQ 1046](https://www.pcisecuritystandards.org/faqs/1046)). PCI SSC also separates two words: critical vulnerabilities under Requirement 6.3.3 "must be resolved within one month of release", while a vulnerability that is "addressed" may be handled with a compensating control or by disabling a vulnerable service ([PCI SSC FAQ 1597](https://www.pcisecuritystandards.org/faqs/1597)). In federal systems, "the authorizing official is the only person who can accept risk" ([NIST SP 800-37 Rev. 2](https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-37r2.pdf), Task R-3). How long an acceptance may run is on the [maximum durations page](https://remediationvsmitigation.com/risk-acceptance-form/maximum-durations/), and who may sign is on the [who may sign page](https://remediationvsmitigation.com/risk-acceptance-form/who-may-sign/). ## How the five answers pick a row Rules are tried in order and the first match wins. Vendor fix exists: not sure is read as no; Internet-reachable: not sure is read as yes; On the KEV catalog: not sure is read as yes; Fix in before due date: not sure is read as no; Tested control in front: not sure is read as no. | Rule | Vendor fix exists | Internet-reachable | On the KEV catalog | Fix in before due date | Tested control in front | Decision Table row | Guidance points to | Assumes | |---|---|---|---|---|---|---|---|---| | 1 | No | Any | Any | Any | Any | kev-no-patch-yet | Mitigate now, fix later | Nothing else is asked once there is no fix. The row is written for an exploited or severe flaw (its KEV example). For a finding outside the KEV catalog the response, mitigate now and fix later, is the same, but the row says nothing about a deadline: set the interim control and a dated plan from the finding's own exposure and impact, and record that it is not a KEV entry. | | 2 | Yes | Any | Yes | Yes | Any | kev-patch-available-exposed | Remediate | The row is written for an asset reachable from the internet. Not reachable gives the same response, remediate; BOD 26-04 Table 1 sets that clock. | | 3 | Yes | No | No | Yes | Any | internal-asset-not-kev-moderate | Remediate | Assumes exploitation is automatable, which the tool does not ask. If it is not, Table 1 gives fix on system upgrade. | | 4 | Yes | Yes | No | Yes | Any | kev-patch-available-exposed | Remediate | No row covers a reachable flaw outside the KEV catalog. The nearest row is written for a KEV entry; the response is the same, remediate, and Table 1 sets the clock. | | 5 | Yes | Any | Any | No | Yes | compensating-control-in-front | Mitigate now, fix later | Matches whatever the KEV and reachability answers are. | | 6 | Yes | No | No | No | Any | not-reachable-low-risk | Accept with a dated exception | Assumes an adversary cannot automate exploitation, which the tool does not ask. | | 7 | Yes | Any | Any | No | Any | patch-breaks-critical-system | Mitigate now, fix later | Catches a missed change window with no tested control: the row lists "an outage window is not available yet" as a signal, so the closest row is the patch-breaks-a-critical-system row, which has the same response, mitigate now and fix later. A federal civilian agency with a KEV date it cannot meet goes to the KEV date row in the Decision Table; the tool does not ask who you are. | Rows the five questions do not reach: end-of-life-software; saas-cannot-patch; decommission-scheduled; pci-cannot-meet-requirement; ot-medical-cannot-patch; kev-due-date-cannot-be-met; false-positive. ## Sources - CISA, BOD 26-04, Table 1 and Appendix A: https://www.cisa.gov/news-events/directives/bod-26-04-prioritizing-security-updates-based-risk (read 19 Nov 2026) - CISA, BOD 26-04 implementation guidance, FAQ: https://www.cisa.gov/news-events/directives/bod-26-04-implementation-guidance-prioritizing-security-updates-based-risk (read 19 Nov 2026) - FedRAMP, Consolidated Rules 2026, Vulnerability Detection and Response: https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-detection-and-response/ (read 19 Nov 2026) - NIST SP 800-40 Rev. 4, Sections 2.1, 2.2, 3.5.3: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf (read 19 Nov 2026) - NIST SP 800-37 Rev. 2, Task R-3: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-37r2.pdf (read 19 Nov 2026) - CISA KEV catalog JSON feed, catalogVersion 2026.11.18: https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) - PCI SSC FAQ 1046: https://www.pcisecuritystandards.org/faqs/1046 (read 19 Nov 2026) - PCI SSC FAQ 1597: https://www.pcisecuritystandards.org/faqs/1597 (read 19 Nov 2026) - NIST CSRC glossary, mitigation: https://csrc.nist.gov/glossary/term/mitigation (read 19 Nov 2026) Last reviewed 19 Nov 2026 --- # About Remediation vs Mitigation Canonical: https://remediationvsmitigation.com/about/ Remediation vs Mitigation is a reference on one choice about a security finding: fix it, reduce it for now, or accept it with a date. First edition 19 Nov 2026. It holds 13 situations, 9 controls and a 15-field risk acceptance form, all read from sources on that date. ## What is on the site - The [Decision Table](https://remediationvsmitigation.com/decision-table/): 13 situations, each with the response the sources point to and the quotes behind it. - Nine compensating control entries, each with what it changes, what it does not fix, how to verify it and the evidence to keep. - The [Risk Acceptance Form](https://remediationvsmitigation.com/risk-acceptance-form/) and the [exception register](https://remediationvsmitigation.com/exception-register/): 15 fields, who may sign, and the longest durations any source sets. - Guides that answer single questions, such as the [difference between remediation and mitigation](https://remediationvsmitigation.com/guides/difference-between-remediation-and-mitigation/), [workaround vs mitigation](https://remediationvsmitigation.com/guides/workaround-vs-mitigation/) and what to do when no patch exists. The site is about security findings only: a vulnerability or weakness found in software, a device or a service. It does not cover environmental cleanup, building work or any other meaning of the two words. ## What it leaves out It does not compare products in a list order, explain how vulnerabilities get severity numbers, or teach vulnerability management as a program. Those subjects have their own references. The site also gives no exploit code, proof-of-concept steps or payloads; a control is described by what it changes and how to verify it, never by how to attack the flaw. ## How it is hosted The site is a set of static files stored on Amazon S3 and served by Amazon CloudFront. It sets no cookies, runs no analytics and loads no third-party scripts. The decision tool and the risk acceptance form run in the browser and send nothing to a server. CloudFront access logs go to a private bucket and are deleted after 400 days; the [privacy page](https://remediationvsmitigation.com/privacy/) lists the fields. Data and forms are published under [Creative Commons Attribution 4.0](https://creativecommons.org/licenses/by/4.0/) (CC BY 4.0). Cite as: Remediation vs Mitigation, "", first edition 19 Nov 2026, remediationvsmitigation.com. Quoted passages from standards stay under their issuers' terms; see the [terms](https://remediationvsmitigation.com/terms/). ## Method and contact Every fact carries a source URL and the read date 19 Nov 2026. The [editorial policy](https://remediationvsmitigation.com/editorial-policy/) sets the rules for sources and wording. Questions and corrections go to [editor@remediationvsmitigation.com](mailto:editor@remediationvsmitigation.com); the [contact page](https://remediationvsmitigation.com/contact/) says what to include. ## Sources - AWS, CloudFront standard logging reference: https://docs.aws.amazon.com/AmazonCloudFront/latest/DeveloperGuide/standard-logs-reference.html (read 19 Nov 2026) - Creative Commons, Attribution 4.0 International deed: https://creativecommons.org/licenses/by/4.0/ (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Editorial Policy: Sources, Labels and Corrections](https://remediationvsmitigation.com/editorial-policy/index.md) - [Privacy: Cookies, Logs and the 400-Day Limit](https://remediationvsmitigation.com/privacy/index.md) - [Decision Table: Remediate, Mitigate or Accept by Situation](https://remediationvsmitigation.com/decision-table/index.md) --- # Access restriction as a compensating control Canonical: https://remediationvsmitigation.com/compensating-controls/access-restriction/ Access restriction narrows who can reach a flaw without removing it. For CVE-2026-0322 (PAN-OS, added to KEV 14 Oct 2026), Palo Alto says limiting the management interface to one jump box moves its CVSS 4.0 base score only from 8.6 to 8.5, so the control buys time, not safety. Read 19 Nov 2026. ## What it covers Access restriction limits which identities and addresses may call a component, including on a flat network. [Network segmentation](https://remediationvsmitigation.com/compensating-controls/network-segmentation/) limits which networks reach it. NIST states the principle at three layers (SP 800-53 Rev. 5.2.0, read 19 Nov 2026): NIST SP 800-53 Rev. 5.2.0 control statements, read 19 Nov 2026. | Control | Layer | Text | |---|---|---| | AC-3 | Logical access | "Enforce approved authorizations for logical access to information and system resources in accordance with applicable access control policies." | | AC-6 | Privilege | "Employ the principle of least privilege, allowing only authorized accesses for users (or processes acting on behalf of users) that are necessary to accomplish assigned organizational tasks." | | SC-7(5) | Network | "Deny network communications traffic by default and allow network communications traffic by exception" | ## When it counts as a mitigation and when it does not Situations drawn from the CVE records and CISA's implementation guidance, read 19 Nov 2026. | Situation | Counts? | Why | |---|---|---| | The flaw sits on a management interface and needs a valid high-privilege account (CVE-2026-0322) | Yes, in part | It cuts who can reach the interface. Administrators and the jump box can still reach it, so the flaw is reachable by them. | | The flaw is in an API and needs no login (CVE-2026-76460, Cisco ISE) | Yes, as far as the list is short | Cisco calls infrastructure ACLs that allow only required management and control plane traffic a mitigation. Any source on the list can still send the request. | | Any authorized caller can trigger the flaw, or the allow-list already covers most of the user base | No | The list no longer separates attackers from users. | | A federal agency has a KEV deadline under BOD 26-04 | Not for the deadline | CISA's guidance says that aside from removing assets from the network, the only known technical mitigation for a KEV entry is the required action in the catalog. Restricting a list moves no input unless it removes public exposure. | | An allowed host or credential is already compromised | No | The attacker is inside the list. | ## The named case: CVE-2026-0322 on PAN-OS Palo Alto Networks published its advisory on 14 Oct 2026. The flaw is code injection (CWE-94) in the PAN-OS management interface: an attacker with network access and a valid account can run code as root on the firewall. It affects PA-Series, VM-Series and Panorama, and "No special configuration is required to be affected". The vendor lists no workaround, and says "The security risk posed by this issue is minimized by restricting access to the management interface to only trusted internal IP addresses". It gives a stronger form in the severity block: "You can reduce the risk of exploitation by restricting access to a jump box that is the only system allowed to access the management interface." The vendor scale reads 8.6 (CVSS:4.0/AV:N) before and 8.5 (AV:A) after. Only the attack vector changes from network to adjacent. The need for a high-privilege account stays, and so does the impact. CISA added the CVE to KEV on 14 Oct 2026 with a due date of 17 Oct 2026 and marked forensic triage as required ([KEV feed](https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json), catalog 2026.11.18). Fixed builds by train, from Palo Alto's advisory as recorded on 19 Nov 2026: Palo Alto Networks advisory CVE-2026-0322 (published 14 Oct 2026), read 19 Nov 2026. Older unsupported versions are affected and have no listed fix. Cloud NGFW and Prisma Access are not affected. | PAN-OS train | First fixed builds (per sub-train) | |---|---| | 12.2 | 12.2.4 | | 12.1 | 12.1.4-h12, 12.1.7-h8, 12.1.10-h2 | | 11.2 | 11.2.4-h22, 11.2.7-h21, 11.2.10-h15, 11.2.13-h3, 11.2.16 | | 11.1 | 11.1.4-h37, 11.1.6-h39, 11.1.7-h11, 11.1.10-h35, 11.1.13-h15, 11.1.16-h3 | | 10.2 | 10.2.7-h38, 10.2.10-h41, 10.2.13-h26, 10.2.16-h11, 10.2.18-h12 | The setting is the permitted IP list of the management interface (Device > Setup > Interfaces > Management Interface Settings > Permitted IP Addresses). A second path is easy to miss: any data-plane interface whose management profile allows HTTPS or SSH (Network > Network Profiles > Interface Mgmt) also exposes the interface. Both places have to be read. Cisco's advisory for ISE (CVE-2026-76460, 16 Sep 2026) names the same step for a different product: infrastructure ACLs that allow only required management and control plane traffic. CISA added it to KEV the same day, due 19 Sep 2026. The case is worked on the [network segmentation page](https://remediationvsmitigation.com/compensating-controls/network-segmentation/). ## A worked case with dates and counts The estate below is illustrative; the advisory, KEV and build facts are real. One firewall runs PAN-OS 11.2.10-h12, inside the affected range, and the fixed build is 11.2.10-h15. 1. Wed 14 Oct 2026: advisory and KEV entry. The KEV due date is Sat 17 Oct 2026. The change window for the upgrade is 28 Oct 2026, 14 days out. 2. Read the permitted IP list. It holds 3 entries: an admin VLAN 10.20.0.0/24 (256 addresses), a VPN pool 10.60.0.0/22 (1,024 addresses) and one contractor address. That is 1,281 addresses allowed to reach the interface. 3. Replace the 3 entries with the jump host 10.20.0.15. The list now holds 1 address. Remove HTTPS and SSH from the two data-plane management profiles that allowed them. 4. Record the vendor figure: 8.6 before, 8.5 after. State in the entry that this is a reduction of 0.1 and that the flaw stays reachable from the jump host. 5. Set the end date at 28 Oct 2026, the upgrade day, and a mid-point review on 21 Oct 2026. After the upgrade, put the list back to the approved standard. For an agency, none of this meets the KEV action by 17 Oct 2026, and forensic triage is still required. For other organizations, the entry is the dated record that supports an exception until 28 Oct. Neither case treats the jump box as a fix. ## Check that it took 1. Read the permitted IP list and every Interface Mgmt profile back from the device and compare with the approved list. 2. From a host your team owns that is not on the list, confirm the management address does not answer. Stop at the refusal. 3. List who holds the jump host and what else it can reach, since the jump host is now the single path. 4. Confirm service accounts, break-glass accounts and automation are on the approved list, since they are the ones most often left out. 5. Repeat after every firewall or routing change. ## What it costs Administrators must connect through the jump host or from the listed addresses, and any script or tool that manages the device from elsewhere stops working. The list also decays. Entries added during an incident and never removed turn a one-host list into a subnet. The jump host becomes a target and needs its own patching and logging. ## Record to keep # Access restriction record - Finding: CVE-2026-0322, PAN-OS management interface (Palo Alto advisory 14 Oct 2026) - Asset and build: [hostname], PAN-OS [build]; fixed build [build] - Control applied: management interface limited to [jump host]; data-plane management profiles [changed or none] - Applied by / date: [name], [date]; ticket [ID]; approver [name] - Before: [number] entries, [number] addresses - After: [number] entries, [number] addresses - Check on [date]: list and profiles read back; refusal confirmed from [test host] - Vendor figure: [before] to [after]; flaw still reachable from [who] - KEV due date and whether it is met: [date], [yes or no]; forensic triage: [status] - End date: [date] (upgrade); review date: [date] - Removal step: restore approved list after upgrade, owner [name] - Next list review: [date], owner [name] ## Sources - Palo Alto Networks security advisory CVE-2026-0322 (published 2026-10-14): https://security.paloaltonetworks.com/CVE-2026-0322 (read 19 Nov 2026) - CISA KEV feed, catalog 2026.11.18: CVE-2026-0322 and CVE-2026-76460: https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) - Cisco advisory cisco-sa-ISE-ABP-VNSW7Tn5 (16 Sep 2026): https://sec.cloudapps.cisco.com/security/center/content/CiscoSecurityAdvisory/cisco-sa-ISE-ABP-VNSW7Tn5 (read 19 Nov 2026) - CISA, BOD 26-04 implementation guidance, alternative mitigation FAQ: https://www.cisa.gov/news-events/directives/bod-26-04-implementation-guidance-prioritizing-security-updates-based-risk (read 19 Nov 2026) - NIST SP 800-53 Rev. 5.2.0, AC-3: https://csrc.nist.gov/projects/cprt/catalog#/cprt/framework/version/SP_800_53_5_2_0/home/AC-3 (read 19 Nov 2026) - NIST SP 800-53 Rev. 5.2.0, AC-6: https://csrc.nist.gov/projects/cprt/catalog#/cprt/framework/version/SP_800_53_5_2_0/home/AC-6 (read 19 Nov 2026) - NIST SP 800-53 Rev. 5.2.0, SC-7(5): https://csrc.nist.gov/projects/cprt/catalog#/cprt/framework/version/SP_800_53_5_2_0/home/SC-7.5 (read 19 Nov 2026) - NIST OSCAL catalog of SP 800-53 Rev. 5.2.0 (last modified 2026-10-22), control text: https://github.com/usnistgov/oscal-content/blob/main/nist.gov/SP800-53/rev5/json/NIST_SP-800-53_rev5_catalog.json (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Network Segmentation as a Mitigation: Cisco ISE Case](https://remediationvsmitigation.com/compensating-controls/network-segmentation/index.md) - [Remove Internet Exposure as a Mitigation: What It Buys](https://remediationvsmitigation.com/compensating-controls/remove-internet-exposure/index.md) - [KEV Required Action: Mitigate or Discontinue Use](https://remediationvsmitigation.com/guides/kev-required-actions-mitigate-or-discontinue/index.md) - [How to Verify a Mitigation Works: Checks for 5 Real CVEs](https://remediationvsmitigation.com/guides/verifying-a-mitigation-works/index.md) --- # Application allow-listing as a compensating control Canonical: https://remediationvsmitigation.com/compensating-controls/application-allow-listing/ No Microsoft advisory or CISA text read on 19 Nov 2026 names application allow-listing as a mitigation for any CVE in this site's data. Its use against CVE-2026-68820 (fixed 11 Aug 2026) is an inference from that flaw's precondition: it helps only when the exploit must start a program the policy has not approved. ## What the control is Microsoft describes application control as a change of default: Windows goes "from a place where all code runs unless your AV solution confidently predicts it's bad, to one where code runs only if your policy says so" ([Microsoft Learn, App Control for Business](https://learn.microsoft.com/en-us/windows/security/application-security/application-control/app-control-for-business/appcontrol)). It adds that application control is not a replacement for antivirus. NIST SP 800-53 Rev. 5.2.0 enhancement CM-7(5), Authorized Software, allow-by-exception, has three parts (read 19 Nov 2026): NIST SP 800-53 Rev. 5.2.0, CM-7(5), OSCAL catalog, read 19 Nov 2026. | Part | What it asks | Text | |---|---|---| | a | Name the authorized software | "Identify" the programs, with an organization-defined scope. NIST's discussion says the list can be limited "to specific versions or from a specific source". | | b | Deny by default | "Employ a deny-all, permit-by-exception policy to allow the execution of authorized software programs on the system;" | | c | Keep the list current | "Review and update the list of authorized software programs" on an organization-defined schedule. | Unlike a [WAF rule](https://remediationvsmitigation.com/compensating-controls/waf-rule/) or an [IPS signature](https://remediationvsmitigation.com/compensating-controls/ips-signature/), it does not depend on recognizing the attack. It limits what an attacker can start afterward. ## When it counts as a mitigation and when it does not The test is where the attacker's code runs. If the exploit needs a new program to start, a deny-by-default policy can refuse it. If the flaw is reached from inside a program that is already approved, nothing new starts and the policy sees nothing to block. Preconditions from the CVE records read 19 Nov 2026. | Flaw | Where the attacker's code runs | Does allow-listing help? | |---|---|---| | CVE-2026-68820, Windows afd.sys use-after-free (Microsoft, 11 Aug 2026) | A locally authorized attacker runs a crafted application | Yes, if enforced on the host and the attacker has no approved program that runs arbitrary code. | | CVE-2026-85880, Windows ALPC heap overflow (Microsoft, 8 Sep 2026) | Code already running in a low-privilege AppContainer, such as a compromised browser renderer | No. The code is inside an approved process. | | CVE-2026-87491, Chromium V8 out-of-bounds write (Google, 8 Sep 2026) | Inside the browser sandbox, from a crafted page | No. The browser is approved and runs the page. | | CVE-2026-65660, SharePoint Server code injection (KEV 25 Sep 2026) | The server application, from a network request by a low-privilege account | No. The vulnerable program is approved and is meant to run. | ## CVE-2026-68820: an inference, not a published mitigation No vendor or CISA entry names application allow-listing as a mitigation: not in the advisories for the four CVEs above, and not in the required actions of the 1,786 entries of the KEV feed of 18 Nov 2026. The use below is this site's inference from the preconditions of CVE-2026-68820, not Microsoft's or CISA's text. Microsoft published the fix on 11 Aug 2026 and marked the flaw exploited that day. CISA added it to KEV the same day with a due date of 25 Aug 2026, a 14-day window, and listed forensic triage as not required. Microsoft lists no workaround. The inference rests on one fact. Microsoft's record says the flaw "allows an authorized attacker to elevate privileges locally" and that "A locally authenticated attacker could run a specially crafted application on an affected system to trigger a race condition". On a host that enforces a deny-by-default policy, that application is not approved. A host where the policy only logs, or where the attacker can use an approved program that runs arbitrary code (next section), gets nothing. The fix is the August 2026 cumulative update for the Windows version in use: Microsoft MSRC CVRF, August 2026 release, CVE-2026-68820, read 19 Nov 2026. Server Core installations take the same update. Where two updates are listed they are the two August 2026 packages Microsoft maps to the same product. Windows Server 2025 and Windows 11 24H2 share the 26100 base but are serviced by different packages with different revisions, as the CVRF shows, so match a host's full build string to the row for its own product. | Windows version | Update | Fixed build | |---|---|---| | Windows Server 2012 | KB5120386 | 6.2.9200.26280 | | Windows Server 2012 R2 | KB5120385 | 6.3.9600.23338 | | Windows 10 1607, Server 2016 | KB5120418 | 10.0.14393.9418 | | Windows 10 1809, Server 2019 | KB5120238 | 10.0.17763.9121 | | Windows 10 21H2 | KB5120249 | 10.0.19044.7663 | | Windows 10 22H2 | KB5120249 | 10.0.19045.7663 | | Windows Server 2022 | KB5120242 or KB5120229 | 10.0.20348.5499 or 10.0.20348.5440 | | Windows 11 23H2 | KB5120240 | 10.0.22631.7517 | | Windows 11 24H2 | KB5121003 or KB5120994 | 10.0.26100.9168 or 10.0.26100.9106 | | Windows 11 25H2 | KB5121003 or KB5120994 | 10.0.26200.9168 or 10.0.26200.9106 | | Windows 11 26H1 | KB5121000 | 10.0.28000.2704 | | Windows Server 2025 | KB5120233 or KB5120228 | 10.0.26100.33296 or 10.0.26100.33222 | Constructed example, not taken from a record: 12 servers carry the finding, 9 with the policy enforced and 3 audit-only. The dates: - 11 Aug 2026: Microsoft publishes the fix and CISA adds the flaw to KEV. This is the finding date. The KEV due date is 25 Aug 2026. - 12 Aug: the endpoint team reads the policy mode on all 12 servers and finds 3 audit-only and 9 enforced. - 13 Aug: the security tester runs the unapproved test program on a pilot server and sees event 3077. The endpoint team lists which of Microsoft's bypass files the policy allows: none on 7 of the enforced servers, msbuild.exe on 2. - 14 Aug: the 3 audit-only servers are patched, 11 days inside the window. They need no exception. - 22 Aug: the 7 enforced servers with no allowed bypass file are patched in the weekly window, 3 days before the due date. - 29 Aug: the 2 enforced servers that allow msbuild.exe are patched after a change freeze, 4 days after the due date. Their exception runs from 11 Aug to 29 Aug, which covers the 14-day window and the 4 days after the due date, and says the control does not cover msbuild.exe. For a federal agency, BOD 26-04 would not count the control as moving the date. ## Approved programs that can run arbitrary code A deny-by-default policy fails for this flaw if the attacker can start the crafted code through a program the policy already allows. Microsoft states the risk: "Microsoft has identified a list of valid applications that an attacker could also potentially use to bypass App Control." Unless the use case requires them, it recommends blocking these (the list names files; check each against the host's allow rules) ([Microsoft Learn, applications that can bypass App Control](https://learn.microsoft.com/en-us/windows/security/application-security/application-control/app-control-for-business/design/applications-that-can-bypass-appcontrol)): File names from Microsoft Learn, read 19 Nov 2026. Microsoft notes: allow msbuild.exe only on a development reference system; if the WebDAV DLLs are blocked, also disable the WebClient service. | Group | Files on Microsoft's list | |---|---| | Script hosts and shells | cscript.exe, wscript.exe, mshta.exe, bash.exe, wmic.exe | | Build and compile tools | msbuild.exe, msbuild.dll, Microsoft.Build.dll, aspnet_compiler.exe, Microsoft.Workflow.Compiler.exe, csi.exe, rcsi.exe, fsi.exe, fsiAnyCpu.exe, dotnet.exe, dnx.exe, texttransform.exe | | Debuggers | cdb.exe, kd.exe, ntkd.exe, ntsd.exe, windbg.exe, dbghost.exe, dbgsvc.exe, dbgsrv.exe | | Windows Subsystem for Linux | wsl.exe, wslconfig.exe, wslhost.exe, lxrun.exe, lxssmanager.dll | | Other | addinprocess.exe, addinprocess32.exe, addinutil.exe, bginfo.exe (before 4.22), infdefaultinstall.exe, kill.exe, powershellcustomhost.exe, runscripthelper.exe, visualuiaverifynative.exe, wfc.exe, system.management.automation.dll, webclnt.dll and davsvc.dll | For each in-scope host, list which of these files the active policy allows. Any allowed entry means the host's row in the record is a dated exception, not a closure. ## Reading enforced or audit mode on a host Microsoft Learn, select types of rules to create, understanding App Control event IDs, and using Event Viewer with AppLocker, read 19 Nov 2026. App Control events for MSI and scripts are in the AppLocker, MSI and Script log. | Product | Where the mode shows | Audit mode (nothing blocked) | Enforced (blocks) | |---|---|---|---| | App Control for Business | Policy rule option 3, Enabled:Audit Mode. Microsoft: "When the Enabled:Audit Mode is deleted, the policy runs in enforced mode." | Event 3076 in Applications and Services Logs, Microsoft, Windows, CodeIntegrity, Operational: the file would have been blocked if the policy was enforced | Event 3077 in the same log: the file did not pass the policy and was blocked | | AppLocker | Enforcement mode of each rule collection: Audit only or Enforce rules | Event 8003 (EXE and DLL) or 8006 (scripts and MSI), level Warning: allowed to run but would have been prevented if enforced | Event 8004 (EXE and DLL) or 8007 (scripts and MSI), level Error: prevented from running | A host whose log shows only 3076 or 8003/8006 events is audit-only and counts as unmitigated in the record below, however many programs it lists as blocked. ## What it costs Every program change touches the list. A server that changes software often needs the list updated each time, or it blocks work (list too narrow) or admits too much (list too broad). The managed installer option cuts this cost where a deployment tool exists: an AppLocker rule collection names the trusted installer binaries, such as Microsoft Configuration Manager or Intune, and with the policy option Enabled:Managed Installer set, files those binaries write are allowed unless a deny rule covers them ([Microsoft Learn, managed installer](https://learn.microsoft.com/en-us/windows/security/application-security/application-control/app-control-for-business/design/configure-authorized-apps-deployed-with-a-managed-installer)). Windows trusts that origin only on the Windows OS volume, so apps installed on another volume need explicit rules. NIST sets no review interval for the list: CM-7(5) part c leaves the schedule to the organization. CM-7(5) part c exists for the list that is built once and never reviewed. That upkeep is the price to weigh against the fix. See the [decision table row for a patch that breaks a critical system](https://remediationvsmitigation.com/decision-table/patch-breaks-critical-system/). ## Check that it took 1. Endpoint security team: read the policy mode on each in-scope host, as in the table above. A policy that logs what it would block protects nothing, so only hosts that log 3077 or 8004/8007 blocks and no audit-only option count. 2. Security tester: on a pilot host your team owns, run a harmless unapproved test program and confirm the block event (3077 or 8004/8007). Stop at the block event. 3. Asset owner: compare the count of enforced hosts with the inventory of affected hosts. Hosts outside the policy are unmitigated. 4. Change manager: confirm who approves a new program for the list and how long that takes. 5. List owner: review the list on the schedule set under CM-7(5)(c). ## Record to keep Application allow-listing record - Finding: [CVE], [product], [date published]; fixed build [build] - Basis: inference from the CVE precondition, no vendor or CISA text names this control - Why allow-listing applies: exploit needs [a new program to run] on [host group] - Why it may not: [files from Microsoft's bypass list that the policy allows, if any] - Policy and product: [name, version]; mode on each host: enforced or audit-only - Hosts in scope: [number]; enforced: [number]; audit-only: [number] - Patch date per host group: [group], [date] - Due date: [date]; days between due date and patch date: [number] - Pilot test: [date], [host], unapproved test program blocked: [yes or no] - Approval route for new programs: [owner], [turnaround] - List owner: [name]; last review [date]; next review [date] - End date: [date of fix install]; fix owner [name] - Approver: [name], [date] ## Sources - Microsoft Learn, App Control for Business: https://learn.microsoft.com/en-us/windows/security/application-security/application-control/app-control-for-business/appcontrol (read 19 Nov 2026) - NIST SP 800-53 Rev. 5.2.0, CM-7(5) Authorized Software, Allow-by-exception: https://csrc.nist.gov/projects/cprt/catalog#/cprt/framework/version/SP_800_53_5_2_0/home/CM-7.5 (read 19 Nov 2026) - NIST OSCAL catalog of SP 800-53 Rev. 5.2.0 (last modified 2026-10-22), control text and discussion: https://github.com/usnistgov/oscal-content/blob/main/nist.gov/SP800-53/rev5/json/NIST_SP-800-53_rev5_catalog.json (read 19 Nov 2026) - CISA KEV feed, catalog 2026.11.18: CVE-2026-68820 (added 2026-08-11, due 2026-08-25): https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) - Microsoft MSRC CVRF, August 2026: CVE-2026-68820 fixed builds and KBs: https://api.msrc.microsoft.com/cvrf/v3.0/cvrf/2026-Aug (read 19 Nov 2026) - Microsoft Learn, App Control policy rules and file rules (rule option 3, Enabled:Audit Mode): https://learn.microsoft.com/en-us/windows/security/application-security/application-control/app-control-for-business/design/select-types-of-rules-to-create (read 19 Nov 2026) - Microsoft Learn, understanding App Control event IDs (3076, 3077): https://learn.microsoft.com/en-us/windows/security/application-security/application-control/app-control-for-business/operations/event-id-explanations (read 19 Nov 2026) - Microsoft Learn, using Event Viewer with AppLocker (8003, 8004, 8006, 8007): https://learn.microsoft.com/en-us/windows/security/application-security/application-control/app-control-for-business/applocker/using-event-viewer-with-applocker (read 19 Nov 2026) - Microsoft Learn, applications that can bypass App Control and how to block them: https://learn.microsoft.com/en-us/windows/security/application-security/application-control/app-control-for-business/design/applications-that-can-bypass-appcontrol (read 19 Nov 2026) - Microsoft Learn, automatically allow apps deployed by a managed installer: https://learn.microsoft.com/en-us/windows/security/application-security/application-control/app-control-for-business/design/configure-authorized-apps-deployed-with-a-managed-installer (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [EDR Rule as a Compensating Control: Coverage](https://remediationvsmitigation.com/compensating-controls/edr-rule/index.md) - [A Patch Would Break a Critical System: Mitigate, Then Fix](https://remediationvsmitigation.com/decision-table/patch-breaks-critical-system/index.md) - [WAF Rule as a Virtual Patch: Atlassian Case and Limits](https://remediationvsmitigation.com/compensating-controls/waf-rule/index.md) - [How to Verify a Mitigation Works: Checks for 5 Real CVEs](https://remediationvsmitigation.com/guides/verifying-a-mitigation-works/index.md) --- # Configuration change as a mitigation for a vulnerability Canonical: https://remediationvsmitigation.com/compensating-controls/configuration-change/ A configuration change removes a precondition the flaw needs, while the code stays. Cisco's advisory for CVE-2026-76418 (28 Oct 2026) lists no workaround but one mitigation: no tunnel-group-list enable, or no secondary-authentication-server-group when the default group uses secondary authentication. Read 19 Nov 2026. ## A setting counts when it removes a precondition Vendors list the conditions a flaw needs. A configuration change is a mitigation when it makes one of those conditions false and keeps it false. It is not one when it only narrows the condition. Vendors also use the words differently. Cisco's advisory for CVE-2026-76418 says "There are no workarounds that address this vulnerability. However, there is a mitigation". Microsoft's bulletin MS10-046 (2010) used "workaround" for a change you make and "mitigation" for a setting "existing in a default state", so the labels do not carry across vendors. The vocabulary is settled on [workaround vs mitigation](https://remediationvsmitigation.com/guides/workaround-vs-mitigation/). Settings and conditions from the CVE records and vendor advisories read 19 Nov 2026. | CVE | Setting | Named by | Does it remove the precondition? | |---|---|---|---| | CVE-2026-76418, Cisco ASA and FTD (KEV 29 Oct 2026) | `no tunnel-group-list enable`, or `no secondary-authentication-server-group` | Cisco | Yes. The flaw needs secondary authentication on a tunnel group in scope. | | CVE-2026-123158, OpenSSH on musl without PAM (KEV 9 Nov 2026) | `PasswordAuthentication no`, including inside Match blocks | OpenSSH and Alpine | Yes. The flaw needs password authentication on. The cost is that nobody can log in with a password. | | CVE-2026-109618, PHP out-of-memory corruption (KEV 26 Oct 2026) | `memory_limit=-1`, then restart every worker | PHP | Only in part. PHP says a failed mmap() can still reach the vulnerable path, and a runaway script is no longer stopped. | ## The named case: CVE-2026-76418 on Cisco ASA and FTD Cisco published the advisory at 19:15 GMT on 28 Oct 2026. A logic error in the Remote Access SSL VPN service lets an unauthenticated attacker reload the device, a denial of service rated 8.6 on CVSS 3.1. The device is affected when it runs a vulnerable release, the SSL VPN login service is reachable, and secondary authentication is configured in tunnel group scope. Cisco's advisory dates exploitation only as "October 2026" and gives no day, so the first dated public points are the advisory on 28 Oct and the CISA KEV entry on 29 Oct 2026, due 1 Nov 2026. Cisco's mitigation has two cases. Case 1: if DefaultWEBVPNGroup does not use secondary authentication, "enter no tunnel-group-list enable" in WebVPN configuration mode. Users then cannot pick a connection profile from a list and "will need to know the correct profile to use". Case 2: if DefaultWEBVPNGroup uses secondary authentication, use `no secondary-authentication-server-group` to remove it. Cisco states the cost: "the security posture is lowered for remote access VPN". It adds that the mitigation was proven in a test environment and that customers "should not deploy any workarounds or mitigations before first evaluating the applicability to their own environment". Cisco advisory cisco-sa-asaftd-tunnel-dos-s6mgkb75 (28 Oct 2026), read 19 Nov 2026. | Product | Train | First fixed release | |---|---|---| | ASA | 9.16 and earlier | 9.16.4.104 | | ASA | 9.18 | 9.18.4.96 | | ASA | 9.20 | 9.20.4.50 | | ASA | 9.22 | 9.22.3.200 | | ASA | 9.23 | 9.23.1.48 | | ASA | 9.24 | 9.24.1.27 (9.24.10.1 also listed) | | FTD | 7.0 and earlier | 7.0.11 | | FTD | 7.2 | 7.2.13 | | FTD | 7.4 | 7.4.9 | | FTD | 7.6 | 7.6.7 | | FTD | 7.7 | 7.7.14 | | FTD | 10.0 | 10.0.3 | | FTD | 10.1 | 10.1.1 | ## A worked read-back that catches the wrong fix The device below is illustrative, running ASA 9.20.4.40, below the first fixed release 9.20.4.50. The facts about Cisco's note are real: Cisco says that "disabling the group list does not remove exposure when that default group uses secondary authentication". 1. Thu 29 Oct 2026: KEV entry, due Sun 1 Nov 2026. Read `show running-config webvpn` and confirm SSL VPN is enabled on an interface. 2. Read `show running-config tunnel-group`. Search every group, including DefaultWEBVPNGroup, for `secondary-authentication-server-group`. Suppose 2 of 5 groups have it, one of them the default group. 3. An engineer sees `tunnel-group-list enable` and runs `no tunnel-group-list enable`. The login page no longer lists profiles. 4. Read `show running-config tunnel-group` again. The default group still has `secondary-authentication-server-group`, so the device is still exposed. The command changed the login page, not the precondition. 5. Apply Case 2: remove secondary authentication from the default group. Remove it from the other group, since the aim is none in scope. Read both groups back. 6. Record the lowered posture, because VPN users are no longer asked for the second credential, and set the end date to the upgrade to 9.20.4.50. ## The baseline rule that governs it SP 800-53 CM-6 asks organizations to "Establish and document configuration settings for components employed within the system that reflect the most restrictive mode consistent with operational requirements" (item a) and to "Identify, document, and approve any deviations from established configuration settings" (item c). Removing secondary authentication is a deviation from the standard: it lowers the posture. It needs an approver and a date, or it is a private change an assessor cannot accept. Item d asks that changes to settings are monitored and controlled, which is the drift rule below. ## Check that it took, and what it costs 1. Read the setting back from the running configuration and compare it with the approved value, as in the worked case. 2. Run the vendor's own check for the condition if one exists (here the configuration checks in Cisco's advisory). 3. Rebuild a test instance from the standard image and confirm the setting is present, since a rebuild from an old image brings the default back. 4. Alert on drift from the approved value so a revert opens a ticket. The cost is what the setting switches off. Cisco's cases cost either the profile list on the login page or the second credential. The OpenSSH setting costs all password logins; test key login in a second session first. Revisit the entry when the fix is installed and the setting can go back, or when exploitation is reported against devices that already carry the setting, in which case it was never enough. ## Record to keep # Configuration change record (deviation from baseline) - Finding: CVE-2026-76418, Cisco ASA and FTD (Cisco advisory 28 Oct 2026) - Asset and release: [hostname], [release]; first fixed release [release] - Precondition removed: secondary authentication in tunnel group scope - Setting changed: [command], on [groups]; before [value]; after [value] - Read-back on [date]: [output summary]; default group checked: [yes or no] - Posture cost accepted: [what was lowered]; approver [name] - Baseline: added to baseline [yes or no]; deviation ID [ID] - Drift alert rule: [rule ID] - Rebuild test: [date], setting present: [yes or no] - KEV due date: [date]; met: [yes or no] - End date: [date of upgrade]; revert step: [command]; owner [name] ## Sources - Cisco advisory cisco-sa-asaftd-tunnel-dos-s6mgkb75 (28 Oct 2026): vulnerable configuration, mitigation, fixed releases: https://sec.cloudapps.cisco.com/security/center/content/CiscoSecurityAdvisory/cisco-sa-asaftd-tunnel-dos-s6mgkb75 (read 19 Nov 2026) - CISA KEV feed, catalog 2026.11.18: CVE-2026-76418 (added 2026-10-29, due 2026-11-01): https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) - OpenSSH 10.6p2 release notes: https://www.openssh.org/releasenotes.html#10.6p2 (read 19 Nov 2026) - Alpine Linux, OpenSSH password authentication bypass (5 Nov 2026): https://alpinelinux.org/posts/2026-11-05-openssh-password-authentication-bypass/ (read 19 Nov 2026) - PHP advisory GHSA-w8hw-w9fx-j79w, CVE-2026-109618: https://github.com/php/php-src/security/advisories/GHSA-w8hw-w9fx-j79w (read 19 Nov 2026) - Microsoft Security Bulletin MS10-046 (2010), archived: definitions of workaround and mitigation: https://learn.microsoft.com/en-us/security-updates/securitybulletins/2010/ms10-046 (read 19 Nov 2026) - NIST SP 800-53 Rev. 5.2.0, CM-6: https://csrc.nist.gov/projects/cprt/catalog#/cprt/framework/version/SP_800_53_5_2_0/home/CM-6 (read 19 Nov 2026) - NIST OSCAL catalog of SP 800-53 Rev. 5.2.0 (last modified 2026-10-22), control text: https://github.com/usnistgov/oscal-content/blob/main/nist.gov/SP800-53/rev5/json/NIST_SP-800-53_rev5_catalog.json (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Workaround vs Mitigation in Real Vendor Advisories](https://remediationvsmitigation.com/guides/workaround-vs-mitigation/index.md) - [Disable a Vulnerable Feature: When It Counts as Mitigation](https://remediationvsmitigation.com/compensating-controls/disable-feature-or-service/index.md) - [Exploited Flaw, No Patch Yet: Mitigate and Set Two Dates](https://remediationvsmitigation.com/decision-table/kev-no-patch-yet/index.md) - [Mitigation vs Remediation Wording in Vendor Advisories](https://remediationvsmitigation.com/guides/mitigation-vs-remediation-in-vendor-advisories/index.md) --- # Disabling a vulnerable feature, service or port as a mitigation Canonical: https://remediationvsmitigation.com/compensating-controls/disable-feature-or-service/ Disabling a feature, service or port counts as a mitigation when the exploit needs it and the business owner accepts losing the function. FortiMail's IBE service in CVE-2026-104286, exploited from 1 Oct 2026, is the worked case. The software stays installed, so the entry needs an end date and a log review. ## Two classes in NIST's own words NIST SP 800-40 Rev. 4 files disabling in two places. Under Mitigate it lists "disabling a vulnerable feature" in the sentence "Reduce the risk by eliminating the vulnerabilities (e.g., patching the vulnerable software, disabling a vulnerable feature, or upgrading to a newer software version without the vulnerabilities)". Under Avoid it lists "disabling computing capabilities in assets that can function without them". The business need separates the two. If nobody uses the capability, the risk does not occur and the publication calls that avoidance. If the capability is wanted and only switched off for now, it is a mitigation, and it needs an end date. Write down which one you mean: only the second needs a removal schedule. PCI SSC FAQ 1597 (May 2025) gives the same action as a way to address a vulnerability rather than resolve it: "e.g., with a compensating control or by disabling a vulnerable service". NIST SP 800-53 CM-7 b is the standing rule behind it: "Prohibit or restrict the use of the following functions, ports, protocols, software, and/or services". ## Which disables count and which miss the path A disable only counts if the thing you turn off is on the path the exploit takes. These cases come from the CVE records read on 19 Nov 2026. Cases from the CVE records and vendor advisories read 19 Nov 2026. Vendor wording is quoted or closely paraphrased from each advisory. Cisco's ASA and FTD settings are on the [configuration change page](https://remediationvsmitigation.com/compensating-controls/configuration-change/). | Case | What the vendor or source says | Counts as a mitigation? | |---|---|---| | FortiMail IBE service off (CVE-2026-104286) | Fortinet's workaround; IBE on is a condition of exposure | Yes. IBE secure mail stops working | | SonicWall SonicOS SSLVPN with SAML: disable or delete every active LDAP, RADIUS and TACACS+ server (CVE-2026-19868) | SonicWall workaround, updated 9 Nov 2026 | Yes. Anything signing in through those servers stops until they return | | Adobe Commerce: disable mail (CVE-2026-75650) | Sansec: "Disabling mail does not stop it either" | No. Mail is not on the exploit path | | Langflow: disable auto-login (CVE-2026-9198) | Langflow documentation: disabling does not revoke sessions or API keys issued earlier | Not for past exposure. Rotate or revoke the credentials too | A KEV required action can itself prescribe a disable, with a scope. For the Service Location Protocol flaw CVE-2023-29552, added to KEV on 8 Nov 2023 with a due date of 29 Nov 2023, CISA's required action reads "Apply mitigations per vendor instructions or disable SLP service or port 427/UDP on all systems running on untrusted networks, including those directly connected to the Internet." The disable applies to systems on untrusted networks, not to every host. ## Worked case: FortiMail IBE, 1 Oct to 7 Oct 2026 Source: Fortinet PSIRT FG-IR-26-175 and its CVRF revision history (1, 5 and 7 Oct 2026), and the CISA KEV entry, read 19 Nov 2026. | Date | Event | |---|---| | 1 Oct 2026 | Fortinet publishes advisory FG-IR-26-175 and says the flaw is exploited. CISA adds CVE-2026-104286 to the KEV catalog the same day with a due date of 4 Oct 2026, a 3-day window (the KEV feed, catalog 2026.11.18, shows dateAdded 2026-10-01 and dueDate 2026-10-04). | | 5 Oct 2026 | Fortinet updates FortiMail Cloud to 7.6.7 GA or 8.0.2 GA. Cloud customers take no action. | | 7 Oct 2026 | Advisory revised to clarify the FortiMail Cloud fix. FortiMail 7.2.10 is listed as upcoming; FortiMail 7.0 is told to migrate to a fixed release. This revision is still the current one on 19 Nov 2026. | The flaw is a path traversal (CWE-22, CVSS 3.1 base score 9.8) that lets an unauthenticated attacker write files on the appliance through HTTP or HTTPS requests. Fixed builds are 8.0.2, 7.6.7 and 7.4.9 or later. Exposure needs three things: an affected build, the IBE service on, and a web interface the attacker can reach. Fortinet's first workaround removes the second condition. In the GUI, set Encryption > IBE > IBE Service to off. In the CLI, run `config system encryption ibe`, then `set status disable`, then `end`. The cost Fortinet's record implies is that recipients can no longer open or reply to IBE-encrypted messages through the FortiMail secure portal, so the mail owner has to sign off. The control matters most for the 7.2 and 7.0 lines. On the advisory's last revision of 7 Oct 2026, no fixed 7.2 build had shipped (7.2.10 is listed as upcoming and is unreleased as of 19 Nov 2026), and 7.0 has none at all. Turning IBE off, or blocking internet access to the webmail interface, is the vendor's only stated step for those appliances. A register entry for a 7.2 appliance therefore ends on a review date, and also when a fixed build is installed. ## What disabling does not tell you Fortinet reported exploitation on the day of publication, and CISA marks forensic triage as Yes for this entry. Turning the service off stops new requests on that path. It does not remove a file already written. Preserve the appliance logs and a configuration backup before you change anything, then search for the lines below. Fortinet's advisory FG-IR-26-175 (read 19 Nov 2026) lists them as indicators of compromise, with addresses defanged here. **Indicator addresses.** Check firewall and proxy logs for connections to `79[.]141[.]169[.]187` and `45[.]129[.]0[.]192`. **System event log (Log > Event > System), three patterns:** - `type=event subtype=system pri=debug user=system ui=cron msg="(root) CMD (/bin/sh -c 'O=/migadmin ...` (the advisory ends the line there) - `type=kevent subtype=admin pri=information user=admin ui=(null) action=logout status=success reason=unknown msg="User admin logged out from (null)."` - `type=kevent subtype=config pri=information user=admin ui=cli module=unknown submodule=unknown msg="Added 'archive234' to 'archive account' : rotation-size[50]rotation-time[1] rotation-hour[14]destination[remote]remote-ip[79[.]141[.]169[.]187]remote-username[archive234]remote-password[***]remote-directory[/uploads] (user: admin, from: cli)"` **Encryption log (Log > Encryption), two lines:** - `FortiMail::IBE::DecrypterMediaIn::DecrypterMediaIn(FortiMail::MediaIn&, const FortiMail::IBE::KeyFinder&, const FortiMail::EmailAddress&, const FortiMail::Buffer&, FortiMail::IBE::DecrypterMediaIn::Version): Caught BufferException(2), BufferImpl.cpp:973, 'Invalid Base64 Encoding at pos 0. Character=0x2a'` - `Internal user *@domain.tld failed to log in.` Fortinet lists these as patterns, so a match is a reason to start triage. The archive-account line is the one that names a remote destination. Matching these on a schedule is the detective half of this entry; see [increased monitoring](https://remediationvsmitigation.com/compensating-controls/increased-monitoring/). ## Check that it took and keeps holding 1. Read the setting back on each appliance. `show system encryption ibe` should now show `set status disable`, where the pre-change export showed `set status enable`. SP 800-40 Rev. 4 asks that additional controls be "functioning as intended". 2. Restart or fail over the appliance and read the setting again. 3. Re-enable IBE only after `get system status` shows a fixed build. SP 800-40 Rev. 4 asks that the response "continues to be in place": no one deactivates it, and an upgrade or restore does not reset it. 4. Add a configuration-drift alert on the IBE status so a re-enable raises a ticket. 5. Get the mail owner's written acceptance that IBE secure mail is off. ## What to record Control: Disable IBE service (FortiMail) Finding: CVE-2026-104286 on , FortiMail Class: Mitigation (capability wanted, turned off for now) or Avoidance (capability unused) Setting before: set status enable (config export ) Setting after: set status disable (config export ) Logs and config backup preserved before the change: Cost accepted by: , loss: IBE secure mail Ends when: fixed build installed (8.0.2, 7.6.7, 7.4.9 or later) or review date Read back by: on ; after restart on Drift alert: Other pages that matter here: [configuration changes](https://remediationvsmitigation.com/compensating-controls/configuration-change/) leave the flaw reachable under non-default use, [removing internet exposure](https://remediationvsmitigation.com/compensating-controls/remove-internet-exposure/) is the other FortiMail workaround, and [how to verify a mitigation works](https://remediationvsmitigation.com/guides/verifying-a-mitigation-works/) covers the read-back in more depth. ## Sources - NIST SP 800-40 Rev. 4, Guide to Enterprise Patch Management Planning, Sections 2.1 and 2.2: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf (read 19 Nov 2026) - PCI SSC FAQ 1597 (May 2025), resolved and addressed: https://www.pcisecuritystandards.org/faqs/1597 (read 19 Nov 2026) - NIST SP 800-53 Rev. 5.2.0, CM-7: https://csrc.nist.gov/projects/cprt/catalog#/cprt/framework/version/SP_800_53_5_2_0/home/CM-7 (read 19 Nov 2026) - Fortinet PSIRT advisory FG-IR-26-175 (vendor): workaround, fixed builds, indicators of compromise: https://fortiguard.fortinet.com/psirt/FG-IR-26-175 (read 19 Nov 2026) - FortiMail CLI reference, system encryption ibe (vendor): https://docs.fortinet.com/document/fortimail/7.6.0/cli-reference/6402/system-encryption-ibe (read 19 Nov 2026) - CISA Known Exploited Vulnerabilities catalog, feed version 2026.11.18: https://www.cisa.gov/known-exploited-vulnerabilities-catalog (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Configuration Change as a Mitigation: Cisco ASA Case](https://remediationvsmitigation.com/compensating-controls/configuration-change/index.md) - [Remove Internet Exposure as a Mitigation: What It Buys](https://remediationvsmitigation.com/compensating-controls/remove-internet-exposure/index.md) - [How to Verify a Mitigation Works: Checks for 5 Real CVEs](https://remediationvsmitigation.com/guides/verifying-a-mitigation-works/index.md) - [Increased Monitoring as a Compensating Control](https://remediationvsmitigation.com/compensating-controls/increased-monitoring/index.md) --- # EDR rule as a compensating control: what it needs Canonical: https://remediationvsmitigation.com/compensating-controls/edr-rule/ An EDR rule counts as a mitigation only on hosts where a healthy agent runs it with a blocking action; in alert-only mode it is monitoring. For PHP2Shell (CVE-2026-109618, KEV 26 Oct 2026), CrowdStrike documents prevention of post-exploitation behavior, while CISA's SIGMA rule is a log detection. ## Coverage decides the class For most register entries the control is a rule. Here it is the rule plus agent coverage. An affected host with no agent, a stale agent or a sensor in alert-only mode sits outside the control, and an assessor will ask for that list. Pull the agent report for the affected hosts only, not the fleet. | State of an affected host | What the rule can do | How to file it | |---|---|---| | Healthy agent, rule action set to block or prevent | Stops the targeted behavior on that host | Mitigation for that host | | Healthy agent, rule action set to alert | Raises an alert, lets the behavior run | Detective control: file it under [increased monitoring](https://remediationvsmitigation.com/compensating-controls/increased-monitoring/) | | No agent, stale agent or sensor disabled | Nothing | Uncovered: an open item with an owner and a date | | Host cannot run an agent (appliance, embedded device) | Nothing | Needs another control; see [OT and medical devices that cannot be patched](https://remediationvsmitigation.com/decision-table/ot-medical-cannot-patch/) | Example with invented counts: 40 hosts run an affected PHP build, 31 report a healthy agent, and 24 of those run the rule with a blocking action. The mitigation covers 24 of 40 hosts (60 percent). Seven hosts are detective only and nine are open. The entry states those three numbers and lists the nine. Filled in for those hosts: - Control: EDR rule. Finding: CVE-2026-109618 on 40 hosts running PHP 8.2 to 8.5 below the fixed versions 8.2.35, 8.3.36, 8.4.27 and 8.5.12 - Mode: block on 24 hosts, alert on 7; uncovered: 9 hosts (no agent, stale agent or sensor off), list attached - Pilot test: one Linux pilot host; the record shows a prevented, not only detected, entry - Alert owner and runbook: named; exceptions with approvers: the 9 uncovered hosts - Ends when: the fixed PHP version is installed on all 40 hosts; review date 29 Oct 2026, the KEV due date ## Reading the action on a Linux host Read the product's own definition of block before filing an entry. Vendors split the same behavior rule into two states. Elastic's documentation for Elastic Defend malicious behavior protection defines them: Detect "Detects malicious behavior on the host and generates an alert. Elastic Defend will not block the malicious behavior." Prevent "Detects malicious behavior on the host, forces the process to stop, and generates an alert" ([Elastic Docs, configure an integration policy for Elastic Defend](https://www.elastic.co/docs/solutions/security/configure-elastic-defend/configure-an-integration-policy-for-elastic-defend), read 19 Nov 2026). Only the second state is a mitigation. Microsoft's documentation for EDR in block mode in Defender for Endpoint lists Windows as the supported operating system, so that setting does not apply to the Linux servers in the case below. PHP2Shell runs on Linux web servers. CrowdStrike's post of 21 Oct 2026 says its indicators of attack "detect and prevent" the post-exploitation behaviors it lists, and names no policy setting. In Elastic Security the state is in the alert's rule name: Elastic's endpoint protection rules include "Behavior - Detected - Elastic Defend" and "Behavior - Prevented - Elastic Defend" ([Elastic Docs, endpoint protection rules](https://www.elastic.co/docs/solutions/security/manage-elastic-defend/endpoint-protection-rules)). In any product, open the detection raised by a test of the behavior on one pilot Linux host and check that the record says the process was blocked or prevented, not only detected. If it says detected, the host is in the alert-only row of the coverage table. A "block" that acts after a detection stops or cleans up a process or artifact; it does not necessarily stop the exploit request that started it. Record what the product's action does and which operating systems it covers. ## Worked case: PHP2Shell, CVE-2026-109618 CVE-2026-109618 is a memory-corruption flaw in PHP (CVSS 3.1 base score 9.8) that lets an unauthenticated request lead to remote code execution. It affects PHP 8.2 before 8.2.35, 8.3 before 8.3.36, 8.4 before 8.4.27 and 8.5 before 8.5.12; PHP 8.1 and older get no upstream fix. Sources: Fastly's PHP2Shell report, the CISA alert (last revised 26 Oct 2026) and the CISA KEV entry, read 19 Nov 2026. | Date and time | Event | |---|---| | 21 Oct 2026, 03:59 UTC | First request from a PHP2Shell-named tool in Fastly's telemetry | | 21 Oct 2026, 14:00 UTC | PHP releases 8.5.12, 8.4.27, 8.3.36 and 8.2.35, 10 hours 1 minute after that first request (release time scheduled for 14:00 UTC, as listed in Fastly's timeline, which cites the PHP Foundation) | | 23 Oct 2026 | CISA updates its alert to add a SIGMA detection rule | | 26 Oct 2026 | CISA adds the CVE to KEV, due 29 Oct 2026, forensic triage marked Yes | Two different things were published for the host, and they sit in different classes. **Prevention, vendor-stated.** On 21 Oct 2026 CrowdStrike wrote that its existing indicators of attack "detect and prevent common post-exploitation activity, including web shell deployment, suspicious child processes spawned by PHP or web server processes, and malicious command execution" ([CrowdStrike, 21 Oct 2026](https://www.crowdstrike.com/en-us/blog/crowdstrike-protects-customers-from-php-rce-zero-days/)). That is a vendor's claim about its own product. The rules are generic, not written for this CVE. They act after a PHP worker has been made to start a shell, so they limit what follows the exploit request and do not stop the request. The post does not say which hosts have the prevent action switched on, so the coverage count above decides whether it is a mitigation for a given host. **Detection, CISA.** CISA's alert, updated 23 Oct 2026, says to "Monitor for suspicious activity, including PHP worker processes launching shells or downloading tools" and points to its SIGMA rule: "Detects PHP2Shell CVE-2026-109618 CVE-2026-109619 RCE Exploitation Attempts", rule ID 741682b9-041e-4a31-b314-c58b15a82cf2, status test, file CISA_SIGMA_261429_r2v1_CVE_2026_109618-109619_PHP2Shell_TLP_CLEAR.yaml at the root of the [cisagov/SIGMA_Rules](https://github.com/cisagov/SIGMA_Rules) repository (the copy read on 19 Nov 2026 is marked modified 5 Nov 2026). It matches HTTP request patterns, public exploit markers, PHP error messages and reported indicators in log sources. It alerts and blocks nothing, so it is a monitoring entry. The log patterns that rule matches, and the host evidence Fastly lists, are printed on [increased monitoring](https://remediationvsmitigation.com/compensating-controls/increased-monitoring/). ## What it costs and where it fails The cost is false positives and agent overhead. NIST SP 800-53 SI-3 says the control must "Address the receipt of false positives during malicious code detection and eradication and the resulting potential impact on the availability of the system." A block rule that stops a legitimate PHP worker spawning a process breaks the application, so a rule tuned to avoid that may be tuned out of usefulness. SI-4 d adds that detections must be analyzed: "Analyze detected events and anomalies;". The rule also fails where the exploit completes before the agent sees a behavior, and where an alert reaches no one. ## Check the rule and the agent 1. Count affected hosts, hosts with a healthy agent, and hosts with the blocking action, as in the example above. Give each gap an owner and a date. 2. On one pilot host, trigger a harmless test of the behavior on a host your team owns and keep the record: the console entry that shows a block or prevent action and not only an alert. 3. Confirm someone receives the alert and has a runbook with a time to first look. 4. Re-run the coverage count after every agent upgrade and every new server build. 5. Set a review date. SP 800-40 Rev. 4 asks that the response "continues to be in place": no one deactivates it. ## What to record - Control: EDR rule - Finding: on , - Rule ID and source: , as documented by on - Mode: block or alert (tested on ) - Affected hosts: ; healthy agent: ; blocking action: ; uncovered list attached - Pilot test record: - Alert owner and runbook: , - Exceptions with approvers: - Ends when: fixed version installed, or review date The paired detective control is [increased monitoring](https://remediationvsmitigation.com/compensating-controls/increased-monitoring/), and the other host-level preventive options are in the register. Network-side blocking is covered under [IPS signatures](https://remediationvsmitigation.com/compensating-controls/ips-signature/). ## Sources - NIST SP 800-53 Rev. 5.2.0, SI-3: https://csrc.nist.gov/projects/cprt/catalog#/cprt/framework/version/SP_800_53_5_2_0/home/SI-3 (read 19 Nov 2026) - NIST SP 800-53 Rev. 5.2.0, SI-4: https://csrc.nist.gov/projects/cprt/catalog#/cprt/framework/version/SP_800_53_5_2_0/home/SI-4 (read 19 Nov 2026) - NIST SP 800-40 Rev. 4, Guide to Enterprise Patch Management Planning, Sections 2.1 and 2.2: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf (read 19 Nov 2026) - Microsoft Learn, EDR in block mode (supported operating system: Windows): https://learn.microsoft.com/en-us/defender-endpoint/edr-in-block-mode (read 19 Nov 2026) - CISA alert, Critical Vulnerabilities in PHP Require Immediate Patching: https://www.cisa.gov/news-events/alerts/2026/10/21/critical-vulnerabilities-php-require-immediate-patching (read 19 Nov 2026) - Fastly, how Fastly detected and responded to PHP2Shell: https://www.fastly.com/blog/how-fastly-detected-and-built-a-coordinated-response-to-php2shell (read 19 Nov 2026) - PHP security advisory GHSA-w8hw-w9fx-j79w (vendor): https://github.com/php/php-src/security/advisories/GHSA-w8hw-w9fx-j79w (read 19 Nov 2026) - CISA Known Exploited Vulnerabilities catalog, feed version 2026.11.18: https://www.cisa.gov/known-exploited-vulnerabilities-catalog (read 19 Nov 2026) - CrowdStrike, CrowdStrike Protects Customers from Threats Delivered via PHP2Shell (21 Oct 2026, vendor): https://www.crowdstrike.com/en-us/blog/crowdstrike-protects-customers-from-php-rce-zero-days/ (read 19 Nov 2026) - CISA SIGMA_Rules repository, PHP2Shell rule file (root of repository): https://github.com/cisagov/SIGMA_Rules (read 19 Nov 2026) - Elastic Docs, configure an integration policy for Elastic Defend (malicious behavior protection levels): https://www.elastic.co/docs/solutions/security/configure-elastic-defend/configure-an-integration-policy-for-elastic-defend (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Increased Monitoring as a Compensating Control](https://remediationvsmitigation.com/compensating-controls/increased-monitoring/index.md) - [IPS Signature as a Mitigation: When It Counts](https://remediationvsmitigation.com/compensating-controls/ips-signature/index.md) - [OT and Medical Devices You Cannot Patch: Mitigate, Review](https://remediationvsmitigation.com/decision-table/ot-medical-cannot-patch/index.md) --- # Increased monitoring as a compensating control Canonical: https://remediationvsmitigation.com/compensating-controls/increased-monitoring/ Increased monitoring is a detective control: it shortens the time to notice an attack and prevents nothing. For PHP2Shell (CVE-2026-109618, KEV 26 Oct 2026) CISA's SIGMA rule gives a monitor eight match branches to run. For F5 CVE-2026-94127 (KEV 22 Sep 2026) neither F5 nor CISA published anything to match. ## What it changes and what it does not Eight of the nine register entries reduce the chance of exploitation. This one reduces the time between exploitation and response. It prevents nothing, and for a flaw already exploited in the wild it is a weak control on its own. NIST SP 800-53 SI-4 e gives the trigger for raising it: "Adjust the level of system monitoring activity when there is a change in risk to organizational operations and assets, individuals, other organizations, or the Nation;". A newly disclosed, unpatched flaw is a change in risk, and the entry should say monitoring was raised for that reason on that date. SI-4 d turns collection into a control: "Analyze detected events and anomalies;". Collection with no analysis and no owner is a log. Do not confuse it with the monitoring step in SP 800-40 Rev. 4 Section 2.2, "Continuously monitor the risk response", which watches that a mitigation stays in place. This page is about watching for exploitation. ## What a monitor has to match A rule needs something to match. The four KEV records below give very different amounts, and the amount decides how much the control is worth. CISA KEV feed 2026.11.18, CISA alert of 21 Oct 2026, and the vendor advisories for each CVE, read 19 Nov 2026. | CVE | Published in the advisory or CISA alert | What a monitor can match | |---|---|---| | CVE-2026-94127 (F5 BIG-IP APM) | Advisory 22 Sep 2026, KEV the same day, due 25 Sep 2026 | No log strings, file paths or indicator addresses from F5 or CISA. Only the configuration and version can be read | | CVE-2026-0322 (PAN-OS) | Advisory and KEV 14 Oct 2026, due 17 Oct 2026 | Three indicator addresses: 66[.]42[.]72[.]14, 208[.]83[.]239[.]114, 149[.]248[.]47[.]234. No log signatures or file indicators | | CVE-2026-104286 (FortiMail) | Advisory and KEV 1 Oct 2026, due 4 Oct 2026 | Two addresses, three System event log patterns and two Encryption log lines, printed on the [disable-a-feature page](https://remediationvsmitigation.com/compensating-controls/disable-feature-or-service/) | | CVE-2026-109618 (PHP) | Alert 21 Oct 2026, SIGMA rule added 23 Oct, KEV 26 Oct 2026 | CISA's SIGMA rule (8 branches, 95 indicator strings), PHP worker processes launching shells, AH00051 error entries, outbound connections from workers | Monitoring fits the last two rows. The PHP2Shell rule is worked next. F5 shows the empty end: CVE-2026-94127 (heap buffer overflow in BIG-IP APM, CVSS 4.0 base score 9.3) came with an iRule on request through an F5 support ticket and no log string or address ([F5 CVE record](https://cveawg.mitre.org/api/cve/CVE-2026-94127), read 19 Nov 2026). CISA's wording is to apply the iRule "as a temporary step to allow proactive forensic triage", then patch. BOD 26-04 defines the text "& forensic triage" as meaning "the agency must complete remediation or mitigation action within the timeline (three days) and carry out a forensic triage of the asset to assess whether the system is compromised." KEV added the CVE on 22 Sep 2026, due 25 Sep. There monitoring cannot be the entry: it only tells you whether the window already cost something. ## Worked case: CISA's SIGMA rule for PHP2Shell, CVE-2026-109618 CISA's alert of 21 Oct 2026 gained a SIGMA rule on 23 Oct. The file read on 19 Nov 2026 is CISA_SIGMA_261429_r2v1_CVE_2026_109618-109619_PHP2Shell_TLP_CLEAR.yaml in the [cisagov/SIGMA_Rules](https://github.com/cisagov/SIGMA_Rules) repository: rule ID 741682b9-041e-4a31-b314-c58b15a82cf2, status test, level critical, dated 21 Oct 2026 and marked modified 5 Nov 2026. It is one condition that ORs eight branches; inside a branch every keyword group must be present. The table lists what each branch matches and the log it needs. CISA SIGMA_Rules repository, PHP2Shell rule r2v1, read 19 Nov 2026. The file's own false-positive note reads: "Rate of FP medium-high with some strings/branches." | Branch | Log it needs | Matches (all groups in the branch) | |---|---|---| | 1 Request | Web or WAF log that carries request headers and body text, which a default access log does not | POST, index.php, PHPSESSID and PHP_SESSION_UPLOAD_PROGRESS, plus one of 15 request markers such as php2shell, B7kQpoc61, multipart/form-data or filename | | 2 Process | EDR or audit process events on the web hosts | A web or PHP process name (php-cgi, apache2, httpd, php-fpm, mod-php) with one of 12 shell or tool words (/bin/sh, /bin/bash, cmd, powershell, python, curl, wget, netcat, nc, whoami, id, uname) | | 3 PHP error log | PHP or Apache error log | "PHP Fatal error:", "Allowed memory" and " in Unknown " | | 4 Web shell line | Web root file contents or response capture | The fragments <, php, echo, system, a $_GET lookup and > in one record | | 5 Crash trace | Core dump or crash log | pDestructor, zif_system and libphp.so | | 6 Upload progress | Web or WAF log with body text | multipart, PHP_SESSION_UPLOAD_PROGRESS, "very large", request or repeated, fields or form-data, and memory | | 7 Cookie | Web or WAF log with response headers | Set-Cookie, PHPSESSID and "unusually large cookie value" | | 8 Indicator list | Any log that holds addresses, names or hashes | Any one of 95 strings: 56 IP addresses, 17 file hashes, and 22 domain names, tool names and request markers | Fastly, which measured the traffic, found no payload string to match, because the exploit is a failure condition, and said source IPs and User-Agent strings go stale within a day. Branch 8 therefore ages fastest; branches 2 and 3 describe what a compromised or failing worker does. ## Checks for this rule 1. Convert the rule with Sigma for your SIEM. Its logsource is the placeholder product cisa, and the file says to edit it only if you know your logsources, so map it and record which of the eight branches have a real log feeding them. A branch with no source cannot fire. 2. Run it over every retained log back to 21 Oct 2026, 03:59 UTC, the first PHP2Shell-named request in Fastly's telemetry. On 19 Nov 2026 that is 29 days. A 30-day retention covers it; a 14-day retention reaches only 5 Nov. 3. Test branch 3 on a test index: add one synthetic line holding the three phrases ("PHP Fatal error:", "Allowed memory", " in Unknown ") and confirm the alert fires, then delete the line. Do not send anything to a production host. 4. Triage hits by branch. A branch 8 hit on an address or marker shows contact only; Fastly found about a third of the PHP2Shell-named traffic it saw labeled itself as authorized exposure or patch-state checking. A branch 2 or branch 3 hit, or an outbound connection from a PHP worker, points to a host that needs forensic triage. Branch 2 keywords such as id, nc and cmd also match inside longer words, so expect noise. 5. Name the alert owner and the time to first look, and set the stop date: the entry ends when PHP 8.5.12, 8.4.27, 8.3.36 or 8.2.35 (or your distribution's patched package) is installed and the triage is closed. ## What to record Control: Increased monitoring (detective) Finding: on , preventive control paired: Raised on: ; reason: change in risk (SI-4 e) Detection rule ID and logic: , Branches with a live log source: ; branches without: Lookback run: from over of retained logs; hits: Match source published by: on , or none found on Test alert: , Owner and runbook: , ; time to first look: Alert review log: Stops or reviewed on: Forensic triage: Alert volume needs an owner who reviews it inside the stated time, because a rule that fires hundreds of times a day without review is not a control. Used alone against a flaw under active exploitation it is not enough: pair it with a preventive entry such as [removing internet exposure](https://remediationvsmitigation.com/compensating-controls/remove-internet-exposure/). It is reasonable as the only control for a low-likelihood finding with a dated exception, which is the case the [exception register](https://remediationvsmitigation.com/exception-register/) is for. The effect of monitoring on the residual figure is in [residual risk after mitigation](https://remediationvsmitigation.com/guides/residual-risk-after-mitigation/), and the [decision table](https://remediationvsmitigation.com/decision-table/) shows when a mitigation may stand in for a fix. ## Sources - NIST SP 800-53 Rev. 5.2.0, SI-4: https://csrc.nist.gov/projects/cprt/catalog#/cprt/framework/version/SP_800_53_5_2_0/home/SI-4 (read 19 Nov 2026) - NIST SP 800-40 Rev. 4, Guide to Enterprise Patch Management Planning, Sections 2.1 and 2.2: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf (read 19 Nov 2026) - CISA BOD 26-04, Appendix A, notes to Table 1: https://www.cisa.gov/news-events/directives/bod-26-04-prioritizing-security-updates-based-risk (read 19 Nov 2026) - F5 CVE record for CVE-2026-94127 on CVE Services (vendor data: workaround, hotfixes): https://cveawg.mitre.org/api/cve/CVE-2026-94127 (read 19 Nov 2026) - Fortinet PSIRT advisory FG-IR-26-175 (vendor): https://fortiguard.fortinet.com/psirt/FG-IR-26-175 (read 19 Nov 2026) - CISA alert, Critical Vulnerabilities in PHP Require Immediate Patching: https://www.cisa.gov/news-events/alerts/2026/10/21/critical-vulnerabilities-php-require-immediate-patching (read 19 Nov 2026) - CISA Known Exploited Vulnerabilities catalog, feed version 2026.11.18: https://www.cisa.gov/known-exploited-vulnerabilities-catalog (read 19 Nov 2026) - CISA SIGMA_Rules, PHP2Shell rule r2v1 (rule text, modified 2026-11-05): https://raw.githubusercontent.com/cisagov/SIGMA_Rules/develop/CISA_SIGMA_261429_r2v1_CVE_2026_109618-109619_PHP2Shell_TLP_CLEAR.yaml (read 19 Nov 2026) - Fastly, how Fastly detected and responded to PHP2Shell: first request time, indicator staleness, look-alike traffic: https://www.fastly.com/blog/how-fastly-detected-and-built-a-coordinated-response-to-php2shell (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [EDR Rule as a Compensating Control: Coverage](https://remediationvsmitigation.com/compensating-controls/edr-rule/index.md) - [Remove Internet Exposure as a Mitigation: What It Buys](https://remediationvsmitigation.com/compensating-controls/remove-internet-exposure/index.md) - [Exception Register: 15 Columns, States and Blank CSV](https://remediationvsmitigation.com/exception-register/index.md) - [Residual Risk After Mitigation: Define, Estimate, Record](https://remediationvsmitigation.com/guides/residual-risk-after-mitigation/index.md) - [How to Verify a Mitigation Works: Checks for 5 Real CVEs](https://remediationvsmitigation.com/guides/verifying-a-mitigation-works/index.md) --- # IPS virtual patching: signature and filter checks Canonical: https://remediationvsmitigation.com/compensating-controls/ips-signature/ An IPS signature counts as a mitigation when it blocks, sits inline and is current. For CVE-2026-0322, Palo Alto named Threat IDs 510046 and 510047 on 14 Oct 2026, calls the coverage limited, and requires inbound decryption to make them work. Check the action, the path and the content version. ## What a signature covers An intrusion prevention system inspects traffic and drops flows that match a signature written for one vulnerability. OWASP's Virtual Patching Cheat Sheet names the device class: "Intermediary devices such as a WAF or IPS appliance". NIST SP 800-53 SI-4 places it in monitoring tooling: "System monitoring capabilities are achieved through a variety of tools and techniques, including intrusion detection and prevention systems, malicious code protection software, scanning tools, audit record monitoring software, and network monitoring software." A [WAF rule](https://remediationvsmitigation.com/compensating-controls/waf-rule/) reads application-layer requests for one web application. A signature sits on the network path, so it can cover many hosts and non-web protocols at once. The weakness is the same: the sensor has to see the traffic in readable form. Signatures are written against known behavior, so a variant can slip past one, and a flaw with no published signature has no protection. ## Worked case: PAN-OS management interface, CVE-2026-0322 On 14 Oct 2026 Palo Alto Networks published its advisory for CVE-2026-0322, a code injection flaw (CWE-94, CVSS 4.0 base score 8.6) in the PAN-OS management interface. An attacker who can reach the interface and holds a valid high-privilege account can run code as root. CISA added it to KEV the same day with a due date of 17 Oct 2026, and Palo Alto's exploitation status reads "Very limited exploitation has been observed targeting the Palo Alto Networks management interface." Every supported train has a fixed build; the per-train fixed builds are in the table on the [access restriction](https://remediationvsmitigation.com/compensating-controls/access-restriction/) page, and the table below compares the two options Palo Alto gives. Palo Alto states that no workaround exists, then names a signature: Threat Prevention subscribers can enable Threat IDs 510046 and 510047 on Applications and Threats content version 9159-10287 or later. Palo Alto calls this limited coverage, and it works only if all four of these hold: - management-port traffic is routed through a data-plane interface, using a management profile on that interface - the certificate for inbound traffic management is replaced - inbound traffic to the management interface is decrypted - threat prevention is enabled on that traffic The cost, per the same record, is SSL decryption load and certificate work on the firewall, and attack variants may still pass. What can be checked on your own sensor is the content version it runs: it must be 9159-10287 or later. Source: Palo Alto Networks advisory CVE-2026-0322, read 19 Nov 2026. | Option | What it needs | What stays open | |---|---|---| | Threat IDs 510046 and 510047 | The four conditions above and content 9159-10287 or later | Variants that do not match; the code remains | | Restrict the management interface to one jump box | See [access restriction](https://remediationvsmitigation.com/compensating-controls/access-restriction/) | Anyone with a high-privilege account who can reach the interface from the jump box; the code remains. Palo Alto rates the vector 8.5 instead of 8.6 | ## Three checks before you rely on it 1. Prevent, not detect. A signature set to alert logs the attack and lets it through. On PAN-OS, go to Objects > Security Profiles > Vulnerability Protection, open the profile the security rule uses, select the Exceptions tab and click Show all signatures. Filter for 510046 and 510047 and read the Action column: it must block (Palo Alto's documentation recommends reset-both for critical, high and medium severity) and not alert. Palo Alto's page notes that an exception can change the response to one signature, so look for an exception on these two IDs. 2. Placement. Confirm the sensor is inline on the path and that decryption happens upstream of it. For CVE-2026-0322 this is the whole control: without decryption of management traffic the signature sees nothing. 3. Freshness. Under Device > Dynamic Updates, find the Applications and Threats row with a check mark in the Currently Installed column. Its version must be 9159-10287 or later; anything older does not carry the Threat IDs. Then apply the two SP 800-40 Rev. 4 checks to the result: the control is "functioning as intended" (confirm a block log entry for the signature, not only that it is loaded), and the response "continues to be in place", so re-check after every content update, route change and policy push. What a block looks like in the log: in Monitor > Logs > Threat, an entry for these signatures has type THREAT and subtype vulnerability, the Threat ID 510046 or 510047, and an Action that blocks. Palo Alto's threat log field list defines the values: "alert" is "threat or URL detected but not blocked", while drop, reset-client, reset-server, reset-both and block-ip are blocking actions (reset-both sends a TCP RST to both client and server). An entry with Action alert means the signature fired and let the traffic through. Filter by the Threat ID column, not by name ([Palo Alto Networks, threat log fields](https://docs.paloaltonetworks.com/pan-os/11-1/pan-os-admin/monitoring/use-syslog-for-monitoring/syslog-field-descriptions/threat-log-fields), read 19 Nov 2026). ## Second case: a FortiOS virtual patch Fortinet advisory FG-IR-26-273 (CVE-2026-126259, published 10 Nov 2026, updated 16 Nov 2026) describes a missing authentication flaw [CWE-306] in FortiOS that may allow an unauthenticated attacker with HA FGSP + standalone-config-sync, or autoscale, to run commands as super admin via crafted HA packets. It is rated CVSS 3.1 9.0, Known Exploited: No, with fixes in FortiOS 8.0.1, 7.6.8, 7.4.13 and 7.2.14 and a migration to a fixed release for 7.0. It also states: "Virtual Patch named \"FG-VD-61445.0day.\" is available in FMWP db update 26.082" ([Fortinet PSIRT FG-IR-26-273](https://www.fortiguard.com/psirt/FG-IR-26-273), read 19 Nov 2026). The three checks carry over with Fortinet's names: the database on the device must be update 26.082 or later, the patch's action must block and not only log, and the device must see the traffic. Here the traffic is HA packets, and Fortinet says that on FGCP clusters exploitation requires an HA-segment adjacent attack vector, so a sensor that is not on the HA segment sees nothing. ## Blind spots and when to revisit - The sensor path changes: a new route, or a cloud migration that moves the workload past the sensor - A bypass variant is published - The date set for the real fix passes Exploited flaws raise the bar. A finding on CISA's KEV catalog has its own handling, set out in the [decision table row for a KEV finding with no patch](https://remediationvsmitigation.com/decision-table/kev-no-patch-yet/). Here CISA marks forensic triage as required, so the three addresses Palo Alto ties to its limited exploit attempts (66[.]42[.]72[.]14, 208[.]83[.]239[.]114 and 149[.]248[.]47[.]234) belong in a search of management and firewall logs whatever the signature does. How long a stand-in may last is on the [maximum durations page](https://remediationvsmitigation.com/risk-acceptance-form/maximum-durations/). ## What to record Control: IPS signature Finding: CVE-2026-0322 on , PAN-OS Filter IDs and action: Threat IDs 510046 and 510047, action Content version on the sensor: installed (needs 9159-10287 or later) Path conditions: management profile on data-plane interface ; certificate replaced ; inbound decryption on ; threat prevention on Policy export: ; proof of prevent mode, not a screenshot of the signature list Block log entry seen: Sensor placement diagram: Change ticket: Ends when: fixed build installed or review date ## Sources - Palo Alto Networks advisory CVE-2026-0322 (vendor): https://security.paloaltonetworks.com/CVE-2026-0322 (read 19 Nov 2026) - OWASP Virtual Patching Cheat Sheet, Virtual Patching Tools: https://cheatsheetseries.owasp.org/cheatsheets/Virtual_Patching_Cheat_Sheet.html (read 19 Nov 2026) - NIST SP 800-53 Rev. 5.2.0, SI-4: https://csrc.nist.gov/projects/cprt/catalog#/cprt/framework/version/SP_800_53_5_2_0/home/SI-4 (read 19 Nov 2026) - NIST SP 800-40 Rev. 4, Guide to Enterprise Patch Management Planning, Sections 2.1 and 2.2: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf (read 19 Nov 2026) - CISA Known Exploited Vulnerabilities catalog, feed version 2026.11.18: https://www.cisa.gov/known-exploited-vulnerabilities-catalog (read 19 Nov 2026) - Palo Alto Networks, Objects > Security Profiles > Vulnerability Protection (Exceptions tab, Action column): https://docs.paloaltonetworks.com/ngfw/help/11-1/objects/objects-security-profiles-vulnerability-protection (read 19 Nov 2026) - Palo Alto Networks, Device > Dynamic Updates (Currently Installed column): https://docs.paloaltonetworks.com/ngfw/help/11-1/device/device-dynamic-updates (read 19 Nov 2026) - Palo Alto Networks, threat log fields (Type, Subtype, Threat ID, Action values): https://docs.paloaltonetworks.com/pan-os/11-1/pan-os-admin/monitoring/use-syslog-for-monitoring/syslog-field-descriptions/threat-log-fields (read 19 Nov 2026) - Fortinet PSIRT FG-IR-26-273 (CVE-2026-126259): https://www.fortiguard.com/psirt/FG-IR-26-273 (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [WAF Rule as a Virtual Patch: Atlassian Case and Limits](https://remediationvsmitigation.com/compensating-controls/waf-rule/index.md) - [Exploited Flaw, No Patch Yet: Mitigate and Set Two Dates](https://remediationvsmitigation.com/decision-table/kev-no-patch-yet/index.md) - [Maximum Duration of a Risk Acceptance: 9 Sources](https://remediationvsmitigation.com/risk-acceptance-form/maximum-durations/index.md) - [Mitigation vs Remediation Wording in Vendor Advisories](https://remediationvsmitigation.com/guides/mitigation-vs-remediation-in-vendor-advisories/index.md) --- # Network segmentation as a compensating control Canonical: https://remediationvsmitigation.com/compensating-controls/network-segmentation/ Cisco's advisory of 16 Sep 2026 for ISE CVE-2026-76460 (CVSS 10.0) lists no workaround and one mitigation: infrastructure ACLs that allow only required management and control plane traffic to the ISE nodes. That is network segmentation. It narrows which networks reach the API and leaves the flaw in place until a fixed patch is installed. ## What segmentation is in the standards' words CISA names it for assets that cannot be updated in time: "Depending on your environment, appropriate isolation techniques may include decommissioning, removal of the vulnerable software product, network segmentation, isolation, software-defined perimeters, and proxies." ([BOD 26-04 implementation guidance](https://www.cisa.gov/news-events/directives/bod-26-04-implementation-guidance-prioritizing-security-updates-based-risk), read 19 Nov 2026.) NIST SP 800-40 Rev. 4 lists it under the Mitigate response: "using firewalls and network segmentation to isolate vulnerable assets, thus reducing the attack surface" (section 2.1), and for unpatchable assets points to micro-segmentation and software-defined perimeters in SP 800-207 (section 3.5.4) ([SP 800-40 Rev. 4](https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf), read 19 Nov 2026). SP 800-53 Rev. 5.2.0 gives the control text. SC-7 b: implement subnetworks for publicly accessible system components that are separated from internal organizational networks. SC-7(5): "Deny network communications traffic by default and allow network communications traffic by exception". SC-7(21): "Employ boundary protection mechanisms to isolate" the components the organization names; its discussion says isolation "provides the capability for increased protection of individual system components and to more effectively control information flows between those components." ([NIST OSCAL catalog of SP 800-53 Rev. 5.2.0](https://github.com/usnistgov/oscal-content/blob/main/nist.gov/SP800-53/rev5/json/NIST_SP-800-53_rev5_catalog.json), last modified 22 Oct 2026, read 19 Nov 2026.) ## When segmentation counts as a mitigation Conditions drawn from SP 800-53 SC-7(5) and the Cisco advisory, read 19 Nov 2026. | It counts when | It does not count when | |---|---| | The vendor or CISA names an access or ACL control, as Cisco does for ISE, and your rule is the one described. | The rule exists only on a diagram, or exists but a second path (another VLAN, a VPN pool, a management network) still reaches the component. | | The attacker's starting point is a network position the rule removes, such as the user or guest VLAN. | The attacker is already inside the allowed set: an admin workstation, another node, a compromised integration server. | | Rules are deny by default, with each allowed source listed and owned. | The rule is allow any with a few exceptions, so the allowed set is larger than anyone has counted. | | The flaw is reached over the network. Cisco's CVSS vector for CVE-2026-76460 is network, no privileges, no user interaction. | The flaw is triggered on the host, through a file or document, or through a service the segment must keep open. | ## Worked case: Cisco ISE CVE-2026-76460 Cisco Identity Services Engine and ISE-PIC have an API endpoint with insufficient authentication control. An unauthenticated attacker can bypass authentication, and Cisco says a successful attacker may obtain root privileges. Cisco published the advisory (cisco-sa-ISE-ABP-VNSW7Tn5) on 16 Sep 2026, states its PSIRT is aware of active exploitation, and gives first fixed releases of 3.1 Patch 12, 3.2 Patch 11, 3.3 Patch 12, 3.4 Patch 7 and 3.5 Patch 4. Release 3.0 has reached end of software maintenance, and Cisco advises migrating to a supported release. CISA added the CVE to KEV the same day with a due date of 19 Sep 2026 and a forensic triage requirement ([Cisco advisory](https://sec.cloudapps.cisco.com/security/center/content/CiscoSecurityAdvisory/cisco-sa-ISE-ABP-VNSW7Tn5), [CISA KEV feed 2026.11.18](https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json), read 19 Nov 2026). On workarounds Cisco writes: "There are no workarounds that address this vulnerability. However, there is a mitigation. To prevent remote exploitation of this vulnerability, use infrastructure access control lists (iACLs) to allow only required management and control plane traffic that is destined to the affected device." It adds that it "considers any workarounds and mitigations (if applicable) to be temporary solutions until an upgrade to a fixed software release is available." The setting is therefore an allow list on the network path to each ISE node, with everything else denied. The advisory gives no port list; the allowed set is yours to inventory. Cisco's advisory does not use the word segmentation. An ACL on the path decides which networks may reach the nodes, so the case sits here. [Access restriction](https://remediationvsmitigation.com/compensating-controls/access-restriction/) covers the other question, which identities and addresses may call a component, and works that through on the PAN-OS flaw CVE-2026-0322. CISA's SSVC data in NVD rates the flaw automatable and total impact, which is row 1 of BOD 26-04 Table 1 for an exposed asset: 3 days with forensic triage. Segmentation does not change that row unless it also removes public reachability, and even then row 9 is also 3 days with forensic triage, so the due date stays 19 Sep 2026. [Removing internet exposure](https://remediationvsmitigation.com/compensating-controls/remove-internet-exposure/) shows which rows move. ## What Cisco says to look for in access.log The ACL does nothing about a node that was already reached. Cisco's advisory gives this under Indicators of Compromise (read 19 Nov 2026): - Review access.log for suspicious usernames, on every node if the deployment is distributed. - Cisco's non-exhaustive example, run at the ISE CLI: `admin#show logging application ise-kong/access.log | include dummyuser`, where dummyuser stands in for a suspicious username. - For more access.log files, collect a support bundle with include debug logs selected and shared key encryption, then decrypt it and find the log at `./ise/logs/apigateway/access.log.gz`. - Cisco: "The presence of any entry in the output may indicate malicious activity." If activity is suspected, Cisco strongly recommends re-imaging the affected nodes and restoring from a configuration backup if needed. - On success the attacker may hold root, so evidence on the node may be removed or hidden. Cisco recommends cross-checking network and firewall logs outside the device for unexpected uploads from the node to external IP addresses and downloads from malicious IP addresses. The advisory lists no indicator addresses. ## What the ACL costs - Anything that talks to ISE from a network you did not list loses access. Inventory the callers first: administrator workstations or a jump host, the other ISE nodes in the deployment, the network devices that send authentication requests, and any integration server. An omission is an outage on the network access control system itself. - A change window and a rollback copy of the ACL on every device in the path, since a wrong deny cuts the system that decides who may join the network. - Ongoing ownership. The allowed list drifts when a new site, device group or integration appears; each addition is a change ticket against this entry. - Cisco states the limit plainly in its own terms: the ACL is temporary until the upgrade, and a compromised allowed source can still reach the API. ## A dated plan for a three-node deployment Illustration: ISE 3.4 at a patch level below Patch 7, three nodes, one admin jump host. Rows marked plan are choices for the example. The 17 Sep ACL leaves the nodes on the network and reachable from the jump host, the network devices and each other, so it is not removal from the network. CISA's guidance says that if the required action cannot be done in the timeframe, "remove the asset from the agency network", and names network segmentation among isolation techniques only for that purpose. This plan therefore does not treat the ACL as meeting the 19 Sep date, and the patch row below is marked late. Example schedule. Only 16 and 19 Sep are events; the other dates are chosen for the illustration. To meet 19 Sep, Patch 7 would have to be installed by that date or the nodes taken off the network. | Date | Event or plan | Record produced | |---|---|---| | 16 Sep 2026 | Event: Cisco advisory published, KEV entry added, due 19 Sep 2026. Plan: caller inventory started from firewall and ACL logs | Inventory sheet with owner per source | | 16 Sep 2026, later | Plan: allow list agreed with the network team; change ticket opened | Draft ACL, ticket, approver | | 17 Sep 2026 | Plan: ACL applied on the device in front of the nodes; reachability checked from the user VLAN, the guest network and the jump host | ACL export, dated; check results per vantage point | | 19 Sep 2026 | Event: KEV due date. Plan: forensic triage done with the access.log review above; Patch 7 window booked | Triage ticket, patch ticket | | 26 Sep 2026 | Plan: 3.4 Patch 7 installed on all three nodes and confirmed with show version. This is 7 days after the 19 Sep due date, so for an agency under BOD 26-04 it is late and the ACL does not excuse it; a non-federal operator records it as the end of the temporary control | Version output per node | | 3 Oct 2026 | Plan: ACL narrowed back to its standing form or removed, as the ticket says | Closing entry | ## How to verify the boundary 1. Export the rule set from the device that enforces it and date the export. Confirm the final rule is a deny and that no earlier rule widens the allowed set. 2. Check reachability of each ISE node's management and API interfaces from a user VLAN, a guest network and any internet-facing network you own, and from one allowed source. Record who could connect and who could not. This is a connection test of your own addresses, not a test of the flaw. 3. Compare the result with the inventory. Every allowed source should connect; every other source should not. 4. Review firewall logs for denied connection attempts to the nodes; they show callers you missed and traffic you meant to stop. Keep the outside-the-node log check from the access.log section with this record. 5. Repeat after every change to the path. ## What to record Network segmentation entry - Finding: CVE-2026-76460, ISE version and patch level per node - Source of the control: Cisco advisory cisco-sa-ISE-ABP-VNSW7Tn5 (iACL mitigation), read 19 Nov 2026 - Boundary device and rule set name; export with date - Allowed sources, each with an owner and a reason - Reachability check: vantage points, date, tester, result - Known gaps: sources inside the allowed set; other paths to the nodes - Change ticket, approver, rollback copy - Forensic triage ticket (KEV entries marked yes) - Fix date, review date, and the date the entry is retired SP 800-40 Rev. 4 says to plan to replace emergency mitigations with permanent fixes and to set schedules for both patch deployment and mitigation removal. Retire the entry when the fixed patch is confirmed on every node. If the ACL is kept, rename it as standing hardening so it is not mistaken for the temporary control. For callers narrowed by identity rather than network, see [access restriction](https://remediationvsmitigation.com/compensating-controls/access-restriction/), and verification steps are in [how to verify a mitigation works](https://remediationvsmitigation.com/guides/verifying-a-mitigation-works/). ## Sources - Cisco security advisory cisco-sa-ISE-ABP-VNSW7Tn5 (vendor): workarounds, fixed releases, indicators: https://sec.cloudapps.cisco.com/security/center/content/CiscoSecurityAdvisory/cisco-sa-ISE-ABP-VNSW7Tn5 (read 19 Nov 2026) - CISA KEV JSON feed, catalog 2026.11.18: CVE-2026-76460: https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) - NVD record for CVE-2026-76460 (CVSS vector, CISA ADP SSVC data): https://services.nvd.nist.gov/rest/json/cves/2.0?cveId=CVE-2026-76460 (read 19 Nov 2026) - CISA BOD 26-04 implementation guidance, FAQ on alternative mitigation (remove the asset from the agency network; isolation techniques): https://www.cisa.gov/news-events/directives/bod-26-04-implementation-guidance-prioritizing-security-updates-based-risk (read 19 Nov 2026) - CISA BOD 26-04, Appendix A, Table 1: https://www.cisa.gov/news-events/directives/bod-26-04-prioritizing-security-updates-based-risk (read 19 Nov 2026) - NIST SP 800-40 Rev. 4, sections 2.1, 3.5.3 and 3.5.4: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf (read 19 Nov 2026) - NIST SP 800-53 Rev. 5.2.0 OSCAL catalog (last modified 2026-10-22): SC-7, SC-7(5), SC-7(21): https://github.com/usnistgov/oscal-content/blob/main/nist.gov/SP800-53/rev5/json/NIST_SP-800-53_rev5_catalog.json (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Access Restriction as a Compensating Control: PAN-OS](https://remediationvsmitigation.com/compensating-controls/access-restriction/index.md) - [Remove Internet Exposure as a Mitigation: What It Buys](https://remediationvsmitigation.com/compensating-controls/remove-internet-exposure/index.md) - [How to Verify a Mitigation Works: Checks for 5 Real CVEs](https://remediationvsmitigation.com/guides/verifying-a-mitigation-works/index.md) --- # Remove internet exposure as a vulnerability mitigation Canonical: https://remediationvsmitigation.com/compensating-controls/remove-internet-exposure/ Removing a product from the internet is a mitigation CISA names in BOD 26-04, but it shortens the clock only in some rows of Table 1. For Atlassian's CVE-2026-21589, added to KEV on 15 Oct 2026, it moves 3 days with forensic triage to 14 days. For FortiMail's CVE-2026-104286 it changes nothing. ## What CISA says removal does BOD 26-04 defines the input this way: "Publicly Exposed is defined as Yes if the asset in scope of this directive is accessible to unauthenticated or untrusted entities via public networks, such as the internet, regardless of its physical or logical location; the value is No otherwise." Its notes to Table 1 then say: "one valid mitigation is to remove the system from the internet; that action changes the value of "Publicly Exposed" from Yes to No and will shift the required timeline for further action, such as remediation, back." ([CISA BOD 26-04](https://www.cisa.gov/news-events/directives/bod-26-04-prioritizing-security-updates-based-risk), Appendix A, read 19 Nov 2026.) Two limits sit in the same text. If any method used to check exposure finds the asset exposed, the value is Yes. And the implementation guidance leaves the final call to the agency, asset by asset ([BOD 26-04 implementation guidance](https://www.cisa.gov/news-events/directives/bod-26-04-implementation-guidance-prioritizing-security-updates-based-risk), read 19 Nov 2026). A product behind a VPN that only authenticated staff can use is for the agency to rate. A product that still answers an internet address is exposed. CISA's industrial control advisories give the same advice as a recommended practice: "Minimize network exposure for all control system devices and/or systems, ensuring they are not accessible from the internet", and when remote access is required, use a VPN while "recognizing VPNs may have vulnerabilities and should be updated to the most current version available". The same advisory adds: "CISA reminds organizations to perform proper impact analysis and risk assessment prior to deploying defensive measures." ([CISA ICSA-26-321-11](https://www.cisa.gov/news-events/ics-advisories/icsa-26-321-11), a republication of Siemens SSA-724261, read 19 Nov 2026.) ## Which rows of Table 1 move Table 1 has 16 rows built from four inputs: Publicly Exposed, In the KEV, Automatable by Adversary, and Technical Impact. Changing only the first input pairs each exposed row with the unexposed row that has the same other inputs. Read from CISA's published table on 19 Nov 2026: BOD 26-04 Table 1, Appendix A, read 19 Nov 2026. Days are calendar days. For a KEV entry whose automatable value is yes and whose impact is total, removal buys no time. | In the KEV / Automatable / Impact | Exposed (row, days) | Not exposed (row, days) | Effect of removal | |---|---|---|---| | Yes / Yes / Total | 1: 3 days & forensic triage | 9: 3 days & forensic triage | None | | Yes / Yes / Partial | 2: 3 days | 10: 14 days | 3 days become 14 | | Yes / No / Total | 3: 3 days & forensic triage | 11: 14 days | 3 days & forensic triage become 14 days | | Yes / No / Partial | 4: 14 days | 12: 14 days | None | | No / Yes / Total | 5: 3 days | 13: 60 days | 3 days become 60 | | No / Yes / Partial | 6: 14 days | 14: 60 days | 14 days become 60 | | No / No / Total | 7: 14 days | 15: fix on system upgrade | 14 days become the next scheduled upgrade | | No / No / Partial | 8: 60 days | 16: fix on system upgrade | 60 days become the next scheduled upgrade | ## Four KEV entries worked through Table 1 The automatable and impact values below are CISA's own, published in the NVD record (CISA Coordinator, SSVC version 2.0.3). The KEV due dates and forensic triage flags are from the KEV feed, catalog 2026.11.18. Each exposed row matches the KEV due date and flag, which is how the row was identified. CISA KEV feed 2026.11.18, NVD records for the four CVEs (CISA ADP SSVC data), and BOD 26-04 Table 1, all read 19 Nov 2026. | CVE and product | KEV added, due | Automatable, impact | Exposed row | Unexposed row | Clock change | |---|---|---|---|---|---| | CVE-2026-21589, Atlassian Data Center products | 15 Oct 2026, 18 Oct 2026; triage yes | No, total | 3: 3 days & forensic triage | 11: 14 days | Due date moves from 18 Oct to 29 Oct 2026 | | CVE-2026-104286, Fortinet FortiMail | 1 Oct 2026, 4 Oct 2026; triage yes | Yes, total | 1: 3 days & forensic triage | 9: 3 days & forensic triage | None | | CVE-2026-76460, Cisco Identity Services Engine | 16 Sep 2026, 19 Sep 2026; triage yes | Yes, total | 1: 3 days & forensic triage | 9: 3 days & forensic triage | None | | CVE-2026-19868, SonicWall SonicOS SSLVPN | 6 Nov 2026, 20 Nov 2026; triage no | No, partial | 4: 14 days | 12: 14 days | None | Only the Atlassian entry gains days. Exposure removal on the other three is still a sound way to cut the chance of compromise while the upgrade is booked, but under the directive it does not change the due date, and the forensic triage requirement on FortiMail and ISE stays. ## Worked case: Atlassian CVE-2026-21589 Atlassian published its advisory on 5 Oct 2026 and wrote: "Remove your instance from the internet until you can patch or apply mitigations, if possible. Instances accessible to the public internet, including those with user authentication, should be restricted from external network access until you can take action." CISA added the CVE to KEV on 15 Oct. A federal agency that had pulled its Confluence Data Center off the internet by 15 Oct, and could show it, would move from row 3 to row 11: the due date from 18 Oct to 29 Oct 2026 (14 days from 15 Oct), and the forensic triage text no longer attached to the row. Three facts keep that from being a bargain. The instance was reachable from 5 Oct, ten days before KEV, and Atlassian says it cannot confirm whether instances were affected and tells you to check all of them for signs of compromise, so the triage still happens. The fix is still due. And the cost is that external users and integrations lose access for as long as the instance is off the internet. A team that still needs those users can use Atlassian's [WAF rule](https://remediationvsmitigation.com/compensating-controls/waf-rule/) instead, which does not change the Table 1 row. FortiMail shows the other side. Fortinet's advisory FG-IR-26-175 lists, besides turning the IBE service off, "Disable access to the FortiMail webmail interface from the internet or limit the access only from trusted private network". The cost is that external users, including IBE recipients, lose webmail and the secure portal. Under Table 1 the due date stays 4 Oct 2026. Fortinet fixed its hosted FortiMail Cloud on 5 Oct 2026, so only self-managed appliances carry this finding. ## How to verify the exposure is gone 1. List every way the product was reached: public DNS names, NAT and port-forward rules, load balancer listeners, cloud security groups, CDN origins and VPN gateways. Remove or restrict each and keep the before and after export, dated. 2. From a network you do not control, such as a mobile hotspot, request the product's own address and hostname and record that nothing answers. This is a reachability check of your own address space, not a test of the flaw. 3. Check DNS and certificate records for stale names that still resolve to the product. 4. Update the asset's exposure tag. BOD 26-04 Phase III has agencies tag assets "that can be reached from outside the agency network and using a routable IP address" and report every seven days when reporting is not automated; the tag change is what moves the timeline. 5. Re-check after every network change and put the check on the change template. ## What to record Exposure removal entry - Finding: CVE, product, version, KEV added and due dates - Table 1 inputs: automatable, impact; exposed row and unexposed row, with days for each - What was removed: DNS names, NAT rules, listeners, security groups, with before and after exports and dates - Replacement access path (VPN, private link) and its owner and patch state - Outside-network check: date, vantage point, result - Exposure tag change date - Dependents checked: partners, integrations, remote support, monitoring - Forensic triage ticket where the KEV entry requires it - Fix date and the date exposure may return Exposure removal stops being enough when the business needs the product reachable again, when an internal path is the likely route in, or when the flaw is exploited before login by anyone already on the internal network. Then use [network segmentation](https://remediationvsmitigation.com/compensating-controls/network-segmentation/) or [access restriction](https://remediationvsmitigation.com/compensating-controls/access-restriction/), and see the decision row for a [KEV finding with no patch yet](https://remediationvsmitigation.com/decision-table/kev-no-patch-yet/). Checks on any mitigation are in [how to verify a mitigation works](https://remediationvsmitigation.com/guides/verifying-a-mitigation-works/). ## Sources - CISA BOD 26-04, Appendix A, Table 1 and notes: https://www.cisa.gov/news-events/directives/bod-26-04-prioritizing-security-updates-based-risk (read 19 Nov 2026) - CISA BOD 26-04 implementation guidance, FAQ on Publicly Exposed: https://www.cisa.gov/news-events/directives/bod-26-04-implementation-guidance-prioritizing-security-updates-based-risk (read 19 Nov 2026) - CISA ICS advisory ICSA-26-321-11 (republishes Siemens SSA-724261), recommended practices: https://www.cisa.gov/news-events/ics-advisories/icsa-26-321-11 (read 19 Nov 2026) - CISA KEV JSON feed, catalog 2026.11.18: https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) - NVD record for CVE-2026-21589 (CISA ADP SSVC data); records for the other three CVEs read the same day: https://services.nvd.nist.gov/rest/json/cves/2.0?cveId=CVE-2026-21589 (read 19 Nov 2026) - Atlassian security advisory for CVE-2026-21589 (vendor): https://confluence.atlassian.com/security/cve-2026-21589-arbitrary-file-access-vulnerability-impacts-multiple-products-1870495748.html (read 19 Nov 2026) - Fortinet PSIRT FG-IR-26-175 (vendor): https://fortiguard.fortinet.com/psirt/FG-IR-26-175 (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Network Segmentation as a Mitigation: Cisco ISE Case](https://remediationvsmitigation.com/compensating-controls/network-segmentation/index.md) - [Exploited Flaw, No Patch Yet: Mitigate and Set Two Dates](https://remediationvsmitigation.com/decision-table/kev-no-patch-yet/index.md) - [Access Restriction as a Compensating Control: PAN-OS](https://remediationvsmitigation.com/compensating-controls/access-restriction/index.md) - [How to Verify a Mitigation Works: Checks for 5 Real CVEs](https://remediationvsmitigation.com/guides/verifying-a-mitigation-works/index.md) --- # WAF virtual patch: what a rule covers and what it misses Canonical: https://remediationvsmitigation.com/compensating-controls/waf-rule/ Atlassian's advisory for CVE-2026-21589 (5 Oct 2026) names one WAF or proxy rule: block two dots next to a slash, backslash or double colon, in plain and URL-encoded form. A rule like that counts as a mitigation only in block mode, on every route to the product, until the fixed version is installed. ## When a WAF rule counts as a mitigation OWASP defines a virtual patch as "A security policy enforcement layer that detects and blocks matching exploitation attempts against a known vulnerability without changing the vulnerable code." It adds the limit in the next sentence: "A virtual patch reduces exposure only for traffic the enforcement layer inspects and attack variants its rules recognize. It does not remove the underlying vulnerability." ([OWASP Virtual Patching Cheat Sheet](https://cheatsheetseries.owasp.org/cheatsheets/Virtual_Patching_Cheat_Sheet.html), read 19 Nov 2026.) This page covers the register entry: one rule, one product, one set of records. Conditions drawn from the OWASP Virtual Patching Cheat Sheet, read 19 Nov 2026. | The rule counts when | The rule does not count when | |---|---| | The flaw is reached by an HTTP request with a shape a filter can match, such as a path sequence or an injected string. OWASP: virtual patching "is ideally suited for injection-type flaws". | The flaw is a logic or authentication error whose requests look like normal ones. OWASP says other attack types may not get "an adequate level of attack surface reduction". | | Every route to the product passes through the WAF or proxy that holds the rule, including the internal hostname and any second load balancer. | Some callers reach the product directly: internal clients, a second hostname, a non-HTTP protocol, or traffic the WAF does not decrypt. | | The rule is in block mode and a test shows it blocks. | The rule is in log-only mode, or a managed rule is disabled by default and nobody switched it on. | | The rule came from the vendor's advisory or was written for the flaw class and tested against encodings. | The rule matches only the one sample request from a report. OWASP says to resist this: "its long term value is significantly decreased". | ## The rule Atlassian named for CVE-2026-21589 CVE-2026-21589 lets an unauthenticated attacker read files under the web application root of Bitbucket, Confluence, Jira Service Management, Jira Software, Bamboo and Crowd Data Center, Crucible and Fisheye, if the attacker knows the exact file name and path. Atlassian published the advisory and fixed versions on 5 Oct 2026. CISA added the CVE to its Known Exploited Vulnerabilities catalog on 15 Oct 2026, ten days later, with a due date of 18 Oct 2026 and a forensic triage requirement ([CISA KEV feed, catalog 2026.11.18](https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json), read 19 Nov 2026). Fixed builds include Confluence Data Center 9.2.26 and 10.2.19, Jira Software Data Center 9.12.40, 10.3.26 and 11.3.12, and Crucible and Fisheye 4.9.15 ([Atlassian advisory](https://confluence.atlassian.com/security/cve-2026-21589-arbitrary-file-access-vulnerability-impacts-multiple-products-1870495748.html), read 19 Nov 2026). Atlassian's temporary measures, in its order: remove the instance from the internet if possible, then Option 1, a rule on the WAF or proxy layer. For Option 1 it says: "Block any URL containing this regex pattern", and "The intent of this regex is to block .. immediately adjacent to / , \ , or ::". Atlassian adds that rule implementation depends on your technology, naming a reverse proxy, AWS WAF and Cloudflare as examples, and tells you to test that the rule blocks that sequence and "handles the URL-encoded patterns". The pattern, as Atlassian printed it: (?is).*(?:/|\\|::|%(?:25)*(?:2f|5c)|(?::|%(?:25)*3a){2})(?:\.|%(?:25)*2e){2}(?:/|\\|::|%(?:25)*(?:2f|5c)|(?::|%(?:25)*3a){2}|;|%(?:25)*3b|$).* **Option 2: Tomcat RewriteValve, for Confluence, Jira Service Management, Jira Software, Bamboo and Crowd.** Atlassian's steps: back up the instance, then for each node in the Data Center cluster: 1. Shut down the node. 2. Find the server.xml file. On Confluence, Jira Service Management, Jira Software and Bamboo it is conf/server.xml. On Crowd it is apache-tomcat/conf/Catalina/localhost/crowd.xml or apache-tomcat/conf/server.xml, whichever holds your application's Context element. Copy the file as a backup. 3. Find the Context element for the application (its docBase attribute is a path that includes the product name; in the standard setup it is the only Context element) and add inside it, if not already there: 4. Find the WEB-INF directory: confluence/WEB-INF on Confluence, atlassian-jira/WEB-INF on Jira Service Management and Jira Software, bamboo/WEB-INF or atlassian-bamboo/WEB-INF on Bamboo, crowd-webapp/WEB-INF on Crowd. 5. In that directory, if rewrite.config exists, copy it as a backup and append the block below. If it does not exist, create it with the block below. 6. Restart the node. # Primary check: Tomcat's decoded, normalized request path (no query string). RewriteCond %{REQUEST_PATH} PATTERN RewriteRule (?s)^.*$ /?denied [F,L] # Raw-URI fallback RewriteCond %{REQUEST_URI} PATTERN RewriteRule (?s)^.*$ /?denied [F,L] In the block, PATTERN stands for the same regex printed under Option 1, character for character. Atlassian prints it on both conditions: the first checks Tomcat's decoded request path, the second the raw URI. A matching request is answered with a forbidden status. **Option 3: urlrewrite.xml, for Bitbucket Data Center only.** Back up the instance. On a clustered system apply the change to all nodes, and to every Bitbucket mirror and mirror farm node. Edit /app/WEB-INF/urlrewrite.xml and add a new rule element as the first rule in the file, before the other rules. The rule's from element holds the Option 1 pattern, its set element sets the status to 404, and its to element is /mvc/error404. Then restart Bitbucket Data Center. Atlassian's cost note covers both options: a legitimate URL with two dots next to a slash, backslash or double colon is denied too. CISA's required action for the entry is to apply mitigations in accordance with vendor instructions. Atlassian's Option 1 is such an instruction, so a correctly deployed rule answers the mitigation half of the KEV entry. It does not answer the forensic triage requirement, and Atlassian says it cannot confirm whether an instance was affected and tells you to check all affected instances for signs of compromise. ## What the rule costs - Atlassian's stated cost: any legitimate URL that contains two dots next to a slash, backslash or double colon is denied. List the product's own URLs and plugin URLs that match before the rule goes to block mode. - A change on the proxy or WAF layer, which usually serves other applications too. A rule written for one product but attached to a shared listener blocks those applications' requests as well. - Testing time. Atlassian asks for a test of plain and URL-encoded forms. OWASP warns that "A passing test for one payload does not establish complete protection". - Upkeep until the fix. SP 800-40 Rev. 4 says "Schedules should be set and enforced for both patch deployment and mitigation removal." ([NIST SP 800-40 Rev. 4](https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf), section 3.5.3, read 19 Nov 2026.) ## A dated plan for one Confluence node Illustration: a team runs Confluence Data Center below 10.2.19 behind a reverse proxy. Rows marked plan are choices made for the example, not events. The dates marked event come from the sources above. Example schedule. The 25 and 27 Oct dates are chosen for the illustration. | Date | Event or plan | Record produced | |---|---|---| | 5 Oct 2026 | Event: Atlassian advisory and fixed versions published | Advisory text quoted in the finding, with read date | | 15 Oct 2026 | Event: KEV entry added, due 18 Oct 2026. Plan: rule deployed in log-only mode on the proxy in front of Confluence | Rule ID, version, mode, change ticket | | 15 Oct 2026, later | Plan: review the log for hits from real users; run the owner's test requests; list URLs that match | Test record with date and tester, list of routes covered | | 16 Oct 2026 | Plan: rule moved to block mode and re-tested | Mode change entry, second test record | | 18 Oct 2026 | Event: KEV due date. Plan: forensic triage of the node under way, upgrade window booked | Triage ticket, upgrade ticket | | 25 Oct 2026 | Plan: Confluence upgraded to 10.2.19 and version confirmed | Version check with date | | 27 Oct 2026 | Plan: rule removed, or kept as a named hardening rule with a new owner | Removal ticket | ## How to verify the rule took 1. Read the rule back from the WAF or proxy configuration and confirm its mode is block, its pattern matches Atlassian's text character for character, and its scope names every hostname that reaches the product. 2. Send one request of the matching shape, using only a harmless path on a system you own, and confirm the response is the block action and that the WAF log records the rule ID. Repeat once with the URL-encoded form and once with the double-encoded form, as Atlassian asks. 3. Compare the proxy's route list with the product's real entry points, including the internal hostname, so that no path skips the rule. 4. Re-run the check after every proxy or WAF change; add it to the change template. ## Other WAF rules named in the advisories on this site Fortinet advisory FG-IR-26-175 and Fastly blog of 16 Sep 2026, read 19 Nov 2026. The Fastly row is the WAF vendor's claim. | CVE | Who named the rule | The rule | Limit stated | |---|---|---|---| | CVE-2026-104286, Fortinet FortiMail (KEV 1 Oct 2026) | Fortinet, advisory FG-IR-26-175 | If a web application firewall sits in front of FortiMail, block POST requests to /ibe that contain '../' | Listed as an alternative to disabling the IBE service or limiting webmail access to trusted private networks; the upgrade is the fix | | CVE-2026-75650, Adobe Commerce and Magento Open Source (KEV 8 Sep 2026) | Fastly (a WAF vendor's own claim) | A virtual patch for Next-Gen WAF covering each stage of the exploit chain | Fastly says it is defense in depth against variants and gives operators time to patch; exploitation began as early as 4 Sep 2026, three days before Adobe's bulletin on 7 Sep | ## What to record WAF rule entry - Finding: CVE-2026-21589, Confluence Data Center, host and version - Source of the rule: Atlassian advisory, Option 1, read 19 Nov 2026 - Rule ID, version, location (proxy or WAF name), attached hostnames and routes - Mode and the date it changed from log-only to block - Test record: date, tester, plain, encoded and double-encoded results - URLs found that match the pattern and how they were handled - Change ticket and approver - Forensic triage ticket (KEV entries marked yes) - Fix date and fixed version - Date the rule is removed or re-owned, and who confirms it Retire or replace the rule when the fixed version is confirmed, when a test finds a path around it (OWASP: go back to analysis), or when the removal date passes with no fix. For the general decision between a rule and a fix, see the companion entries for [IPS signatures](https://remediationvsmitigation.com/compensating-controls/ips-signature/) and [removing internet exposure](https://remediationvsmitigation.com/compensating-controls/remove-internet-exposure/). ## Sources - Atlassian security advisory for CVE-2026-21589 (vendor), mitigation options 1 to 3: https://confluence.atlassian.com/security/cve-2026-21589-arbitrary-file-access-vulnerability-impacts-multiple-products-1870495748.html (read 19 Nov 2026) - CISA KEV JSON feed, catalog 2026.11.18: CVE-2026-21589 and CVE-2026-104286: https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) - OWASP Virtual Patching Cheat Sheet: https://cheatsheetseries.owasp.org/cheatsheets/Virtual_Patching_Cheat_Sheet.html (read 19 Nov 2026) - NIST SP 800-40 Rev. 4, section 3.5: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf (read 19 Nov 2026) - Fortinet PSIRT FG-IR-26-175 (vendor), workaround list: https://fortiguard.fortinet.com/psirt/FG-IR-26-175 (read 19 Nov 2026) - Fastly, what is Style Smuggler, CVE-2026-75650 (WAF vendor): https://www.fastly.com/blog/what-is-style-smuggler-cve-2026-75650 (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [When Virtual Patching Is Enough, and When It Is Not](https://remediationvsmitigation.com/guides/virtual-patching-and-when-it-is-enough/index.md) - [IPS Signature as a Mitigation: When It Counts](https://remediationvsmitigation.com/compensating-controls/ips-signature/index.md) - [Remove Internet Exposure as a Mitigation: What It Buys](https://remediationvsmitigation.com/compensating-controls/remove-internet-exposure/index.md) - [How Virtual Patching Works: Six Phases and a Real Rule](https://remediationvsmitigation.com/guides/how-virtual-patching-works/index.md) - [How to Verify a Mitigation Works: Checks for 5 Real CVEs](https://remediationvsmitigation.com/guides/verifying-a-mitigation-works/index.md) --- # Contact Canonical: https://remediationvsmitigation.com/contact/ Mail editor@remediationvsmitigation.com for corrections, a source the site should read, or a question about reuse under CC BY 4.0. There is no contact form, because the site stores nothing from visitors; a message exists only in the mailbox. First edition 19 Nov 2026. ## Where to write The one address is [editor@remediationvsmitigation.com](mailto:editor@remediationvsmitigation.com). The site has no contact form, no chat widget and no tracking script, so a message you send reaches only that mailbox and the mail systems that carry it. The decision tool and the risk acceptance form run in the browser; nothing typed into them is transmitted, so there is nothing to attach from them. Four reasons to write and what to include. Read 19 Nov 2026. | You want to | Put in the message | Where it is handled | |---|---|---| | Report an error | Page address, the sentence, the source URL with the right text, the date you read it | [Corrections](https://remediationvsmitigation.com/corrections/) | | Suggest a source | The document, its URL, and the situation or control it bears on | [Editorial policy](https://remediationvsmitigation.com/editorial-policy/) | | Reuse data or a form | What you will reuse and where the credit line will sit | [Terms](https://remediationvsmitigation.com/terms/) | | Ask about log data | The date and time range, to match against CloudFront access log entries | [Privacy](https://remediationvsmitigation.com/privacy/) | ## What helps a report get checked - Quote the exact text you think is wrong, not a summary of it. - Give the source URL and the date you read it. Directives such as CISA BOD 26-04 and rules such as FedRAMP's 2026 consolidated rules can change, and the first edition reads them as of 19 Nov 2026. - Say which situation or control the point bears on, for example the nine compensating control entries. - Leave out exploit code, payloads and live vulnerability details; the site does not publish them and a report does not need them. ## What the address does not do It does not give case-by-case advice on whether to patch, mitigate or accept a particular finding; the pages report where published guidance points, and the [decision table](https://remediationvsmitigation.com/decision-table/) shows the sources. A message and your address are kept in the mailbox like any email. Nothing is added to a mailing list. See the [privacy page](https://remediationvsmitigation.com/privacy/) for what the web server logs and for how long. ## Sources - AWS, CloudFront standard logging reference: https://docs.aws.amazon.com/AmazonCloudFront/latest/DeveloperGuide/standard-logs-reference.html (read 19 Nov 2026) - CISA, BOD 26-04: https://www.cisa.gov/news-events/directives/bod-26-04-prioritizing-security-updates-based-risk (read 19 Nov 2026) - FedRAMP, Consolidated Rules 2026: https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-detection-and-response/ (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Corrections Log: How to Report an Error](https://remediationvsmitigation.com/corrections/index.md) - [Privacy: Cookies, Logs and the 400-Day Limit](https://remediationvsmitigation.com/privacy/index.md) - [Terms of Use and CC BY 4.0 License](https://remediationvsmitigation.com/terms/index.md) --- # Corrections Canonical: https://remediationvsmitigation.com/corrections/ No correction has been made since the first edition on 19 Nov 2026; the log below is empty. A factual error, a broken source link or a quote that no longer matches its source can be reported to editor@remediationvsmitigation.com. Each fix is logged with its date and the old and new text. ## Corrections log Corrections made after the first edition of 19 Nov 2026. Empty on 19 Nov 2026. | Date | Page | Was | Now | Source of the fix | |---|---|---|---|---| | None so far | | | | | ## What counts as an error - A quote that does not match the source text on the read date, or a quote cut so that its meaning changes. - A number with a wrong value or without its source, such as a KEV count that no longer matches catalog version 2026.11.18 as read. - A source URL that fails or points to a different document. - A label that puts a situation in the wrong response, with the source text that shows it. ## What to send Mail [editor@remediationvsmitigation.com](mailto:editor@remediationvsmitigation.com) with the page address, the sentence, the source URL that shows the right text and the date you read it. Opinions about which response is wise for a particular organization are not corrections; the pages report where published guidance points. ## What happens next A confirmed error is fixed on the page, the page's modified date changes, and a row is added to the log with the old and new text. A report that turns out not to be an error gets a reply with the source text that supports the page. No response time is promised. The rules for sources are on the [editorial policy](https://remediationvsmitigation.com/editorial-policy/); other ways to reach the site are on the [contact page](https://remediationvsmitigation.com/contact/). ## Sources - CISA KEV catalog JSON feed, catalogVersion 2026.11.18: https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Editorial Policy: Sources, Labels and Corrections](https://remediationvsmitigation.com/editorial-policy/index.md) - [Contact: editor@remediationvsmitigation.com](https://remediationvsmitigation.com/contact/index.md) - [About Remediation vs Mitigation: Scope and Hosting](https://remediationvsmitigation.com/about/index.md) --- # Remediate, mitigate or accept: the Decision Table Canonical: https://remediationvsmitigation.com/decision-table/ Thirteen situations, each with the decision, the deadline, who signs and the record to keep (read 19 Nov 2026). Published guidance points to remediation in 3, mitigation now with a dated fix in 7, a dated exception in 2 and verify-then-close in 1. CISA BOD 26-04 grants federal agencies no waiver. ## The 13 situations **Decision** is the response the cited text points to most directly. It is not an order of preference. **Deadline** is a date or interval a source sets, and the cell says so where none does. **Who signs** names the role a source gives. Each linked situation page quotes the rule, works one real CVE through it and says when the usual answer is wrong. The sources for each linked row are named on its page. Decision Table, 13 situations. Sources: CISA BOD 26-04 and its implementation guidance, CISA BOD 26-02, FedRAMP Consolidated Rules for 2026, PCI SSC FAQs 1046 and 1597, NIST SP 800-40 Rev. 4 and SP 800-37 Rev. 2. Read 19 Nov 2026. | Situation | Decision | Deadline | Who signs | Record to keep | |---|---|---|---|---| | [A patch exists, the flaw is on the KEV catalog and the asset is internet-reachable](https://remediationvsmitigation.com/decision-table/kev-patch-available-exposed/) | Remediate | Federal: 3 days from the KEV listing or the agency's own detection, whichever comes first; 14 days where exploitation is not automatable and control is partial. Forensic triage rides with the 3-day rows for total control | No one for a federal agency: CISA issues no waivers. Elsewhere, the risk owner | Clock start, fix date, forensic triage report | | [Exploited or severe, and the vendor has no patch yet](https://remediationvsmitigation.com/decision-table/kev-no-patch-yet/) | Mitigate now, fix later | Mitigation at once. NIST: set and enforce schedules for patch deployment and for mitigation removal. A KEV entry keeps its due date | The risk owner. On a federal system the authorizing official accepts what remains | Mitigation, its removal date, planned patch date | | [A patch would break a critical system](https://remediationvsmitigation.com/decision-table/patch-breaks-critical-system/) | Mitigate now, fix later | The next outage window after testing. On a federal KEV entry the due date stands | The risk owner for the delay. No one for a federal KEV entry | Regression result, outage window, interim control | | [The software is end of life](https://remediationvsmitigation.com/decision-table/end-of-life-software/) | Remediate (replace or remove) | Federal edge devices under BOD 26-02 (issued 5 Feb 2026): identified end-of-support devices decommissioned within 18 months, by 5 Aug 2027. Elsewhere, the replacement date you set | The authorizing official or risk owner, for any SA-22 exception | Replacement date, isolation evidence, SA-22 reason | | [Not reachable, not on the KEV catalog, not automatable](https://remediationvsmitigation.com/decision-table/not-reachable-low-risk/) | Accept with a dated exception | Fix on system upgrade: the next scheduled major upgrade or rebuild. The timeline restarts when exposure, KEV status or automation changes | The risk owner, within stated tolerance. In BOD 26-04 scope the timeline is the rule | Exposure basis, the upgrade, the conditions that reopen it | | [SaaS or a managed service you cannot patch](https://remediationvsmitigation.com/decision-table/saas-cannot-patch/) | Mitigate now, fix later | The provider's fix. FedRAMP providers must categorize a flaw not mitigated or remediated within 192 days of evaluation as accepted | The provider records it. The agency authorizing official signs any agency POA&M | Provider notice, your configuration changes, provider status date | | [A control already sits in front](https://remediationvsmitigation.com/decision-table/compensating-control-in-front/) | Mitigate now, fix later | A fix date written the day the control goes in. A KEV due date still stands; FedRAMP's 192 days is the outer limit | The risk owner. In PCI DSS the QSA judges the control | Control, test date, exposure value, fix date | | [The asset is scheduled to be retired](https://remediationvsmitigation.com/decision-table/decommission-scheduled/) | Accept with a dated exception | The decommission date, no later. On a federal KEV entry the due date stands | The risk owner. The authorizing official on a federal system | Decommission date, interim control, confirmation of removal | | A PCI DSS system cannot meet a requirement | Mitigate now, fix later | Critical patches within one month of release (Requirement 6.3.3, PCI SSC FAQ 1597) | The QSA judges a compensating control. PCI SSC does not approve one | Assessor's finding, corrective actions, targeted risk analysis | | [A medical or OT device cannot be patched on cycle](https://remediationvsmitigation.com/decision-table/ot-medical-cannot-patch/) | Mitigate now, fix later | None set. NIST: reevaluate the alternatives to patching periodically | Security architects review the long-term mitigations in advance. The risk owner accepts what remains | Approved long-term methods, re-evaluation date, impact analysis | | A federal agency cannot meet a KEV date | Mitigate: remove the asset from the network | The due date stands. Removing exposure moves the asset to another Table 1 row | No one: CISA issues no waivers. Agencies contact CISA about special cases | Removal from the network, contact with CISA, new clock | | [Internal asset, patch exists, not on the KEV catalog, automatable](https://remediationvsmitigation.com/decision-table/internal-asset-not-kev-moderate/) | Remediate | 60 days (Table 1 rows 13 and 14) | No one for a federal agency. Elsewhere, the risk owner | Clock start, 60-day fix date, reason for any exception | | [The scanner reports a flaw that is not there](https://remediationvsmitigation.com/guides/false-positive-handling/) | Verify, then close | None set | No risk is accepted. FedRAMP's legacy playbook holds a false positive that is not validated as pending approval | Evidence of absence, validator, closing date | ## Federal deadlines in BOD 26-04 Table 1 Most deadline cells above come from this table. CISA publishes it as an image; the rows below are transcribed from that image. Days are calendar days. In the directive's words, "The timelines for each asset are dynamic as facts change": removing a system from the internet changes "Publicly Exposed" from Yes to No and moves the asset to another row, while a new KEV listing shortens the timeline. The clock starts when CISA adds the flaw to the KEV catalog or the agency identifies it on an asset, whichever comes first. BOD 26-04 Table 1, Remediation Timelines, transcribed 19 Nov 2026. Totals: 5 rows at 3 days (3 of them with forensic triage), 6 at 14 days, 3 at 60 days, 2 at fix on system upgrade. | Row | Publicly exposed | In the KEV | Automatable by adversary | Technical impact | Agency timeline | |---|---|---|---|---|---| | 1 | Yes | Yes | Yes | Total | 3 days and forensic triage | | 2 | Yes | Yes | Yes | Partial | 3 days | | 3 | Yes | Yes | No | Total | 3 days and forensic triage | | 4 | Yes | Yes | No | Partial | 14 days | | 5 | Yes | No | Yes | Total | 3 days | | 6 | Yes | No | Yes | Partial | 14 days | | 7 | Yes | No | No | Total | 14 days | | 8 | Yes | No | No | Partial | 60 days | | 9 | No | Yes | Yes | Total | 3 days and forensic triage | | 10 | No | Yes | Yes | Partial | 14 days | | 11 | No | Yes | No | Total | 14 days | | 12 | No | Yes | No | Partial | 14 days | | 13 | No | No | Yes | Total | 60 days | | 14 | No | No | Yes | Partial | 60 days | | 15 | No | No | No | Total | Fix on system upgrade | | 16 | No | No | No | Partial | Fix on system upgrade | The directive defines remediation as eliminating the vulnerability "through patching, decommissioning the system, or another action". "Fix on system upgrade" means remediation "the next time the vulnerable asset receives a scheduled major upgrade or rebuild". These two definitions are why the table treats retirement as a fix and a compensating control as a step. ## Who signs depends on the rule that binds you For a federal system under NIST's risk management framework, "The explicit acceptance of risk is the responsibility of the authorizing official and cannot be delegated to other officials within the organization." ([SP 800-37 Rev. 2](https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-37r2.pdf), Task R-4). For BOD 26-04 deadlines nobody signs: "CISA does not issue waivers or exceptions for actions required in cyber directives." ([implementation guidance](https://www.cisa.gov/news-events/directives/bod-26-04-implementation-guidance-prioritizing-security-updates-based-risk)). In PCI DSS the entity proposes a compensating control and the assessor judges it: PCI SSC "is not able to approve specific configurations or compensating controls" ([FAQ 1046](https://www.pcisecuritystandards.org/faqs/1046)). Everyone else decides inside its own tolerance: NIST SP 800-39 says organizations "determine acceptable degrees of residual risk based on organizational risk tolerance and the specific risk tolerances of particular decision makers". Titles for that last case are on the [who may sign page](https://remediationvsmitigation.com/risk-acceptance-form/who-may-sign/). ## KEV due dates since BOD 26-04 In catalog 2026.11.18 (1,786 entries), 169 entries were added on or after 10 Jun 2026, the date of BOD 26-04. For 131 of them the due date is 3 days after the add date and for 38 it is 14 days. None is at 60 days. The feed marks 108 of the 169 for forensic triage and 61 not. Counted from the CISA feed on 19 Nov 2026. How the older required-action wording splits is on the [KEV required actions guide](https://remediationvsmitigation.com/guides/kev-required-actions-mitigate-or-discontinue/). ## What the table does not decide Only federal civilian agencies are bound by BOD 26-04. For everyone else CISA "strongly recommends all organizations review and monitor the KEV catalog and prioritize remediation of the listed vulnerabilities". FedRAMP's 2026 rules go further for cloud providers: "A fully mitigated vulnerability will still exist (with negligible risk) until it has been remediated." That is why mitigation is a step in most rows and the end state in none. Where a row ends in an exception, the [risk acceptance form](https://remediationvsmitigation.com/risk-acceptance-form/) and the [exception register](https://remediationvsmitigation.com/exception-register/) hold the fields and the [maximum durations page](https://remediationvsmitigation.com/risk-acceptance-form/maximum-durations/) lists every limit found. The definitions behind the labels are in [remediation versus mitigation](https://remediationvsmitigation.com/guides/difference-between-remediation-and-mitigation/). The table is also available as [CSV](https://remediationvsmitigation.com/decision-table.csv) and [JSON](https://remediationvsmitigation.com/decision-table.json). ## Sources - CISA, BOD 26-04, Table 1 and Appendix A: https://www.cisa.gov/news-events/directives/bod-26-04-prioritizing-security-updates-based-risk (read 19 Nov 2026) - CISA, BOD 26-04 implementation guidance, FAQ: https://www.cisa.gov/news-events/directives/bod-26-04-implementation-guidance-prioritizing-security-updates-based-risk (read 19 Nov 2026) - CISA, BOD 26-02, Required Actions: https://www.cisa.gov/news-events/directives/bod-26-02-mitigating-risk-end-support-edge-devices (read 19 Nov 2026) - CISA, KEV JSON feed, catalog 2026.11.18: https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) - CISA, Known Exploited Vulnerabilities Catalog: https://www.cisa.gov/known-exploited-vulnerabilities (read 19 Nov 2026) - FedRAMP, Vulnerability Detection and Response rules 2026, VDR-CSO-RES: https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-detection-and-response/ (read 19 Nov 2026) - FedRAMP, Vulnerability Evaluation and Reporting rules 2026, VER-TFR-MAV: https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-evaluation-and-reporting/ (read 19 Nov 2026) - NIST SP 800-40 Rev. 4, Sections 2.1, 3.5.3 and 3.5.4: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf (read 19 Nov 2026) - NIST SP 800-37 Rev. 2, Task R-4: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-37r2.pdf (read 19 Nov 2026) - NIST SP 800-39, Task 3-3: https://nvlpubs.nist.gov/nistpubs/Legacy/SP/nistspecialpublication800-39.pdf (read 19 Nov 2026) - PCI SSC, FAQ 1046: https://www.pcisecuritystandards.org/faqs/1046 (read 19 Nov 2026) - PCI SSC, FAQ 1597: https://www.pcisecuritystandards.org/faqs/1597 (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Risk Acceptance Form: Free Template With 15 Fields](https://remediationvsmitigation.com/risk-acceptance-form/index.md) - [Exception Register: 15 Columns, States and Blank CSV](https://remediationvsmitigation.com/exception-register/index.md) - [Difference Between Remediation and Mitigation in IT Security](https://remediationvsmitigation.com/guides/difference-between-remediation-and-mitigation/index.md) - [Who Can Accept Risk: Signer by Standard (2026)](https://remediationvsmitigation.com/risk-acceptance-form/who-may-sign/index.md) - [Vulnerability False Positive: Close It With Evidence](https://remediationvsmitigation.com/guides/false-positive-handling/index.md) --- # A compensating control already sits in front of the flaw Canonical: https://remediationvsmitigation.com/decision-table/compensating-control-in-front/ Treat the finding as mitigated, not fixed, and write a fix date the day the control goes in. FedRAMP's 2026 rules say a fully mitigated vulnerability still exists until remediated, and a KEV due date stands even then. NIST SP 800-40 Rev. 4 lets an organization accept the risk by relying on existing controls. Guidance points to: Mitigate now, fix later. Also allowed: Accept with a dated exception. Signals: Exploit path blocked at a choke point; The vulnerable code or version is unchanged. ## The situation A vendor fix exists and is not yet installed. Between an attacker and the vulnerable component sits a control you operate: a WAF rule, a network segment, an access restriction. The row applies once the control has been tested against this flaw. A control nobody tested is not yet a control, and the finding stays on whichever row its other facts select. ## The decision and its conditions Mitigate now, fix later. The control covers the days before the fix, and the fix date is written when the control goes in, not after. Accept with a dated exception is the other allowed response, and it fits only where no fix will come, such as a vendor that never ships one (see the [risk acceptance form](https://remediationvsmitigation.com/risk-acceptance-form/)). The decision holds while four conditions hold: 1. The control was tested against this specific flaw, with a date and a named tester. 2. The control sits on every path to the component, not only the main one. 3. A fix date, an owner and a date to remove the control are on the record. 4. The control is tested again whenever its rule set, the product version or the network path changes. ## What the sources say a control is worth Four issuers, six statements about a control in front of a flaw. Quotes read 19 Nov 2026. | Source | What it says a control does | Quote | |---|---|---| | NIST SP 800-40 Rev. 4, Section 2.1 | Can support acceptance | "Accept the risk from vulnerable software as is, such as by relying on existing security controls to prevent vulnerability exploitation" | | FedRAMP VER-EVA-EIR | Can change the exposure label | "once this prevention is in place the vulnerability should no longer be considered an internet-reachable vulnerability." | | FedRAMP VDR-CSO-RES | Leaves the flaw in place | "A fully mitigated vulnerability will still exist (with negligible risk) until it has been remediated." | | FedRAMP VDR-TFR-KEV | Does not excuse a KEV due date | "Providers SHOULD remediate Known Exploited Vulnerabilities according to the due dates in the CISA Known Exploited Vulnerabilities Catalog (even if the vulnerability has been fully mitigated)" | | OWASP Virtual Patching Cheat Sheet | Covers only what it inspects | "A virtual patch reduces exposure only for traffic the enforcement layer inspects and attack variants its rules recognize." | | PCI SSC FAQ 1046 | Is judged by the assessor | "QSAs are responsible to determine whether a compensating control is sufficient to meet the intent of a requirement" | NIST and FedRAMP are not in conflict once each is read for its job. NIST describes what an organization may decide about its own risk. FedRAMP and CISA describe what a regulated party must be able to show. The same control can justify acceptance in a private risk register and still not satisfy a KEV due date. CISA's guidance goes further for agencies: "Aside from removing affected assets from the network, the only known technical mitigation to vulnerabilities in the KEV Catalog is to apply the required actions listed in the catalog." BOD 26-04 also sets the exposure value harder than FedRAMP does. "Publicly Exposed" is Yes if the asset is "accessible to unauthenticated or untrusted entities via public networks", and if any one method finds it exposed the value is Yes. A filter in front of an asset reachable from the internet is therefore not on its own a change to No, while FedRAMP lets a prevention in place drop the internet-reachable label. A restriction that leaves no public path at all is the case where the BOD 26-04 value can change, as the worked case shows. ## Worked case: CVE-2026-0322 on a PAN-OS firewall CVE-2026-0322 is code injection in the PAN-OS management interface (CVSS 4.0 base score 8.6, from Palo Alto), worked through on the [access restriction](https://remediationvsmitigation.com/compensating-controls/access-restriction/) page. Here a firewall on 11.2.12 has its management interface open to the internet and gets the jump-box restriction; the fix is 11.2.13-h3 or 11.2.16. CVE-2026-0322, PAN-OS on PA-Series, VM-Series and Panorama, read 19 Nov 2026. CISA's SSVC data on the record reads exploitation active, automatable no, technical impact total. | Date | Event | Basis | |---|---|---| | 14 Oct 2026 | Palo Alto publishes the advisory with exploit maturity ATTACKED and notes very limited exploitation. CISA adds the CVE to the KEV catalog the same day, due 17 Oct 2026, forensic triage required | Palo Alto advisory; CISA KEV feed | | 14 Oct 2026 | The team restricts the management interface to one jump box and tests from an outside network that the address no longer answers | Control and test date assumed | | 17 Oct 2026 | KEV due date, 3 days after listing | CISA KEV feed | | 28 Oct 2026 | Day 14. Table 1 date for a federal agency whose exposure value is now No: exposed, KEV, not automatable, total control is row 3 (3 days); unexposed is row 11 (14 days) | BOD 26-04 Table 1, counted | | 24 Apr 2027 | Outer limit if the team were a FedRAMP provider and its evaluation date were 14 Oct 2026: 192 days later | VER-TFR-MAV, counted | The control changes the answer differently under each rule. A federal agency that can show no other method finds the interface public moves from row 3 to row 11: 28 Oct instead of 17 Oct. The agency, not the catalog, makes the final exposure call, so the evidence of the test is the record. A FedRAMP provider gets no such move: VDR-TFR-KEV tells it to remediate by the CISA date, 17 Oct, "even if the vulnerability has been fully mitigated". Either way the fix date stays. ## What to record - The control, where it sits on the traffic path, and the flaw it covers (CVE ID). - The test: the variants tried, the result, the date and the tester. - What the control does not cover, such as an internal route to the same service. - The exposure value under the rule that binds you: internet-reachable or not under FedRAMP, Publicly Exposed under BOD 26-04. - The fix date, the owner, the date the control is removed and the date of the last re-test. - For a KEV entry, the due date and the log review that covers the time before the control existed. Entries for a [WAF rule](https://remediationvsmitigation.com/compensating-controls/waf-rule/) and [network segmentation](https://remediationvsmitigation.com/compensating-controls/network-segmentation/) show these fields filled in. To prove the control holds, see [how to verify a mitigation](https://remediationvsmitigation.com/guides/verifying-a-mitigation-works/). ## When the usual answer is wrong - **The flaw is added to the KEV catalog.** The due date replaces your fix date, and a provider's control does not move it, as on 14 Oct above. Go to [patch available and exposed](https://remediationvsmitigation.com/decision-table/kev-patch-available-exposed/). - **The control sees only part of the traffic.** OWASP adds that "Code level fixes and Virtual Patching are NOT mutually exclusive" and that a passing test for one payload does not establish complete protection. A restriction on one interface does nothing for a path that reaches the service another way. Palo Alto's Threat IDs 510046 and 510047 for this flaw are called limited coverage by Palo Alto. - **The control fails its re-test.** The finding returns to the row its other facts select, with no credit for the control. - **The vendor never ships a fix.** Then the control becomes the stated mitigating factor in a dated exception, and the [residual risk after mitigation](https://remediationvsmitigation.com/guides/residual-risk-after-mitigation/) is what the approver signs for. ## Sources - NIST SP 800-40 Rev. 4, Section 2.1: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf (read 19 Nov 2026) - FedRAMP, VER-EVA-EIR and VER-TFR-MAV: https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-evaluation-and-reporting/ (read 19 Nov 2026) - FedRAMP, VDR-CSO-RES and VDR-TFR-KEV: https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-detection-and-response/ (read 19 Nov 2026) - OWASP, Virtual Patching Cheat Sheet: https://cheatsheetseries.owasp.org/cheatsheets/Virtual_Patching_Cheat_Sheet.html (read 19 Nov 2026) - PCI SSC, FAQ 1046: https://www.pcisecuritystandards.org/faqs/1046 (read 19 Nov 2026) - CISA, BOD 26-04, Appendix A, Publicly Exposed: https://www.cisa.gov/news-events/directives/bod-26-04-prioritizing-security-updates-based-risk (read 19 Nov 2026) - CISA, BOD 26-04 implementation guidance, alternative mitigation FAQ: https://www.cisa.gov/news-events/directives/bod-26-04-implementation-guidance-prioritizing-security-updates-based-risk (read 19 Nov 2026) - Palo Alto Networks, CVE-2026-0322 advisory (fixed builds, mitigation, indicator addresses): https://security.paloaltonetworks.com/CVE-2026-0322 (read 19 Nov 2026) - CISA, KEV JSON feed, catalog 2026.11.18, entry CVE-2026-0322: https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) - NVD record for CVE-2026-0322, including CISA SSVC entry: https://services.nvd.nist.gov/rest/json/cves/2.0?cveId=CVE-2026-0322 (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Decision Table: Remediate, Mitigate or Accept by Situation](https://remediationvsmitigation.com/decision-table/index.md) - [KEV Entry, Patch Out, Internet-Exposed: 3 or 14 Days](https://remediationvsmitigation.com/decision-table/kev-patch-available-exposed/index.md) - [WAF Rule as a Virtual Patch: Atlassian Case and Limits](https://remediationvsmitigation.com/compensating-controls/waf-rule/index.md) - [Network Segmentation as a Mitigation: Cisco ISE Case](https://remediationvsmitigation.com/compensating-controls/network-segmentation/index.md) - [When Virtual Patching Is Enough, and When It Is Not](https://remediationvsmitigation.com/guides/virtual-patching-and-when-it-is-enough/index.md) - [How to Verify a Mitigation Works: Checks for 5 Real CVEs](https://remediationvsmitigation.com/guides/verifying-a-mitigation-works/index.md) - [Residual Risk After Mitigation: Define, Estimate, Record](https://remediationvsmitigation.com/guides/residual-risk-after-mitigation/index.md) --- # The asset is scheduled to be retired on a known date Canonical: https://remediationvsmitigation.com/decision-table/decommission-scheduled/ Accept with a dated exception that expires on the decommission date, and add an interim control if the asset is reachable. BOD 26-04 counts decommissioning as remediation only once the asset is gone, so a federal KEV due date that falls first still binds. In the worked case below the gap is 8 days. Guidance points to: Accept with a dated exception. Also allowed: Mitigate now, fix later. Signals: Decommission date is set; Asset has no long-term role. ## The situation The asset has a retirement date in a project plan or change ticket, and a vulnerability is found on it before that date. The asset has no long-term role. The row does not cover software that is past support with no retirement date; that is [end-of-life software](https://remediationvsmitigation.com/decision-table/end-of-life-software/). ## The decision and its conditions Accept with a dated exception that ends on the decommission date. Mitigate now, fix later is also allowed and is the better choice if the asset is reachable from outside. The exception holds only while four conditions hold: 1. The date sits in an approved plan with a named owner, not in a hope. 2. An interim control is in place if the asset is reachable from the internet. For an unexposed asset, write down the basis for saying so. 3. The fix being skipped costs more than the exposure it leaves for the days that remain. 4. No shorter date binds you, such as a KEV due date on a federal system. ## The rules it follows Two sources agree that taking the asset away resolves the flaw. BOD 26-04: "the vulnerability shall be remediated (that is, the vulnerability is eliminated through patching, decommissioning the system, or another action)". NIST SP 800-40 Rev. 4 lists Avoid among its four responses: "Ensure that the risk does not occur by eliminating the attack surface, such as by uninstalling the vulnerable software, decommissioning assets with the vulnerabilities". Both mean the asset is gone. Until then the finding is open and the exception covers the gap. No rule sets a length for this exception, so the date is the length. NIST asks for tracking: "Organizations should closely track and monitor all exceptions to maintenance plans." and "All exceptions to maintenance plans should be reviewed regularly to determine if the maintenance plan can be implemented now." It also names the last step of avoidance: verify "that vulnerable assets were decommissioned or replaced" and check that no one "restarts the decommissioned asset". ## Worked case: CVE-2026-85880 on a file server A heap overflow in Windows ALPC lets code that already runs as a low-privileged local user inside an AppContainer escape the sandbox and reach SYSTEM (CVSS 3.1 base score 7.8, vector AV:L/AC:L/PR:L). Microsoft marked it exploited on the day it shipped the fix. The affected asset is an internal Server 2016 file server that the change ticket retires on 30 Sep 2026. CVE-2026-85880, Windows Advanced Local Procedure Call, on an internal Server 2016 file server. Read 19 Nov 2026. | Date | Event | Basis | |---|---|---| | 8 Sep 2026 | Microsoft publishes the fix (for Server 2016, KB5123099) and marks the flaw exploited. CISA adds the CVE to the KEV catalog the same day | Microsoft advisory; CISA KEV feed | | 22 Sep 2026 | KEV due date, 14 days after listing: BOD 26-04 Table 1 row 11 (not publicly exposed, in the KEV, not automatable, total control). The feed's own due date is the same | CISA KEV feed; Table 1, counted | | 30 Sep 2026 | Decommission date on the change ticket, 22 days after listing and 8 days after the due date | Case assumption | | 15 Nov 2026 | Date the project slips to, 68 days after listing | Case assumption | A federal agency has no exception to write. The due date of 22 Sep comes 8 days before the retirement, CISA issues no waivers, and its guidance gives one fallback: if the action cannot be done in time, "remove the asset from the agency network". The choices are KB5123099 and a restart before 22 Sep, or taking the server off the network by then. Any other organization may write the exception, 8 Sep to 30 Sep, if its policy allows it. Microsoft lists no workaround. The interim control is to limit who can sign in to the server, which lowers the chance that attacker code runs on the host at all; the cost is that users lose interactive access. Against that stands the cost of the fix, which is one cumulative update and one restart. For a file server that cost is often less than the exception's paperwork and its interim control, and the usual answer, accept until retirement, is the weaker choice. If the project slips to 15 Nov, the exception still ends on 30 Sep. Running the server to the new date is a new request, with the patch as its default. ## What to record **Exception: scheduled decommission** - Tracking ID and finding reference (CVE ID, scanner finding ID): - Asset, environment, exposure (internet-facing or internal): - Reason the asset is not patched: scheduled for retirement - Decommission date, and the project or change ticket that carries it: - Cost and downtime of the fix that is being skipped: - Interim control while the asset runs (or: asset not reachable, and the basis): - Expiry of this exception: the decommission date - Number of times the date has moved (allowed: 0 or 1): - Approver, title, date: - Person who will confirm removal, and the date: - Reopen if: the date moves a second time, the asset becomes internet-exposed, the CVE is added to the KEV catalog - Closed on: the date the asset was confirmed gone from the network and the inventory The last two lines carry the control: an expiry on the decommission date, and a named person who confirms removal. Fields not listed here are on the [risk acceptance form](https://remediationvsmitigation.com/risk-acceptance-form/) and the [exception register](https://remediationvsmitigation.com/exception-register/). What happens when a date passes is covered in [risk exception expiry](https://remediationvsmitigation.com/guides/exception-expiry-and-renewal/). ## When the usual answer is wrong - **A federal KEV due date falls before retirement.** BOD 26-04 allows no exception. Patch, or remove the asset from the network, before the due date. - **The asset is reachable from the internet.** Exposure moves a KEV entry to a shorter row: the same flaw on an exposed asset would be row 3 or 4, with 3 or 14 days. The directive says the timelines "are dynamic as facts change", and a retirement planned for next quarter does not pause them. Put the interim control in before the exception. For an unexposed asset that is not in the KEV, the slower row applies, such as the 60 days of the [internal row](https://remediationvsmitigation.com/decision-table/internal-asset-not-kev-moderate/); this case is a KEV entry, so row 11 and its 14 days apply. - **The date has already moved once.** A second move ends the exception and sends the finding back to its normal response. The one-move limit is a rule to write into your own policy; no standard sets it. - **The asset is powered off but still on the network or in the inventory.** Retirement counts only when the asset is confirmed gone. - **A listing on the KEV catalog, a change in exposure or a vendor mitigation the asset can take.** Each of these reopens the entry. ## Sources - CISA, BOD 26-04, Appendix A, notes to Table 1: https://www.cisa.gov/news-events/directives/bod-26-04-prioritizing-security-updates-based-risk (read 19 Nov 2026) - CISA, BOD 26-04 implementation guidance, alternative mitigation and waiver FAQs: https://www.cisa.gov/news-events/directives/bod-26-04-implementation-guidance-prioritizing-security-updates-based-risk (read 19 Nov 2026) - NIST SP 800-40 Rev. 4, Sections 2.1, 3.3 and 3.5.5: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf (read 19 Nov 2026) - CISA, KEV JSON feed, catalog 2026.11.18, entry CVE-2026-85880: https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) - Microsoft Security Response Center, September 2026 CVRF, CVE-2026-85880: https://api.msrc.microsoft.com/cvrf/v3.0/cvrf/2026-Sep (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Decision Table: Remediate, Mitigate or Accept by Situation](https://remediationvsmitigation.com/decision-table/index.md) - [Risk Acceptance Form: Free Template With 15 Fields](https://remediationvsmitigation.com/risk-acceptance-form/index.md) - [Exception Register: 15 Columns, States and Blank CSV](https://remediationvsmitigation.com/exception-register/index.md) - [Maximum Duration of a Risk Acceptance: 9 Sources](https://remediationvsmitigation.com/risk-acceptance-form/maximum-durations/index.md) - [Accepting Risk on End-of-Life Software: Limits and Record](https://remediationvsmitigation.com/guides/accepting-risk-on-end-of-life-software/index.md) - [Risk Exception Expiry: What Happens and How to Renew](https://remediationvsmitigation.com/guides/exception-expiry-and-renewal/index.md) - [End-of-Life Software: Replace, Isolate or Accept With Dates](https://remediationvsmitigation.com/decision-table/end-of-life-software/index.md) --- # End-of-life software: replace, isolate or accept Canonical: https://remediationvsmitigation.com/decision-table/end-of-life-software/ Remediate by replacing or removing the product, because no patch will ship. CISA BOD 26-02, issued 5 Feb 2026, has federal agencies decommission end-of-support edge devices by 5 Aug 2027. NIST SP 800-53 SA-22 allows an exception only for a critical capability with no replacement or for a system too isolated to replace. Guidance points to: Remediate. Also allowed: Mitigate now, fix later; Accept with a dated exception. Signals: Vendor has ended support; No updates exist. ## The situation The vendor has ended support, so no update exists or will. NIST puts the cause in one sentence: "The vendor may no longer support the vulnerable software, meaning that a patch for it will never be released because the software is at end-of-life." Every later finding on the product stays open for good, so the decision is about the product, not the single CVE. ## The decision and its conditions Remediate: replace the product with a supported one, or remove it. Mitigate now, fix later and Accept with a dated exception are allowed while the replacement is built. The exception is narrow, and the conditions are these: 1. A replacement project exists, with a date and an owner. 2. Until then the product is isolated from public and uncontrolled networks, and the isolation was checked. 3. If replacement is blocked, the block is one of the two SA-22 reasons below, and the approver signs it. 4. If the product is a federal edge device, the BOD 26-02 dates are met. ## The rules it follows BOD 26-02 covers edge devices, meaning devices on the network boundary that the vendor or CISA considers end of support. It binds federal civilian agencies, not contractors, and it "does not apply to Operational Technology devices and FedRAMP authorized cloud computing" services. Its dates run from 5 Feb 2026. BOD 26-02, Required Actions, issued 5 Feb 2026. The right column counts months from the issue date; CISA prints the intervals, not the calendar dates. Read 19 Nov 2026. | Deadline | Action required of federal civilian agencies | Date by count | |---|---|---| | Immediately | Update each vendor-supported edge device running end-of-support software to a supported version, where this does not adversely affect mission critical functionality | 5 Feb 2026 | | 3 months | Inventory devices on the CISA end-of-support edge device list | 5 May 2026 | | 12 months | Decommission listed devices whose end-of-support date is on or before the deadline | 5 Feb 2027 | | 18 months | "Decommission all identified EOS edge devices from agency networks, replacing devices as needed with vendor-supported devices that can receive current security updates." | 5 Aug 2027 | | 24 months | Keep continuous discovery running and have decommissioned devices on or before their end-of-support dates | 5 Feb 2028 | NIST SP 800-53 SA-22 states when replacement may wait: "Exceptions to replacing unsupported system components include systems that provide critical mission or business capabilities where newer technologies are not available or where the systems are so isolated that installing replacement components is not an option." It adds that the added risk "can be mitigated, for example, by prohibiting the connection of such components to public or uncontrolled networks, or implementing other forms of isolation". Its assessment objects include "documented approvals (including justification) for the continued use of unsupported system components", which is the record to keep. CISA's KEV catalog shows how often the end-state is removal, not a patch. CISA explains the practice in its guidance: it "does not assume that all running legacy products are fully patched", and "actors will continue exploiting vulnerabilities until they are remediated or mitigated". Required-action wording in the CISA KEV feed, catalog 2026.11.18 (1,786 entries), counted 19 Nov 2026. None of the 83 entries was added in 2026; 54 were added in 2022. | What the required action tells you to do | Entries | |---|---| | Disconnect the product if still in use | 62 | | Discontinue use of the product | 12 | | Remove the product from service or from agency networks | 5 | | Retire and replace the product (legacy D-Link hardware) | 4 | | Total with end-of-life wording | 83 | ## Worked case: CVE-2026-111497 on an end-of-life Node.js line A flaw in the Brotli decoder bundled in Node.js lets an attacker who controls compressed input corrupt process memory, which can lead to a crash, memory disclosure or remote code execution (CVSS 3.1 base score 9.8 from NVD). A service runs on Node.js 20, an end-of-life line. Node.js 22, 24 and 26 get fixes; Node.js 20 gets none. CVE-2026-111497, Brotli decoder bundled in Node.js. Read 19 Nov 2026. | Date | Event | Basis | |---|---|---| | 21 Oct 2026 | Node.js releases 22.23.4, 24.22.1 and 26.12.1. Its advisory says end-of-life lines (25.x, 23.x, 20.x and earlier) are always affected and get no fix | Node.js security release | | 26 Oct 2026 | ELLIO records exploitation in the wild at 15:08 UTC | ELLIO | | 27 Oct 2026 | CISA adds the CVE to the KEV catalog | CISA KEV feed | | 10 Nov 2026 | KEV due date, 14 days after listing. Forensic triage is marked not required | CISA KEV feed | | 19 Nov 2026 | Today: 9 days past the due date for a federal agency whose Node.js 20 service has not moved | Counted | There is no patch to wait for, so the remediation is a move to a supported line: 22.23.4 or later, then a rebuild of images and a restart. The stopgap the researchers at Aikido and ELLIO describe is to reject inbound requests with Content-Encoding br at the reverse proxy; the cost is an error for clients that send Brotli request bodies. It does not cover Brotli responses that the service itself fetches from other servers. BOD 26-02 does not set the clock here, because an application runtime is not an edge device; for a federal agency the clock is the KEV due date of 10 Nov 2026. For anyone else it is the date the policy sets. An SA-22 exception does not fit this service. A supported Node.js line exists, so "newer technologies are not available" is false, and the service is reachable by the attackers who send it requests, so "so isolated" is false too. The honest record is a dated plan to move, with the proxy rule as the interim control. ## What to record - Product, version, vendor end-of-support date, and the CVE or CVEs that raised the question. - The replacement: project, date, owner. For federal edge devices, the BOD 26-02 date that applies. - The isolation in place and how it was checked, or the basis for saying the product is not reachable. - If replacement is blocked: which of the two SA-22 reasons applies, who approved it and the expiry date. Use the [risk acceptance form](https://remediationvsmitigation.com/risk-acceptance-form/). - The review date. For federal edge devices the dates in the table are the ceiling; no source sets one for other assets, so review no later than the replacement date and earlier if exposure changes. ## When the usual answer is wrong - **Extended support exists.** A paid support contract is an alternative source of support under SA-22, and the product stops being end of life for that purpose. Record the contract and its end date. - **The product is in the KEV catalog.** If a patch exists for a supported version, go to [patch available and exposed](https://remediationvsmitigation.com/decision-table/kev-patch-available-exposed/); for a version past support the required action is usually to disconnect or discontinue, and the due date still applies to federal agencies. - **The product is an OT or medical device.** BOD 26-02 excludes OT. Use [OT and medical devices](https://remediationvsmitigation.com/decision-table/ot-medical-cannot-patch/), where replacement is often not an option. - **The stopgap is partial.** The Node.js proxy rule covers one path in. If the service also fetches Brotli data from outside, the finding is still open. How to write the exception when replacement is blocked, and what the feed shows by year, is on [accepting risk on end-of-life software](https://remediationvsmitigation.com/guides/accepting-risk-on-end-of-life-software/). ## Sources - CISA, BOD 26-02, Scope and Required Actions: https://www.cisa.gov/news-events/directives/bod-26-02-mitigating-risk-end-support-edge-devices (read 19 Nov 2026) - NIST SP 800-40 Rev. 4, Section 2.1: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf (read 19 Nov 2026) - NIST SP 800-53 Rev. 5 catalog (OSCAL), SA-22 guidance and assessment objects: https://github.com/usnistgov/oscal-content/blob/main/nist.gov/SP800-53/rev5/json/NIST_SP-800-53_rev5_catalog.json (read 19 Nov 2026) - CISA KEV JSON feed, requiredAction field and entry CVE-2026-111497, catalog 2026.11.18: https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) - CISA, BOD 26-04 implementation guidance, old CVEs FAQ: https://www.cisa.gov/news-events/directives/bod-26-04-implementation-guidance-prioritizing-security-updates-based-risk (read 19 Nov 2026) - Node.js, October 2026 security releases: https://nodejs.org/en/blog/vulnerability/october-2026-security-releases (read 19 Nov 2026) - ELLIO, node2shell analysis of CVE-2026-111497: https://ellio.tech/en/blog/node2shell-cve-2026-111497-rce-on-stock-nodejs-builds/ (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Decision Table: Remediate, Mitigate or Accept by Situation](https://remediationvsmitigation.com/decision-table/index.md) - [OT and Medical Devices You Cannot Patch: Mitigate, Review](https://remediationvsmitigation.com/decision-table/ot-medical-cannot-patch/index.md) - [KEV Entry, Patch Out, Internet-Exposed: 3 or 14 Days](https://remediationvsmitigation.com/decision-table/kev-patch-available-exposed/index.md) - [Risk Acceptance Form: Free Template With 15 Fields](https://remediationvsmitigation.com/risk-acceptance-form/index.md) - [Accepting Risk on End-of-Life Software: Limits and Record](https://remediationvsmitigation.com/guides/accepting-risk-on-end-of-life-software/index.md) --- # Internal asset, patch available, not on the KEV, automatable Canonical: https://remediationvsmitigation.com/decision-table/internal-asset-not-kev-moderate/ Remediate within 60 days. BOD 26-04 Table 1 rows 13 and 14 give 60 calendar days to an unexposed asset when the CVE is not in the KEV catalog but an adversary can automate it. The clock shortens to 3 days the day CISA lists it: Langflow went from 18 Sep to 7 Aug 2026. Guidance points to: Remediate. Also allowed: Accept with a dated exception. Signals: Publicly exposed: no; In the KEV: no; Automatable: yes. ## The situation The asset is not publicly exposed, the vendor has a fix, CISA has not listed the CVE in the Known Exploited Vulnerabilities catalog, and CISA's analysis says an adversary can automate every step. Under BOD 26-04 the inputs are asset exposure, catalog status, automatable by adversary and technical impact. CISA publishes the last three for every CVE through its Vulnrichment program; the agency answers exposure. ## The decision and why it is 60 days The decision is remediate, with a longer clock than for exposed assets or listed flaws. Table 1 has three rows that give 60 days, and only two apply here: BOD 26-04 Table 1, selected rows, read from the published table image on 19 Nov 2026. | Row | Exposed | In the KEV | Automatable | Impact | Timeline | This page? | |---|---|---|---|---|---|---| | 8 | Yes | No | No | Partial | 60 days | No, exposed | | 13 | No | No | Yes | Total | 60 days | Yes | | 14 | No | No | Yes | Partial | 60 days | Yes | | 15 and 16 | No | No | No | Total or partial | Fix on system upgrade | No, see [not reachable, low risk](https://remediationvsmitigation.com/decision-table/not-reachable-low-risk/) | | 9 | No | Yes | Yes | Total | 3 days & forensic triage | No, the CVE is listed | Automation is what separates 60 days from "fix on system upgrade". NIST SP 800-39 gives the reasoning for a longer clock on lower-priority risk: "This does not necessarily mean that the lower-priority risks would not be addressed. Rather, it could mean that fewer resources might be directed at the lower-priority risks (at least initially), or that the lower-priority risks would be addressed at a later time." Later is not never, and an organization may set a shorter period of its own. ## The rule, quoted - Remediation: "the vulnerability is eliminated through patching, decommissioning the system, or another action". - Exposure: the directive counts an asset as public if it is "accessible to unauthenticated or untrusted entities via public networks, such as the internet, regardless of its physical or logical location". An instance with a public IP address in a cloud account is exposed, however internal its role. - Clock start: the earlier of CISA adding the CVE to the catalog and the agency enumerating it on an asset in the CDM dashboard. The days are calendar days. - Outside the directive: PCI SSC FAQ 1597 says critical patches "must be resolved within one month of release" and other patches follow the entity's risk assessment. FedRAMP's 2026 rules (launched 24 Jun 2026) tell providers to mark any vulnerability not fixed within 192 days of evaluation as an accepted vulnerability. ## Worked case: Langflow, CVE-2026-9198 Langflow OSS 1.0.0 through 1.10.0 issues superuser tokens to any network caller when auto-login is on, and a code-validation endpoint then runs submitted Python, so an unauthenticated caller reaches remote code execution. Langflow 1.10.1 shipped on 23 Jun 2026, IBM's bulletin followed on 2 Jul, and NVD published the CVE on 17 Jul 2026, the day CISA's SSVC entry recorded automatable yes and impact total. An instance on a private network is row 13. CISA added the CVE to the catalog on 4 Aug 2026. Dates from IBM's bulletin, Langflow's release notes, NVD and the KEV feed, catalog 2026.11.18, read 19 Nov 2026. The scanner finding on 20 Jul is assumed for illustration. | Date | Event | Row and deadline | |---|---|---| | 23 Jun 2026 | Langflow 1.10.1 released with the fix | Not yet a CVE | | 17 Jul 2026 | NVD publishes the CVE; CISA's SSVC data lists automatable yes, impact total | Row 13: 60 days if the clock starts here | | 20 Jul 2026 | The agency's scanner finds an internal Langflow 1.9.x and updates the CDM dashboard (assumed date) | Clock starts. 60 days end 18 Sep 2026; midpoint 19 Aug | | 4 Aug 2026 | CISA adds the CVE to the catalog, forensic triage required. Day 15 of 60 | Row 9: 3 days & forensic triage. Catalog due date 7 Aug 2026 | A patch window planned for September fits the 60-day row and misses the new date by six weeks. CISA's guidance says to remediate "by the 'dueDate'" in the catalog, so 7 Aug 2026 is the date to meet. Counted from the 20 Jul start, the 3-day timeline would already have ended on 23 Jul; the directive does not say how a row change in mid-clock is counted, so take the catalog date and raise the earlier-start reading with CISA if it matters. The row changed because of CISA's action alone; nothing about the asset changed. Watching the catalog on the day it updates is part of owning a 60-day row. Contrast a CVE that stays on the unlisted side: fastjson 1.2.68 to 1.2.83 (CVE-2026-16723) has CISA's SSVC entry of automatable no and impact total, and is not in the catalog. An unexposed deployment is row 15, fix on system upgrade, and the 1.2.84 release of 29 Jul 2026 can wait for the next rebuild. Automatable yes versus no is the whole difference between that and Langflow. ## When an exception is allowed For federal systems under BOD 26-04 the exception route is closed: CISA grants no waivers or exceptions to directive actions, and nothing in the text carves out the 60-day rows. The fallback for a missed date is to remove the asset from the network. Others may accept the risk. NIST SP 800-39 says "organizations may accept substantially greater risk (in the moderate/high range) due to compelling mission, business, or operational needs". The condition is the compelling need; a wish to avoid a reboot is not one. The exception names the CVE, a signer with the authority to accept it, the compensating step and an expiry date; see the [risk acceptance form](https://remediationvsmitigation.com/risk-acceptance-form/). ## What to record - CVE ID, asset, and the evidence for each Table 1 input: exposure (how it was established), catalog status on the day, automatable and impact as CISA published them. - The clock start and the source of that date (dashboard entry or catalog addition). - The planned patch date inside the 60 days, the test that supports it, and who approves the change window. - Review dates: the midpoint, day 60, and a catalog check at each review. If you set a shorter period than the table, write down that it is your own. - The patched version and the date it was reached. ## When the usual answer is wrong - The CVE joins the catalog. The row becomes 9 to 12 and the date comes from the catalog (Langflow above). See [patch available and exposed](https://remediationvsmitigation.com/decision-table/kev-patch-available-exposed/) for the 3-day rows. - The asset is not as internal as recorded. A load balancer, a vendor tunnel or a public address makes it exposed; if any one method says so, the value is Yes and rows 5 to 8 apply. - Automation changes. If CISA revises the entry to automatable no, the row moves to 15 or 16. - Another path in. An internal attacker is anyone on the network, including a compromised workstation. Langflow's tokens went to any network caller, which shows why internal is not a safe label for an unauthenticated service. - The patch cannot be applied. Go to [patch breaks a critical system](https://remediationvsmitigation.com/decision-table/patch-breaks-critical-system/). ## Sources - CISA, BOD 26-04, definitions, Table 1 and Appendix A notes: https://www.cisa.gov/news-events/directives/bod-26-04-prioritizing-security-updates-based-risk (read 19 Nov 2026) - CISA, BOD 26-04 implementation guidance, FAQ on waivers and due dates: https://www.cisa.gov/news-events/directives/bod-26-04-implementation-guidance-prioritizing-security-updates-based-risk (read 19 Nov 2026) - CISA KEV JSON feed, catalog 2026.11.18, CVE-2026-9198 dateAdded and dueDate: https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) - IBM security bulletin for Langflow OSS, CVE-2026-9198 (vendor advisory): https://www.ibm.com/support/pages/node/7278927 (read 19 Nov 2026) - Langflow, release notes for 1.10.1 (vendor): https://github.com/langflow-ai/langflow/releases/tag/v1.10.1 (read 19 Nov 2026) - NIST SP 800-39, risk response and risk acceptance: https://nvlpubs.nist.gov/nistpubs/Legacy/SP/nistspecialpublication800-39.pdf (read 19 Nov 2026) - PCI SSC, FAQ 1597: https://www.pcisecuritystandards.org/faqs/1597 (read 19 Nov 2026) - FedRAMP 2026 rules, VER-TFR-MAV: https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-evaluation-and-reporting/ (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Decision Table: Remediate, Mitigate or Accept by Situation](https://remediationvsmitigation.com/decision-table/index.md) - [Not Exposed, Not on the KEV, Not Automatable: Accept](https://remediationvsmitigation.com/decision-table/not-reachable-low-risk/index.md) - [KEV Entry, Patch Out, Internet-Exposed: 3 or 14 Days](https://remediationvsmitigation.com/decision-table/kev-patch-available-exposed/index.md) - [Maximum Duration of a Risk Acceptance: 9 Sources](https://remediationvsmitigation.com/risk-acceptance-form/maximum-durations/index.md) --- # Exploited flaw, no patch released yet Canonical: https://remediationvsmitigation.com/decision-table/kev-no-patch-yet/ Mitigate now and plan the fix. NIST SP 800-40 Rev. 4 says to set and enforce two schedules, one for deploying the patch and one for removing the mitigation. FortiMail 7.2 shows the stakes: listed as exploited on 1 Oct 2026, its fix was still marked upcoming on 19 Nov 2026, 49 days later. Guidance points to: Mitigate now, fix later. Also allowed: Accept with a dated exception. Signals: A patch may not be available yet; Mitigation instructions exist from the vendor. ## The situation The flaw is exploited or severe, the vendor has published a fix for some versions or for none, and your version has no fix yet. NIST SP 800-40 Rev. 4, Section 2.1, lists this among the reasons patching is not immediately viable: "A patch may not be available yet. For example, a vulnerability may be announced before a patch is ready, and it could be days, weeks, or months before the patch is released." This page is for a fix that is coming. A product that will never be fixed belongs on [end-of-life software](https://remediationvsmitigation.com/decision-table/end-of-life-software/), and the general options when no fix is planned are in mitigation when no patch exists. ## The decision and its conditions The decision is mitigate now, fix later. It holds when all of these are true: - The vendor, or a standard such as CISA's required action, names a mitigation you can apply: a feature turned off, access narrowed, or a rule in front of the service. - You can say what the mitigation breaks and who accepts that. - You can set a date by which the permanent fix replaces the mitigation, even if that date is a vendor check-in date and not a release date. If no mitigation exists and the product cannot be stopped, the response is an exception recorded on the [risk acceptance form](https://remediationvsmitigation.com/risk-acceptance-form/), with the duration limits on the [maximum durations page](https://remediationvsmitigation.com/risk-acceptance-form/maximum-durations/). ## The rule, quoted NIST SP 800-40 Rev. 4 defines the practice: "Emergency mitigation. This is the emergency procedure in a crisis situation, like those described above for the emergency patching scenario, to temporarily mitigate vulnerabilities before a patch is available." Section 3.5.3 adds the obligation: "Once a permanent fix, such as a patch, is available, the patch will need to be deployed and the mitigation removed. Schedules should be set and enforced for both patch deployment and mitigation removal." It also says: "Organizations should plan to replace emergency mitigations with permanent fixes." For a federal civilian agency, the catalog entry's required action tells you what is acceptable. The text on entries added since the directive reads, in part: "Apply mitigations in accordance with vendor instructions, ensuring compliance with CISA's BOD 26-04 ... or discontinue use of the product if mitigations are unavailable." CISA's implementation guidance adds that "the only known technical mitigation to vulnerabilities in the KEV Catalog is to apply the required actions listed in the catalog", so a vendor mitigation meets the directive and a homemade one does not. CISA issues no waivers. For a product embedded in another vendor's offering with no patch, the guidance says to "contact that vendor directly to obtain the patch or other operational guidance". Legacy FedRAMP guidance (reference only after 24 Jun 2026) treated this as a vendor dependency: "CSPs are required to check in with the vendor at least once a month to determine the status of the patch/fix", with a date of last check-in recorded. Copying that cadence is a choice, not a current rule. ## Worked case: FortiMail, CVE-2026-104286 A path traversal in FortiMail lets an unauthenticated attacker write files on the appliance through crafted HTTP or HTTPS requests. Fortinet rates it 9.8 and, in advisory FG-IR-26-175, says it "has been reported to be exploited in the wild". CISA's SSVC entry gives automatable yes and technical impact total, so an exposed appliance is a Table 1 row 1 case. Dates from Fortinet PSIRT FG-IR-26-175 and the KEV feed, catalog 2026.11.18, read 19 Nov 2026. | Date | Event | What a defender with a FortiMail 7.2.9 appliance does | |---|---|---| | 1 Oct 2026 | Fortinet publishes the advisory with a workaround; CISA adds the CVE to the catalog the same day, due 4 Oct 2026 | Apply the workaround inside the 3 days. Start the triage record; the catalog entry marks forensic triage as required | | 5 Oct 2026 | Advisory solution update; Fortinet fixes FortiMail Cloud itself (no customer action) | Cloud tenants close the record. Self-managed appliances stay open | | 7 Oct 2026 | Last advisory revision: 8.0.2, 7.6.7 and 7.4.9 are fixed builds; the 7.2 train reads "Upgrade to upcoming 7.2.10 or above"; 7.0 reads "Migrate to a fixed release" | 8.0, 7.6 and 7.4 owners start the patch clock. The 7.2 owner keeps the mitigation and sets a check date | | 19 Nov 2026 | The advisory page still reads "upcoming 7.2.10", updated date 7 Oct | The 7.2 appliance has run on the workaround for 49 days with the removal date unset | The vendor's three mitigations differ in what they cost. Turning the IBE (identity-based encryption) service off stops secure-mail delivery through the portal, so recipients can no longer open or reply to IBE messages. Blocking the webmail interface from the internet removes access for external users, including IBE recipients. A web application firewall rule on POST requests to /ibe containing a parent-directory sequence must be written and tested by the owner. The choice of mitigation is a business decision and belongs in the record. The 7.0 line has no fix to wait for: Fortinet says to migrate. That appliance has left this page and moved to [end-of-life software](https://remediationvsmitigation.com/decision-table/end-of-life-software/). ## What to record Mitigation record, exploited flaw with no patch for this version CVE ID: CVE-2026-104286 Asset and version: FortiMail 7.2.9, mail gateway Table 1 inputs: exposed yes, in the KEV yes, automatable yes, impact total Clock start: 1 Oct 2026 (catalog addition) or the earlier identification date Vendor source and date read: Fortinet FG-IR-26-175, revision of 7 Oct 2026, read 19 Nov 2026 Mitigation applied, date and by whom: IBE service off, 1 Oct 2026 What it breaks and who accepted that: IBE secure-mail recipients lose the portal; accepted by the mail service owner Evidence the mitigation is in effect: configuration export showing IBE service off Forensic triage: report reference and date; logs and configuration backup preserved before the change Fix status check: next vendor check date, last check date, evidence of the check Patch deployment date, set when the fix ships: ____ Mitigation removal date, set when the fix ships: ____ Close condition: patch applied and mitigation removed, both dated Close the record only when both dates are filled. A mitigation with no removal date reads later as an unrecorded exception. No source sets a number of days for how long an emergency mitigation may stand, so the review rule is the earlier of the vendor's release and a date you fix when you apply it, with any extension tied to a signed exception. ## When the usual answer is wrong - The mitigation closes one route, not the flaw. Fortinet's workaround stops the IBE path, and it does not undo a compromise from before 1 Oct. The workaround and the triage are separate tasks. - The mitigation costs more than the exposure. If the workaround stops a service the business cannot lose, the choice becomes an exception with a signer who can accept that, not a quiet pass. - The fix arrives for some trains only. Fortinet shipped three trains and left a fourth waiting; track each version line, not the product. - The vendor's mitigation is a configuration change in a system you do not run. For SaaS, see [SaaS you cannot patch](https://remediationvsmitigation.com/decision-table/saas-cannot-patch/). ## Sources - NIST SP 800-40 Rev. 4, Sections 2.1, 3.3 and 3.5.3: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf (read 19 Nov 2026) - Fortinet PSIRT, FG-IR-26-175 (vendor advisory): https://fortiguard.fortinet.com/psirt/FG-IR-26-175 (read 19 Nov 2026) - CISA KEV JSON feed, catalog 2026.11.18, CVE-2026-104286 entry and requiredAction field: https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) - CISA, BOD 26-04 implementation guidance, FAQ on alternative mitigations, vendors and waivers: https://www.cisa.gov/news-events/directives/bod-26-04-implementation-guidance-prioritizing-security-updates-based-risk (read 19 Nov 2026) - FedRAMP legacy POA&M playbook, vendor dependency (reference only after 24 Jun 2026): https://www.fedramp.gov/legacy/playbook/csp/authorization/poam/ (read 19 Nov 2026) - NVD record for CVE-2026-104286, including CISA SSVC entry: https://services.nvd.nist.gov/rest/json/cves/2.0?cveId=CVE-2026-104286 (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Decision Table: Remediate, Mitigate or Accept by Situation](https://remediationvsmitigation.com/decision-table/index.md) - [Maximum Duration of a Risk Acceptance: 9 Sources](https://remediationvsmitigation.com/risk-acceptance-form/maximum-durations/index.md) - [KEV Required Action: Mitigate or Discontinue Use](https://remediationvsmitigation.com/guides/kev-required-actions-mitigate-or-discontinue/index.md) - [End-of-Life Software: Replace, Isolate or Accept With Dates](https://remediationvsmitigation.com/decision-table/end-of-life-software/index.md) --- # Patch available, on the KEV catalog, internet-exposed Canonical: https://remediationvsmitigation.com/decision-table/kev-patch-available-exposed/ Remediate. CISA BOD 26-04 (10 Jun 2026) gives a federal civilian agency 3 days, or 3 days plus forensic triage, for an exposed asset with a KEV entry and a patch; one of the four rows allows 14. The clock starts at the earlier of the catalog addition and the agency's own finding. Guidance points to: Remediate. Also allowed: Mitigate now, fix later. Signals: Publicly exposed: yes; In the KEV: yes; A vendor update exists. ## The situation Three facts are true at once: a vendor update exists, CISA lists the CVE in the Known Exploited Vulnerabilities (KEV) catalog, and the asset is publicly exposed. The directive defines the last term: "Any agency-owned or agency-managed IT resource accessible to unauthenticated or untrusted entities via public networks, such as the internet, regardless of its physical or logical location." A server in a private subnet that a load balancer, a NAT rule or a vendor tunnel makes reachable from the internet counts. Where several tools disagree, the directive settles it: "If one such method finds that a system is publicly exposed, the value for the decision point should be set to Yes." ## The decision and what sets the number of days The decision is remediate: the directive defines it as "the vulnerability is eliminated through patching, decommissioning the system, or another action". Table 1 holds four rows for an exposed asset on the catalog. Which one applies depends on two facts CISA publishes for each CVE through its Vulnrichment program: whether an adversary can automate every step, and whether exploitation gives total or partial control. The examples below are CVEs on this site whose catalog dates were read on 19 Nov 2026. BOD 26-04 Table 1, rows 1 to 4 (publicly exposed: yes, in the KEV: yes), read from the published table image on 19 Nov 2026. Automatable and impact values for the examples are from CISA's SSVC entries in NVD; due dates are from the KEV feed, catalog 2026.11.18. | Table 1 row | Automatable | Technical impact | Timeline | Example from the catalog | Added to catalog | Catalog due date | |---|---|---|---|---|---|---| | 1 | Yes | Total | 3 days & forensic triage | CVE-2026-97229, GitLab CE and EE | 29 Oct 2026 | 1 Nov 2026 | | 2 | Yes | Partial | 3 days | CVE-2026-76418, Cisco ASA and FTD | 29 Oct 2026 | 1 Nov 2026 | | 3 | No | Total | 3 days & forensic triage | CVE-2026-87491, Chromium V8 | 9 Sep 2026 | 23 Sep 2026 | | 4 | No | Partial | 14 days | CVE-2026-41709, VMware ESX | 28 Oct 2026 | 11 Nov 2026 | The V8 line shows how to read a due date. Row 3 would give 3 days, yet the catalog shows 14 (9 Sep to 23 Sep). CISA's guidance explains that it calculates the catalog date "using available data to determine whether there is evidence that the vulnerability is present in a publicly exposed asset", and that "the final determination for asset exposure is determined by the agency". CISA's 14 days matches row 11, an unexposed asset. If your V8 build sits on an exposed asset, row 3 applies to that asset, not the date in the feed. ## The rule, in the directive's words - Clock start: "The timelines defined in Table 1 begin when either (1) CISA adds the vulnerability to the KEV Catalog, or (2) pursuant to BOD 23-01, the agency enumerates or identifies the vulnerability on an asset and updates the Continuous Diagnostics and Mitigation (CDM) Program Agency and Federal Dashboard. Whichever event occurs first starts the remediation timeline." - Triage: "The text '& forensic triage' means that the agency must complete remediation or mitigation action within the timeline (three days) and carry out a forensic triage of the asset to assess whether the system is compromised." - Scope: the directive applies to federal civilian agency assets and, "unless directed by the governing procurement contract", not to contractors. CISA issues no waivers: "CISA does not issue waivers or exceptions for actions required in cyber directives." - Cloud providers under FedRAMP's 2026 rules (launched 24 Jun 2026) are told that "Providers SHOULD remediate Known Exploited Vulnerabilities according to the due dates in the CISA Known Exploited Vulnerabilities Catalog (even if the vulnerability has been fully mitigated)". Everyone else has no CISA deadline. The 3-day and 14-day figures are still a published benchmark for how fast a cataloged flaw on an exposed asset is expected to move, and an auditor or insurer may cite them. ## Worked case: GitLab, CVE-2026-97229 GitLab shipped the fix in its critical patch release of 12 Oct 2026 (19.4.2, 19.3.5 and 19.2.8) and rates the flaw 10.0. CISA's SSVC entry, stamped 13 Oct, lists automatable yes and technical impact total, so an exposed self-managed instance falls in row 1. CISA added the CVE to the catalog on 29 Oct 2026, 17 days after the patch, with a due date of 1 Nov 2026 (29 Oct plus 3 days). Take two agencies with the same exposed instance on 19.4.1. Two assumed agencies applying BOD 26-04's clock rule to the real GitLab dates. The enumeration dates are invented for illustration; everything else is from GitLab, NVD and the KEV feed. | | Agency A | Agency B | |---|---|---| | Enumerates the instance and updates the CDM dashboard | 20 Oct 2026 | 30 Oct 2026 | | Table 1 row on that day | Row 5: exposed, not yet in the KEV, automatable, total: 3 days | Row 1: 3 days & forensic triage | | Clock started | 20 Oct 2026, the earlier event | 29 Oct 2026, the catalog addition | | Date the 3 days end | 23 Oct 2026 | 1 Nov 2026 | | Position when CISA adds the CVE on 29 Oct | Already 6 days past the date, and triage is now also required | Within the catalog due date, 2 days left | Agency B must patch and also triage by 1 Nov 2026. GitLab states that 19.4.1 had an incomplete earlier fix, so the target is 19.4.2 or later. The patch itself is a restart of GitLab services after a backup. GitLab also warns that a successful exploit can be quiet, so a clean log does not prove safety, and its detections name the Workhorse log, the CI lint endpoint and the GraphQL ciLint mutation. CISA's implementation guidance puts the triage steps in order: scope, preserve volatile data, patch, contain, analyze, then a report. It gives target hours (scoping within two hours of the catalog addition, the report within 48 to 72 hours) as recommendations only; the requirement is that "an adequate forensic triage analysis is performed". Preserve evidence before patching: "Collect all required evidence prior to this step as patching may jeopardize the availability of artifacts." ## What to record - CVE ID, asset, and the date and tool that first identified the CVE on it, so the clock start can be shown (the earlier of that date and the catalog addition). - Table 1 inputs as answered for this asset: exposure with its evidence, catalog status, automatable and impact as CISA published them, and the row and number of days they give. - The date the update was applied or the system was decommissioned, and the version reached. - In a triage row: the forensic triage report and its date. CISA's guidance says it "will serve as an agency record" and should hold the incident timeline, the actions taken, technical findings, containment and a recommendation. - If you mitigated first by taking the system off the internet: the date and the evidence. That changes the exposure value. There is no review date here, only a due date. A mitigation is a step inside the clock, not a replacement for it. ## When the usual answer is wrong - The catalog date is later than 3 days but your asset is exposed. The feed reflects CISA's evidence, not your inventory; the agency decides exposure asset by asset (Chromium V8 above). - The patch exists but the first fix was incomplete. GitLab 19.4.1 is the case on this page: check the vendor's fixed-version list, not just that you applied an update. - The patch is applied but the asset may already be compromised. In a triage row the patch alone does not meet the directive. Rows 2 and 4 carry no triage requirement, but exploitation before the patch is still possible for any cataloged flaw. - The patch breaks something the mission needs. The deadline stays; go to [a patch that breaks a critical system](https://remediationvsmitigation.com/decision-table/patch-breaks-critical-system/). - The vendor has not shipped a patch for your version. Go to [no patch yet](https://remediationvsmitigation.com/decision-table/kev-no-patch-yet/). - The asset is not exposed after all. The timeline lengthens: row 9 to 12 for an unexposed asset on the catalog, or [internal asset, not on the KEV](https://remediationvsmitigation.com/decision-table/internal-asset-not-kev-moderate/) if the CVE is not listed. The newest catalog entries carry a required action that names BOD 26-04 and ends with an exposure duty: "Stakeholders are responsible for evaluating each asset's internet exposure and ensuring adherence to BOD 26-04 patching guidelines." In the feed read on 19 Nov 2026 (catalog 2026.11.18), 169 of 1,786 entries carry it: 131 with a 3-day due date and 38 with 14 days. The wording of those actions, including the discontinue clause, is covered in [KEV required actions: mitigate or discontinue](https://remediationvsmitigation.com/guides/kev-required-actions-mitigate-or-discontinue/). ## Sources - CISA, BOD 26-04, definitions, Table 1 and Appendix A notes: https://www.cisa.gov/news-events/directives/bod-26-04-prioritizing-security-updates-based-risk (read 19 Nov 2026) - CISA, BOD 26-04 implementation guidance, forensic triage steps and FAQ on due dates and waivers: https://www.cisa.gov/news-events/directives/bod-26-04-implementation-guidance-prioritizing-security-updates-based-risk (read 19 Nov 2026) - CISA KEV JSON feed, catalog 2026.11.18, dateAdded, dueDate and requiredAction: https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) - FedRAMP 2026 rules, VDR-TFR-KEV: https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-detection-and-response/ (read 19 Nov 2026) - GitLab, critical patch release 19.4.2, 19.3.5, 19.2.8 (vendor advisory): https://docs.gitlab.com/releases/patches/patch-release-gitlab-19-4-2-released/ (read 19 Nov 2026) - NVD record for CVE-2026-97229, including CISA SSVC entry: https://services.nvd.nist.gov/rest/json/cves/2.0?cveId=CVE-2026-97229 (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Decision Table: Remediate, Mitigate or Accept by Situation](https://remediationvsmitigation.com/decision-table/index.md) - [Exploited Flaw, No Patch Yet: Mitigate and Set Two Dates](https://remediationvsmitigation.com/decision-table/kev-no-patch-yet/index.md) - [Internal Asset, Not on the KEV, Automatable: 60 Days](https://remediationvsmitigation.com/decision-table/internal-asset-not-kev-moderate/index.md) - [KEV Required Action: Mitigate or Discontinue Use](https://remediationvsmitigation.com/guides/kev-required-actions-mitigate-or-discontinue/index.md) --- # Not internet-reachable, not on the KEV, not automatable Canonical: https://remediationvsmitigation.com/decision-table/not-reachable-low-risk/ Accept with a dated exception. CISA BOD 26-04 Table 1 gives rows 15 and 16 the timeline "Fix on system upgrade": asset not publicly exposed, CVE not in the KEV, exploitation not automatable. Guidance says agencies need not prioritize them, but a flip in any answer moves the row to 14 or 60 days. Guidance points to: Accept with a dated exception. Also allowed: Mitigate now, fix later. Signals: Publicly exposed: no; In the KEV: no; Automatable: no. ## The situation and the decision Three answers are all no: the asset is not publicly exposed, the CVE has no entry in CISA's Known Exploited Vulnerabilities (KEV) catalog, and an adversary cannot automate every step of exploitation. The response guidance points to is to accept the finding with a signed, dated exception and let the fix ride the next scheduled major upgrade or rebuild. Mitigating now and fixing later is also allowed. The decision holds only while the three answers hold. Table 1 of [BOD 26-04](https://www.cisa.gov/news-events/directives/bod-26-04-prioritizing-security-updates-based-risk) shows what one flipped answer does to each of the two "Fix on system upgrade" rows (technical impact is total control in row 15, partial control in row 16): BOD 26-04 Table 1, read from CISA's published table image on 19 Nov 2026. Timelines are calendar days. | Row | Baseline | If it becomes publicly exposed | If it is added to the KEV | If exploitation becomes automatable | |---|---|---|---|---| | 15 (total control) | Fix on system upgrade | Row 7: 14 days | Row 11: 14 days | Row 13: 60 days | | 16 (partial control) | Fix on system upgrade | Row 8: 60 days | Row 12: 14 days | Row 14: 60 days | ## How each answer is decided **Publicly exposed.** The directive sets the value to yes if the asset is "accessible to unauthenticated or untrusted entities via public networks, such as the internet, regardless of its physical or logical location". If any one of several overlapping methods finds it exposed, the value is yes, so one scanner saying no does not outweigh another saying yes. Where an agency has not supplied exposure data, [CISA's implementation guidance](https://www.cisa.gov/news-events/directives/bod-26-04-implementation-guidance-prioritizing-security-updates-based-risk) treats an asset as public if it uses a publicly routable IP address or has the asset type "network", and as internal only after those tests fail. **In the KEV.** The CVE ID has an entry in the catalog. Catalog version 2026.11.18, read on 19 Nov 2026, holds 1,786 entries. **Automatable and technical impact.** Both come from CISA's Vulnrichment data for the CVE. The guidance adds a default: when that data is missing and the CVE is not in the KEV, CISA treats automatable as "no" and technical impact as "total" until CISA or a third party supplies it. An unenriched CVE on an unexposed asset therefore lands on row 15. Internal is not the same as unreachable. FedRAMP's 2026 evaluation rules focus on "internet-reachable (rather than internet-accessible)" flaws because a service behind a load balancer can still receive a payload that the vulnerable code processes, and they give SQL injection as a classic example on systems "not typically internet-accessible". Once a filter that deflects the triggering payload is in place, the same text says the flaw "should no longer be considered an internet-reachable vulnerability". ## The rule it follows BOD 26-04, Appendix A: "Fix on system upgrade means that, unless conditions change as described in item (e) above, the vulnerability should be remediated the next time the vulnerable asset receives a scheduled major upgrade or rebuild." Item (e) is the reset clause: removing a system from the internet moves it to a slower row, but a CVE added to the KEV "will shorten according to Table 1". The implementation guidance puts it plainly: "agencies do not need to prioritize these vulnerabilities and can wait for the next system upgrade". It also says the timelines "represent maximum duration until a remediation is applied" and that an agency with lower tolerance "may set shorter timelines". For anyone outside the federal rule, the test is NIST SP 800-39: "Risk acceptance is the appropriate risk response when the identified risk is within the organizational risk tolerance." The guide on [risk acceptance versus risk tolerance](https://remediationvsmitigation.com/guides/risk-acceptance-vs-risk-tolerance/) covers who sets that limit. ## A worked case: CVE-2026-16723 in fastjson 1.x CVE-2026-16723 is a remote code execution flaw (CWE-502, CVSS 3.1 base score 9.0 from the CNA) in fastjson 1.2.68 through 1.2.83, a Java JSON library. Alibaba's advisory is dated 21 Jul 2026. NVD published the record on 23 Jul 2026, and the same day CISA's coordinator entry on it read exploitation none, automatable no, technical impact total. Alibaba released fastjson 1.2.84 with the fix on 29 Jul 2026 and archived the 1.x repository the same day. The CVE is not in KEV catalog 2026.11.18. Take a batch service that parses JSON from an internal queue, runs on a private address, and is not tagged as a network device. The advisory's trigger is a Spring Boot executable fat-jar that passes attacker-supplied JSON to the parser. Suppose the agency's tooling first lists the CVE on that service on 29 Jul 2026 (an assumed date for the example). The answers are exposure no, KEV no, automatable no, technical impact total: row 15, fix on system upgrade. Dates from NVD, the CISA entry on the CVE record, Alibaba's release notes and the KEV feed, read 19 Nov 2026. The last two rows are computed from the assumed 29 Jul start. | Date | Event | Effect on the row | |---|---|---| | 21 Jul 2026 | Alibaba publishes the advisory | None yet | | 23 Jul 2026 | NVD record; CISA entry: exploitation none, automatable no, technical impact total | Row 15 for an unexposed asset | | 29 Jul 2026 | fastjson 1.2.84 released; 1.x repository archived | Fix exists; no further 1.x fixes will come | | 19 Nov 2026 | KEV catalog 2026.11.18 checked: no entry | Row 15 stands | | If exposed on 29 Jul 2026 | Row 7 applies | Remediate by 12 Aug 2026 | | If automatable on 29 Jul 2026 | Row 13 applies | Remediate by 27 Sep 2026 | The exception names the next scheduled rebuild of the service as the fix date. Because fastjson 1.x is archived, that rebuild has to be a move to 1.2.84 or to fastjson2 (2.0.63 or later, which the advisory says is not affected by this CVE), so the plan names which one. ## What to record - The three answers, the method behind each, and the date of each check (the exposure check, the KEV feed version, the date the Vulnrichment values were read). - The upgrade or rebuild the fix waits for, with a calendar date even if approximate, and the version it moves to. - The triggers that reopen the entry: an exposure change, a KEV listing, a change in the Vulnrichment values, or a new proof of exploitation. - The approver. On a federal system under BOD 26-04 the timeline is the rule and nobody signs a longer one; the authorizing official (the role NIST SP 800-37 gives the explicit acceptance of risk) signs only an agency's own written exception. Elsewhere, the risk owner whose stated tolerance covers the finding signs, with a title and a date. - A review date. BOD 26-04 sets none for an agency; it asks CISA to reassess its own timelines once per fiscal year, a cadence that is easy to copy. This site's rule: use the sooner of the planned upgrade and a yearly exposure re-check. The [risk acceptance form](https://remediationvsmitigation.com/risk-acceptance-form/) holds the fields and the [exception register](https://remediationvsmitigation.com/exception-register/) tracks the entry. ## When accepting is the wrong answer - The fix is cheap. For CVE-2026-16723 the fix is a dependency bump to 1.2.84 on a service already scheduled for a release, so a dated exception costs more paperwork than the change. BOD 26-04 permits waiting; it does not ask for it. - The service is reachable through something in front of it. If the batch queue is fed by an internet-facing form, the parser still handles attacker-supplied data and the exposure answer should be yes under the FedRAMP reading above. - One of the three answers rests on a single source. An asset tagged internal from a spreadsheet but seen on a public address by a scanner is exposed. - **Exploitation is automatable and a patch exists.** Automation, not exposure, is the test here. Rows 13 and 14 (not exposed, not in the KEV, automatable) give 60 days, so the [60-day internal row](https://remediationvsmitigation.com/decision-table/internal-asset-not-kev-moderate/) applies and the exception route is closed. - The finding may be absent rather than unreachable; see [false positive handling](https://remediationvsmitigation.com/guides/false-positive-handling/). ## Sources - CISA, BOD 26-04, Table 1 and Appendix A definitions: https://www.cisa.gov/news-events/directives/bod-26-04-prioritizing-security-updates-based-risk (read 19 Nov 2026) - CISA, BOD 26-04 implementation guidance: exposure fallback logic, defaults, fix on system upgrade: https://www.cisa.gov/news-events/directives/bod-26-04-implementation-guidance-prioritizing-security-updates-based-risk (read 19 Nov 2026) - FedRAMP, Vulnerability Evaluation and Reporting rules 2026, VER-EVA-EIR: https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-evaluation-and-reporting/ (read 19 Nov 2026) - NIST SP 800-39, Risk Acceptance: https://nvlpubs.nist.gov/nistpubs/Legacy/SP/nistspecialpublication800-39.pdf (read 19 Nov 2026) - NVD record for CVE-2026-16723, with the CISA coordinator entry: https://services.nvd.nist.gov/rest/json/cves/2.0?cveId=CVE-2026-16723 (read 19 Nov 2026) - Alibaba, fastjson 1.2.84 release notes: https://github.com/alibaba/fastjson/releases/tag/1.2.84 (read 19 Nov 2026) - CISA KEV feed, catalog version 2026.11.18: https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) - NIST SP 800-37 Rev. 2, Task R-4: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-37r2.pdf (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Decision Table: Remediate, Mitigate or Accept by Situation](https://remediationvsmitigation.com/decision-table/index.md) - [Internal Asset, Not on the KEV, Automatable: 60 Days](https://remediationvsmitigation.com/decision-table/internal-asset-not-kev-moderate/index.md) - [Risk Acceptance Criteria: 7 Tests From NIST and CISA](https://remediationvsmitigation.com/guides/risk-acceptance-criteria/index.md) - [Exception Register: 15 Columns, States and Blank CSV](https://remediationvsmitigation.com/exception-register/index.md) - [Risk Acceptance vs Risk Tolerance: Who Sets the Limit](https://remediationvsmitigation.com/guides/risk-acceptance-vs-risk-tolerance/index.md) - [Vulnerability False Positive: Close It With Evidence](https://remediationvsmitigation.com/guides/false-positive-handling/index.md) --- # Medical and OT devices that cannot follow the patch cycle Canonical: https://remediationvsmitigation.com/decision-table/ot-medical-cannot-patch/ Mitigate with a pre-approved set of long-term methods and re-evaluate them on a schedule; accept the residual risk in a dated exception. Siemens advisory SSA-724261, published 10 Nov 2026 for CVE-2026-109830, says "Currently no fix is available", and CISA's republication on 17 Nov 2026 lists exposure and impact-analysis steps instead. Guidance points to: Mitigate now, fix later. Also allowed: Accept with a dated exception. Signals: Vendor-controlled update schedule; Safety or availability impact. ## The situation and the decision The device is a controller, gateway, infusion pump, imaging system or similar asset whose maker sets the update schedule, because each release needs testing or certification first, or because the maker has issued no fix. The decision is to isolate and restrict the device with methods reviewed in advance, to record the residual risk as a dated exception, and to patch when the maker releases a fix that the site can test and approve. Quotes in the next section; sources read 19 Nov 2026. | Condition | Response | Source | |---|---|---| | Maker has no fix yet | Long-term methods now; re-evaluate when the maker revises the advisory | NIST SP 800-40 Rev. 4, 3.5.4 | | Maker has a fix but certification or testing delays it | Same methods; get the maker's release date in writing and set the exception to end then | 800-40 Rev. 4, 2.1 | | Site could patch on its own, but only by voiding the warranty or support | Treat as unpatchable unless the maker agrees | 800-40 Rev. 4, 2.1 | | Product is end of life and will never be fixed | Replace on a dated plan; isolate meanwhile | [End-of-life software](https://remediationvsmitigation.com/decision-table/end-of-life-software/) | ## The rules it follows NIST SP 800-40 Rev. 4, Section 2.1 gives the reason and its price: "The manufacturer may require customers to update the software on a delayed schedule, such as for assets with human safety implications in a highly regulated sector, because of the extensive testing and certification that must be performed first. In these cases, organizations that choose to implement updates on their own may be voiding the product warranty and preventing future support from the manufacturer." Section 3.5.4 turns it into a plan: "There should be an approved set of methods for each maintenance group, and these methods should have been reviewed and analyzed in advance by security architects/engineers to determine their adequacy in mitigating risk." It cites micro-segmentation and software-defined perimeters from SP 800-207 as examples, and asks for extra mitigations "to prevent exploitation of specific vulnerabilities that the long-term risk mitigation methods don't adequately address". Section 3.4 adds that such assets belong in their own maintenance group rather than being labeled exceptions, and Section 3.5.5 asks that "All exceptions to maintenance plans should be reviewed regularly to determine if the maintenance plan can be implemented now." For medical devices the duty to produce patches sits with the maker. Under [21 U.S.C. 360n-2](https://www.law.cornell.edu/uscode/text/21/360n-2) (section 524B of the Federal Food, Drug, and Cosmetic Act), the sponsor of a submission for a cyber device must "make available postmarket updates and patches to the device and related systems to address (A) on a reasonably justified regular cycle, known unacceptable vulnerabilities; and (B) as soon as possible out of cycle, critical vulnerabilities that could cause uncontrolled risks". A hospital waiting on a maker can therefore ask which of those two the flaw falls under and get a date. ## A worked case: SSA-724261 and CVE-2026-109830 Siemens ProductCERT published SSA-724261 on 10 Nov 2026 (status interim, revision 1). It covers meta-iot2050, the Siemens meta-iot2050 project, from V01.03.01 with no upper version: "Affected devices take parts from a configuration file and execute them unchecked." An attacker able to change that file can run commands as root. The CVSS 3.1 base score is 7.3 (vector AV:L/AC:L/PR:L/UI:R) and the weakness is CWE-78. The remediation field reads "Currently no fix is available". CISA republished the advisory verbatim as ICSA-26-321-11 on 17 Nov 2026. NVD lists the record as awaiting analysis, and the CVE is not in KEV catalog 2026.11.18. CISA's recommended practices are the pattern the row follows: - "Minimize network exposure for all control system devices and/or systems, ensuring they are not accessible from the internet." - "Locate control system networks and remote devices behind firewalls and isolating them from business networks." - "When remote access is required, use more secure methods, such as Virtual Private Networks (VPNs), recognizing VPNs may have vulnerabilities and should be updated to the most current version available. Also recognize VPN is only as secure as the connected devices." - "CISA reminds organizations to perform proper impact analysis and risk assessment prior to deploying defensive measures." The last line matters on a plant floor or a ward: the mitigation itself can stop a process. CISA also says the text is republished "as-is" and that it is "not responsible for the editorial or technical accuracy of republished advisories", so Siemens' own advisory settles the affected versions. The flaw needs local access and a user action, so the exposure that matters is who can write the configuration file: engineering laptops, update tooling and remote-support paths. A site with these devices on an isolated cell network and file changes limited to a controlled engineering station already has most of what the advisory asks for. Dates from the Siemens CSAF advisory, the CISA page and NVD, read 19 Nov 2026. The last row is a trigger, not a date. | Date | Event | Action on the record | |---|---|---| | 10 Nov 2026 | Siemens publishes SSA-724261; NVD publishes CVE-2026-109830 | Inventory units at V01.03.01 or later; open the entry | | 17 Nov 2026 | CISA republishes as ICSA-26-321-11 | Confirm the republication adds no new instruction | | 19 Nov 2026 | KEV catalog 2026.11.18 holds no entry; advisory still shows no fix | Entry stays open; mitigation list approved | | On revision of SSA-724261 | A fix or new mitigation appears | Re-run the risk and cost-benefit steps below | ## A worked case for a medical device: the request to the maker The device is a networked infusion pump. The hospital's biomedical team learns on 2 Nov 2026 (assumed) that a vulnerability affects the pump firmware and that no update is yet released. The maker cannot be forced to a date, but 21 U.S.C. 360n-2 lets the hospital ask a pointed question. If the pump is a cyber device under the section's definition, its sponsor must "submit to the Secretary a plan to monitor, identify, and address, as appropriate, in a reasonable time, postmarket cybersecurity vulnerabilities and exploits" and make updates available either on a "reasonably justified regular cycle" for known unacceptable vulnerabilities or "as soon as possible out of cycle" for critical vulnerabilities "that could cause uncontrolled risks". The section took effect 90 days after 29 Dec 2022, so it covers submissions made after 29 Mar 2023; for an older pump, ask what the maker commits to. Dates and counts in this table are assumed for illustration. Day counts computed by this site. The statute sets no reply time. | Date | Step | Result on the record | |---|---|---| | 2 Nov 2026 | Hospital learns of the flaw; asset list shows 140 pumps on firmware below the fixed release | Entry opened; pumps moved to their own maintenance group | | 3 Nov 2026 | Pump VLAN limited to the infusion server and the biomedical engineering station; internet and general hospital access removed. Biomedical engineering checks pump delivery at the clinical test bench | Long-term methods applied; impact analysis signed by clinical engineering | | 4 Nov 2026 | Written request to the maker: is this a known unacceptable vulnerability (regular cycle) or a critical one (out of cycle), and what is the release date of the update | Request and send date kept | | 18 Nov 2026 | Maker answers: regular cycle, firmware release 15 Jan 2027 | Answer filed; 14 days after the request | | 29 Jan 2027 | Exception end: release date plus 14 days for bench testing, before any pump is updated | Exception ends 88 days after 2 Nov 2026 | | Each maker notice | Maker revises the advisory or answers another request | Re-run risk and cost-benefit steps | The exception is signed by the risk owner for the clinical service the pumps support, with clinical engineering's impact analysis and the security architect's review of the segmentation attached. NIST SP 800-40 Rev. 4 names the architect's review but no signer, so the title is this site's rule. If the maker later calls the flaw critical, the out-of-cycle duty applies and the end date moves up. ## Re-evaluation has two parts The one NIST sentence behind the review date: "Organizations should plan on periodically reevaluating their alternatives to patching." The two parts, from the same section, are a risk assessment of whether the alternatives are still effective and a cost-benefit analysis of whether the assets are worth the added cost of "mitigating, transferring, or accepting the risk of unpatchable assets". No interval is given. As this site's own rule, re-evaluate at every revision of the maker's advisory and at every new flaw in the asset, since Section 3.3 notes that "a new vulnerability in an asset might necessitate a change in the methods needed to mitigate its risk", and put a calendar date on the entry as a backstop. How to prove a control still works is in [verifying the mitigation works](https://remediationvsmitigation.com/guides/verifying-a-mitigation-works/). ## What to record - Device, maker, firmware version, the maker's update schedule and the reason it blocks patching. - The maintenance group and its approved long-term methods, the architect's review date and the last test date. [Network segmentation](https://remediationvsmitigation.com/compensating-controls/network-segmentation/) is the usual first method. - The impact analysis done before each defensive measure, and who signed it off. - The next re-evaluation date and its owner. Residual risk goes in a [dated exception](https://remediationvsmitigation.com/risk-acceptance-form/), signed by the risk owner for the process the device supports (the authorizing official on a federal system), with the architect's review attached. ## When the usual answer is wrong - The device is reachable from the internet. Remove that first; no other method makes up for it, and a VPN is only as secure as the devices behind it. - The mitigation breaks the process. Test it on a spare or in a maintenance window before it goes on a live line, as CISA's impact-analysis line says. - The maker has shipped a certified fix. The exception has run out of reason; schedule the update. - **The flaw is already in the KEV.** A federal KEV due date still binds a federal agency: BOD 26-04 allows no waiver, and if the update cannot be applied by the date, CISA's fallback is to remove the asset from the network. For a device that cannot leave the network, contact CISA about the special case. Elsewhere, isolation is a stopgap and the maker's fix date decides how long the exception can run. The risk owner signs it, with the authorizing official on a federal system. - Nobody has asked the maker for dates. A note of the request and the reply belongs on the record, and under 21 U.S.C. 360n-2 it is a fair request for a medical device. ## Sources - NIST SP 800-40 Rev. 4, Sections 2.1, 3.3, 3.4, 3.5.4, 3.5.5: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf (read 19 Nov 2026) - CISA, ICS advisory ICSA-26-321-11, Recommended Practices and revision history: https://www.cisa.gov/news-events/ics-advisories/icsa-26-321-11 (read 19 Nov 2026) - Siemens ProductCERT, SSA-724261 (CSAF), affected range and remediation: https://cert-portal.siemens.com/productcert/csaf/ssa-724261.json (read 19 Nov 2026) - NVD record for CVE-2026-109830: https://services.nvd.nist.gov/rest/json/cves/2.0?cveId=CVE-2026-109830 (read 19 Nov 2026) - 21 U.S.C. 360n-2, Ensuring cybersecurity of devices (FD&C Act section 524B): https://www.law.cornell.edu/uscode/text/21/360n-2 (read 19 Nov 2026) - CISA KEV feed, catalog version 2026.11.18: https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) - CISA, BOD 26-04 implementation guidance, waiver and alternative mitigation FAQs: https://www.cisa.gov/news-events/directives/bod-26-04-implementation-guidance-prioritizing-security-updates-based-risk (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Decision Table: Remediate, Mitigate or Accept by Situation](https://remediationvsmitigation.com/decision-table/index.md) - [Network Segmentation as a Mitigation: Cisco ISE Case](https://remediationvsmitigation.com/compensating-controls/network-segmentation/index.md) - [Risk Acceptance Form: Free Template With 15 Fields](https://remediationvsmitigation.com/risk-acceptance-form/index.md) - [End-of-Life Software: Replace, Isolate or Accept With Dates](https://remediationvsmitigation.com/decision-table/end-of-life-software/index.md) - [How to Verify a Mitigation Works: Checks for 5 Real CVEs](https://remediationvsmitigation.com/guides/verifying-a-mitigation-works/index.md) --- # What to do when a patch would break a critical system Canonical: https://remediationvsmitigation.com/decision-table/patch-breaks-critical-system/ Mitigate now and fix on a dated plan, unless the flaw is exploited and exposed. NIST SP 800-40 Rev. 4 accepts testing and outage windows as reasons to wait, but PHP2Shell (CVE-2026-109618) reached the KEV on 26 Oct 2026 with a 29 Oct due date, and CISA grants no waivers. Guidance points to: Mitigate now, fix later. Also allowed: Accept with a dated exception. Signals: Testing shows regression; An outage window is not available yet. ## The situation and the decision A vendor patch exists. A test, a pilot or a past incident shows that installing it would disrupt a system the business cannot lose, or no outage window is available yet. The decision is to put an interim control on the exposure now, record the test failure, and set a dated plan for the patch. How long that plan can run depends on whether the flaw is exploited and who can reach the system. NIST names three different troubles under the word "broken", and each calls for a different record: Categories from NIST SP 800-40 Rev. 4, Section 3.3, read 19 Nov 2026. | What is wrong | Response | What the record holds | |---|---|---| | The patch is flawed and does not correct the vulnerability | Emergency mitigation; vendor ticket; re-check when the vendor reissues | Vendor reply, the re-test result | | The patch disrupts other software or systems | Emergency mitigation; fix the conflict or schedule an outage window | Regression evidence, new test plan, next window date | | The patch itself is compromised | Incident response, outside this page | Incident number | ## What the sources allow Quotes read 19 Nov 2026. | Source | What it allows or requires | Quote | |---|---|---| | NIST SP 800-40 Rev. 4, Section 2.1 | Waiting for testing and outage windows | "The organization may need to wait for a scheduled outage window, perform testing first, update other software that interacts with the software to be patched, or train employees on new features or interfaces." | | NIST SP 800-40 Rev. 4, Section 3.3 | Emergency mitigation when the patch is the problem | "Emergency mitigations are sometimes needed because of issues with a patch." | | NIST SP 800-40 Rev. 4, Section 3.5.3 | The mitigation is temporary | "Schedules should be set and enforced for both patch deployment and mitigation removal." | | NIST SP 800-53, SI-2 b | Testing is required | "Test software and firmware updates related to flaw remediation for effectiveness and potential side effects before installation;" | | CISA, BOD 26-04 implementation guidance | No waiver for federal directive dates | "CISA does not issue waivers or exceptions for actions required in cyber directives." | For high-availability systems the same guidance says agency authorizing officials "should establish change management processes and continuity of operations plans facilitating remediation timelines in BOD 26-04", for example by putting staff time on manual testing of urgent patches. Section 3.5.2 of SP 800-40 Rev. 4 adds that even in an emergency a patch can go first to a small number of canary assets "to confirm that the patch is not corrupted and does not break the software", a stage of minutes to hours. A canary is the cheapest way to find the regression before it reaches the critical system. ## A worked case: PHP2Shell, CVE-2026-109618 CVE-2026-109618 is an out-of-memory flaw in PHP that allows unauthenticated remote code execution (CVSS 3.1 base score 9.8). The PHP Group released 8.5.12, 8.4.27, 8.3.36 and 8.2.35 on 21 Oct 2026 at 14:00 UTC. CISA's alert the same day urged immediate patching, and CISA added the CVE and its sibling CVE-2026-109619 to the KEV on 26 Oct 2026, with a due date of 29 Oct 2026 and forensic triage required. PHP 8.1 and older get no fix. What the patch changes bears on testing. PHP's advisory says it removes longjmp() for out-of-memory errors and replaces it with "a delayed, graceful soft OOM triggered at the next VM safe point, and an immediate hard OOM using _exit() if memory consumption reaches a second, higher threshold". A critical application that runs close to its memory limit is the one to run through staging first. Suppose an order-processing system on PHP 8.3 fails its staging test on 22 Oct 2026 (an assumed date) because a memory-heavy batch job now ends on the hard limit. The interim controls come from PHP's advisory, and they are incomplete by its own account: setting memory_limit=-1 is "not a complete mitigation" since a failed mmap() can still reach the flaw, and the advisory adds limits on request size and multipart parts. Release, KEV and tooling times from PHP's advisory, the KEV feed and Fastly's report; the staging row is an assumed example. | Date | Event | What the record shows | |---|---|---| | 21 Oct 2026, 14:00 UTC | Fixed releases published; exploit tooling already seen at 03:59 UTC | Patch exists; clock for the site's own policy starts | | 22 Oct 2026 (assumed) | Staging regression on the memory-heavy batch job | Failed test result attached; interim controls applied and tested | | 26 Oct 2026 | KEV listing; due date 29 Oct 2026 | Federal due date set; the interim control is no longer a plan for an agency | | 29 Oct 2026 | Federal due date | Patched, or offline; no waiver | Here the usual answer is a bridge of days, not a plan of weeks. CISA's alert, dated 21 Oct 2026 and revised on 23 and 26 Oct, says: "Organizations that cannot update immediately, or that continue to use PHP 8.1 or earlier unsupported versions, should take PHP-based systems offline until they can be patched or updated. A web application firewall may provide some protection until systems are updated or patched." For an agency, the KEV listing made 29 Oct a hard date, and the route when it cannot be met is removal from the network. ## What to record, and when it expires - The failed test: date, build, what broke and the evidence. - The interim control, its test date, and what it leaves open (here, the advisory says the configuration mitigation is not complete). - The next outage window or the date the conflict is expected to be fixed, and the approver. The [risk acceptance form](https://remediationvsmitigation.com/risk-acceptance-form/) carries the fields. - Expiry: the date of the next outage window. If it passes without the patch, the entry has expired; NIST asks that "All exceptions to maintenance plans should be reviewed regularly to determine if the maintenance plan can be implemented now." ## When the usual answer is wrong - The flaw is in the KEV and the system is exposed. See [patch available and exposed](https://remediationvsmitigation.com/decision-table/kev-patch-available-exposed/): the due date stands and CISA says no waivers. - The interim control is partial and the flaw is exploited. A web application firewall rule is a bridge of days; the [WAF rule entry](https://remediationvsmitigation.com/compensating-controls/waf-rule/) says what it leaves unfixed. - Exploitation began before the fix, as with PHP2Shell, where exploit tooling appeared about ten hours before the release. Delay is then measured against live attacks, not a calendar. - The patch is flawed rather than disruptive. Mitigation continues, and the record is the vendor ticket, not a test failure. ## Sources - NIST SP 800-40 Rev. 4, Sections 2.1, 3.3, 3.5.2, 3.5.3, 3.5.5: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf (read 19 Nov 2026) - NIST SP 800-53 Rev. 5.2.0, SI-2 b (CPRT catalog): https://csrc.nist.gov/projects/cprt/catalog#/cprt/framework/version/SP_800_53_5_2_0/home (read 19 Nov 2026) - CISA, BOD 26-04 implementation guidance: waivers and high-availability systems: https://www.cisa.gov/news-events/directives/bod-26-04-implementation-guidance-prioritizing-security-updates-based-risk (read 19 Nov 2026) - PHP Group, security advisory GHSA-w8hw-w9fx-j79w, 21 Oct 2026: https://github.com/php/php-src/security/advisories/GHSA-w8hw-w9fx-j79w (read 19 Nov 2026) - CISA alert, Critical Vulnerabilities in PHP Require Immediate Patching: https://www.cisa.gov/news-events/alerts/2026/10/21/critical-vulnerabilities-php-require-immediate-patching (read 19 Nov 2026) - CISA KEV feed, catalog version 2026.11.18: CVE-2026-109618 added 26 Oct 2026, due 29 Oct 2026: https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) - Fastly, PHP2Shell detection and response (first tooling at 03:59 UTC on 21 Oct 2026): https://www.fastly.com/blog/how-fastly-detected-and-built-a-coordinated-response-to-php2shell (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Decision Table: Remediate, Mitigate or Accept by Situation](https://remediationvsmitigation.com/decision-table/index.md) - [Risk Acceptance Form: Free Template With 15 Fields](https://remediationvsmitigation.com/risk-acceptance-form/index.md) - [KEV Entry, Patch Out, Internet-Exposed: 3 or 14 Days](https://remediationvsmitigation.com/decision-table/kev-patch-available-exposed/index.md) - [Mitigation vs Remediation Wording in Vendor Advisories](https://remediationvsmitigation.com/guides/mitigation-vs-remediation-in-vendor-advisories/index.md) --- # SaaS or a managed service you cannot patch Canonical: https://remediationvsmitigation.com/decision-table/saas-cannot-patch/ When the vendor owns the fix, change what you control and track the vendor's date. If the vendor has already patched, close the entry on its written evidence and check your tenant for earlier misuse. FedRAMP's 192-day acceptance limit and the 7 Dec 2026 FedRAMP deadline set the outer dates. Guidance points to: Mitigate now, fix later. Also allowed: Accept with a dated exception. Signals: Vendor controls the fix; Your lever is configuration, access or exposure. ## Who owes what Quotes read on 19 Nov 2026 from FedRAMP's agency POA&M page and CISA's BOD 26-04 implementation guidance. | Party | Duty | Source and quote | |---|---|---| | Cloud provider | Detect, evaluate, report, mitigate and remediate | FedRAMP: "A FedRAMP Certified cloud service provider is responsible for detecting, evaluating, reporting, mitigating, and remediating vulnerabilities in its cloud service offering." | | Customer (agency) | Write its own plan only for actions it owns | FedRAMP: agencies create POA&Ms "only when there are actions the agency needs to take, track, fund, manage, or accept" | | Customer (agency) | Stay engaged with providers | CISA guidance: agencies "remain responsible for engaging service providers directly, as needed, to ensure compliance with CISA Directives." | | Commercial provider | Report outstanding vulnerabilities to the customer | CISA guidance: the provider "must report the status of outstanding vulnerabilities to the agency receiving the service" | NIST SP 800-40 Rev. 4 puts SaaS under Transfer: "replacing conventional software installations with software-as-a-service (SaaS) usage where the SaaS vendor/managed service provider takes care of patching." Its definition of Transfer is "Reduce the risk by sharing some of the consequences with another party". The exposure is still there until the vendor fixes it, so while a vendor fix is expected the answer is mitigation and a tracked fix, not acceptance. Acceptance comes in only when the vendor says it will not fix the flaw, or the fix passes the 192-day limit below. The customer-side controls for tenants are listed in [SaaS vulnerability mitigation](https://remediationvsmitigation.com/guides/mitigation-in-cloud-and-saas/); this row decides which response to record and when to close it. ## A hosted service patched before the customer knew: BeyondTrust, 29 Aug to 16 Oct 2026 CVE-2026-102764 is an origin-validation flaw in BeyondTrust Remote Support that lets an unauthenticated remote attacker perform file operations on the appliance. BeyondTrust patched its hosted (SaaS) customers itself, so no hosted customer could patch at any point. The advisory dates the SaaS patch as of 9 Oct and the blog dates the patch release 8 Oct. Day counts are from 29 Aug. Sources: BeyondTrust BT26-05 and blog, CISA KEV feed 2026.11.18, all read 19 Nov 2026. | Date | Event | Day | What a hosted customer could do | |---|---|---|---| | 29 Aug 2026 | A suspected China-nexus actor exploits the flaw against one BeyondTrust-hosted instance and one self-hosted appliance | 0 | Nothing: no notice existed | | 28 Sep 2026 | BeyondTrust threat hunting finds the hosted-instance intrusion and contains it within 27 minutes | 30 | Nothing: no notice existed | | 8 Oct 2026 | BeyondTrust releases the patch; SaaS instances start to be patched automatically (blog) | 40 | Nothing to install | | 9 Oct 2026 | Advisory: "As of Oct 9, 2026, a patch that remediates this vulnerability has been applied to all Remote Support SaaS customers" | 41 | Nothing to install | | 15 Oct 2026 | Coordinated disclosure: CVE, advisory BT26-05 (CVSS v4 9.3), blog post, Mandiant analysis | 47 | Decide the register entry | | 16 Oct 2026 | CISA adds the CVE to KEV with a due date of 19 Oct 2026, 3 days later | 48 | Federal agencies: follow the BOD 26-04 cloud guidance | On 15 Oct a hosted customer faces a choice between two records. The vulnerability entry can close at once: the advisory confirms the SaaS patch as of 9 Oct, and the blog says "BeyondTrust SaaS customers have already been patched, and no further action is required." Keep the advisory page and its 15 Oct issue date as the closing evidence. No mitigation entry is needed for a flaw that is already fixed. The second record is not closed by the patch. The flaw was exploited for 40 days before the patch was released and 41 before the advisory confirms it on every SaaS instance, and BeyondTrust writes that it is "only aware of the two impacted customers" and has "not seen evidence of adversary activity since August". That is the vendor's statement about its own telemetry. Open an entry in the incident or security-review log (it is a check for past misuse, not a control), ask the provider in writing whether your tenant was among the affected instances, and close it with the reply and the review below. ## What a customer can still review in its own tenant BeyondTrust's blog lists what to hunt for after this actor. Four of its items can be read in a hosted tenant's own audit and session records, and each is a customer-side check: - Privileged or administrative sessions from residential broadband or commercial VPN addresses that nobody can tie to a known administrator or a ticket. - Unauthorized account, permission or session-policy changes, "particularly modified permissions for 'Jump Items' or 'Teams'". - New multi-factor registrations on privileged accounts. - Stored credentials authenticating outside a managed session or from unfamiliar locations. If anything is found, BeyondTrust says: "Rotate every credential the appliance stored or injected." Record the review in that incident or review log with the date it was run and the period covered, 29 Aug to 9 Oct 2026. It does not belong among the compensating controls, which stand in for a fix. For a flaw with no vendor fix yet, controls do apply, and those go in that register: switch off the affected feature, limit which networks and accounts reach the tenant, and rotate any credential the advisory names. FedRAMP's rules describe the provider's duty in the same terms: "If it is not possible to fully mitigate vulnerabilities or remediate vulnerabilities, providers SHOULD instead partially mitigate vulnerabilities promptly, progressively, and persistently." ## When the vendor has not fixed it FedRAMP's rule VER-TFR-MAV says providers "MUST categorize any vulnerability that is not or will not be fully mitigated or remediated within 192 days of evaluation as an accepted vulnerability." Counted from an evaluation on 15 Oct 2026, 192 days ends on 25 Apr 2027 (our count). CISA's BOD 26-04 guidance says FedRAMP "will require mandatory adoption of the new Vulnerability Detection and Response (VDR) and Vulnerability Evaluation and Reporting (VER) rules by December 7, 2026". The older FedRAMP playbook, reference only since 24 Jun 2026, asked for high-risk vendor dependencies to be "mitigated to a Moderate level through compensating controls within thirty (30) days" and for a vendor check-in "at least once a month". No source sets a number for non-federal customers, so set your own review date from the provider's committed fix date and write the reason beside it. ## Record to keep, review date, what changes the answer - Record: the provider's advisory or status notice with its date, your configuration changes and the dates they were tested, the contact who confirmed the vendor's fix plan, and the date of the last vendor check-in. - Review date: the provider's committed fix date, or your own date with the reason written down. For a vendor with no fix date, review monthly. - Changes the answer: the provider publishes the fix (record the close date); the flaw reaches the CISA catalog (the provider's due date then applies, see [patch available and exposed](https://remediationvsmitigation.com/decision-table/kev-patch-available-exposed/)); or the provider will not fix it, or the 192-day limit passes, which turns the entry into a [dated exception](https://remediationvsmitigation.com/risk-acceptance-form/) or an exit plan. That is the only point at which acceptance is the right record. The limits of transfer are covered in [risk transfer versus risk acceptance](https://remediationvsmitigation.com/guides/risk-acceptance-vs-risk-transfer/), and the plan format for agencies in [the POA&M guide](https://remediationvsmitigation.com/guides/poam-plan-of-action-and-milestones/). ## Sources - BeyondTrust, security advisory BT26-05 (CVE-2026-102764): https://www.beyondtrust.com/trust-center/security-advisories/bt26-05 (read 19 Nov 2026) - BeyondTrust, ARMORY CASTLE and BT26-05: What Security Teams Need to Know: https://www.beyondtrust.com/blog/entry/bt26-05 (read 19 Nov 2026) - CISA, KEV JSON feed, catalogVersion 2026.11.18: https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) - NIST SP 800-40 Rev. 4, Section 2.1 (Transfer): https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf (read 19 Nov 2026) - FedRAMP, Managing POA&Ms for agencies: https://www.fedramp.gov/2026/agencies/use/ongoing/poams/ (read 19 Nov 2026) - CISA, BOD 26-04 implementation guidance, cloud and FedRAMP FAQs: https://www.cisa.gov/news-events/directives/bod-26-04-implementation-guidance-prioritizing-security-updates-based-risk (read 19 Nov 2026) - FedRAMP, Vulnerability Detection and Response rules 2026: https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-detection-and-response/ (read 19 Nov 2026) - FedRAMP, Vulnerability Evaluation and Reporting rules 2026, VER-TFR-MAV: https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-evaluation-and-reporting/ (read 19 Nov 2026) - FedRAMP legacy POA&M playbook (reference only after 24 Jun 2026): https://www.fedramp.gov/legacy/playbook/csp/authorization/poam/ (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Decision Table: Remediate, Mitigate or Accept by Situation](https://remediationvsmitigation.com/decision-table/index.md) - [Risk Acceptance Form: Free Template With 15 Fields](https://remediationvsmitigation.com/risk-acceptance-form/index.md) - [SaaS Vulnerability Mitigation: Steps for the Customer](https://remediationvsmitigation.com/guides/mitigation-in-cloud-and-saas/index.md) - [Risk Transfer vs Risk Acceptance for Security Findings](https://remediationvsmitigation.com/guides/risk-acceptance-vs-risk-transfer/index.md) - [POA&M Example: A Filled Plan of Action and Milestones Entry](https://remediationvsmitigation.com/guides/poam-plan-of-action-and-milestones/index.md) --- # Editorial policy Canonical: https://remediationvsmitigation.com/editorial-policy/ Every claim rests on text read from its source on a stated date, 19 Nov 2026 for the first edition. Standards and government text come first, quotes stay short and exact, vendor statements are labeled as the vendor's, and a page changes its modified date only when a fact on it changes. ## How each kind of statement is treated | Kind of statement | Treatment | Example on the site | |---|---|---| | A definition | Quoted from the issuer in quotation marks, with the document and section | Remediation as "eliminated through patching, decommissioning the system, or another action" (CISA BOD 26-04) | | A count | Computed by script from a named version of the source, with the read date | 1,786 KEV entries in catalog version 2026.11.18 | | A date | Taken from the issuer's own page; a day count is marked as arithmetic by this site | 7 Dec 2026 to 7 Mar 2027 is 90 days | | A vendor claim | Shown as the vendor's and linked to the vendor page | Cloudflare managed ruleset documentation | | An absence | Written as "None found on" plus the date, a statement about one day's reading | No KEV required action read on 19 Nov 2026 names MFA | | An inference | Labeled as where the cited text points, never as a ruling on an organization | A decision row's guidance label | Each fact in the data files carries one label: primary (text from the standard, law or directive itself), vendor (the vendor's own statement), third-party (a report by someone else) or none-found (a dated search that returned nothing). ## Sources and quotes - **Primary text first.** Directives, standards and regulator pages from CISA, NIST, FedRAMP and PCI SSC are quoted from their own pages. Where a PDF was read, the quote was matched to the extracted text. - **Short quotes, exact wording.** A quote keeps the source's words; only typographic quotation marks are changed to plain ones. Anything longer than a sentence is paraphrased and linked. - **Standards with closed text.** Where a standard's own text cannot be opened, the page cites an official public page of the issuer that states the requirement, such as a PCI SSC FAQ, or leaves the clause out. A source that was not opened is never cited, and no reader page carries a note about it. - **Numbers only with a source.** A number appears with the document, the section or row, and the read date, as in BOD 26-04 Table 1, read from the table image. ## Wording and safety The two key words are used as the sources use them. Remediation means the vulnerability is eliminated (BOD 26-04 wording); mitigation means exposure or impact is reduced while the vulnerability remains. Where a source uses them differently, such as NIST SP 800-40 Rev. 4 counting patching as a form of mitigating risk (Section 2.1), the page says so. The [difference guide](https://remediationvsmitigation.com/guides/difference-between-remediation-and-mitigation/) sets this out. The site contains no exploit code, proof-of-concept steps, payloads or scanning and probing instructions. Controls are described by what they change, how to verify them and what evidence to keep. ## Changes and licensing A fact that changes at its source is updated with a new read date, and the change is logged on the [corrections page](https://remediationvsmitigation.com/corrections/), which is empty so far. Report an error to [editor@remediationvsmitigation.com](mailto:editor@remediationvsmitigation.com). Data and forms are under [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/); the [terms](https://remediationvsmitigation.com/terms/) give the cite line. For the scope of the site see the [about page](https://remediationvsmitigation.com/about/). ## Sources - CISA, BOD 26-04 (definition of remediation): https://www.cisa.gov/news-events/directives/bod-26-04-prioritizing-security-updates-based-risk (read 19 Nov 2026) - NIST SP 800-40 Rev. 4, Section 2.1: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf (read 19 Nov 2026) - CISA KEV catalog JSON feed, catalogVersion 2026.11.18: https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) - Creative Commons, Attribution 4.0 International deed: https://creativecommons.org/licenses/by/4.0/ (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Corrections Log: How to Report an Error](https://remediationvsmitigation.com/corrections/index.md) - [About Remediation vs Mitigation: Scope and Hosting](https://remediationvsmitigation.com/about/index.md) --- # Exception register Canonical: https://remediationvsmitigation.com/exception-register/ The register has 15 columns, a filled example row and a blank CSV header. Two fields hold status: exception_state tracks the exception itself, and vulnerability_state uses FedRAMP's six 2026 vulnerability definitions. Keeping them apart lets an expired exception and a still-detected finding both show. Read 19 Nov 2026. ## The 15 columns, with one row filled in One row is one signed [Risk Acceptance Form](https://remediationvsmitigation.com/risk-acceptance-form/) or one interim mitigation waiting on a fix. The filled row is an invented organization's record of a real finding, as it stands on 19 Nov 2026: CVE-2026-65660, a SharePoint Server code-injection flaw that CISA added to KEV on 25 Sep 2026 with a due date of 28 Sep 2026 (CISA KEV feed 2026.11.18). Microsoft's fixed build for SharePoint 2019 is 16.0.10417.20198, delivered as two packages: KB5002894 (Core) and KB5002896 (Language Pack). A private company can sign this exception. A federal civilian agency cannot waive a KEV due date, because "CISA does not issue waivers or exceptions for actions required in cyber directives" (BOD 26-04 implementation guidance). Column definitions and the form field each comes from. Example values are invented except the CVE, the KEV dates and the fixed build, which are real (read 19 Nov 2026). The row is in renewal due because its expiry, 12 Dec 2026, is 23 days after the read date. | Column | Holds | Filled row | Form field | |---|---|---|---| | exception_id | Your own number, unique and never reused | EX-2026-031 | 1 | | finding_ref | CVE ID and scanner finding IDs | CVE-2026-65660 | 1 | | asset | System IDs, environment, exposure | SP-FARM-02, SharePoint Server 2019 below build 16.0.10417.20198, prod, internal with partner VPN access | 3 | | response | accepted, interim mitigation or vendor dependency | interim mitigation | 7, 12 | | exception_state | active, renewal due, lapsed, escalated or closed (defined below) | renewal due | Not on the form | | vulnerability_state | One of the six FedRAMP states (defined below), as of the last scan | partially mitigated | Not on the form | | approver_role | Role that signed; the name stays on the form | CISO | 13 | | decision_date | Date of signature, as YYYY-MM-DD | 2026-09-28 | 13 | | expiry_date | Date the exception lapses unless renewed | 2026-12-12 | 14 | | next_review_date | Next review under the cadence in field 10 | 2026-11-30 | 10 | | control_ids | Compensating control entries used | CC-007 (anonymous viewing off), CC-012 (partner access through the VPN only) | 8 | | plan_milestone_date | Date of the next milestone | 2026-11-26 (patch in the change window) | 9 | | vendor_last_checkin | Date of the last vendor contact; blank if no vendor fix is awaited | blank: the fix exists | 12 | | closed_date | Date the exception was closed | blank | Not on the form | | closure_evidence | Rescan reference, change ticket or decommission record | blank | Not on the form | Blank header and empty row for a spreadsheet or database import: exception_id,finding_ref,asset,response,exception_state,vulnerability_state,approver_role,decision_date,expiry_date,next_review_date,control_ids,plan_milestone_date,vendor_last_checkin,closed_date,closure_evidence ,,,,,,,,,,,,,, The same header is in the [downloadable CSV](https://remediationvsmitigation.com/exception-register/exception-register.csv). ## Two status fields, kept apart A finding has a state, and so does the exception that covers it. A register with one status column loses one of them: an exception can lapse while the scanner still detects the flaw, and a flaw can be fixed while the paperwork is still open. FedRAMP's 2026 definitions describe the finding. The exception lifecycle below is this register's own; no standard defines it. Exception lifecycle defined by this register. | exception_state | Meaning | Who moves it | Moves to | |---|---|---|---| | active | Signed and inside its expiry date | Approver signs | renewal due, escalated or closed | | renewal due | Expiry is 30 days away or less | Register owner, on the monthly review | closed (renewed as a new row), lapsed or escalated | | lapsed | Expiry date passed with no signature | Register owner | escalated, or closed if the finding is remediated | | escalated | Sent to a senior owner under criteria written in advance | Register owner | closed, or a new row with a new decision | | closed | Remediated, false positive, decommissioned or renewed as a new row | Register owner, only with closure_evidence | Terminal | A renewal is a new signature. Copy the row to a new exception_id with the new decision_date and expiry_date, then set the old row to closed with closure_evidence "renewed as EX-...". The old row stays as the audit trail and no ID is reused. FedRAMP Consolidated Rules for 2026, definitions, read 19 Nov 2026. FedRAMP writes them for cloud providers; the register borrows the vocabulary. | vulnerability_state | FedRAMP definition (2026) | Expected response | |---|---|---| | accepted | "A vulnerability that the provider does not intend to fully mitigate or remediate, OR that has not or will not be fully mitigated or remediated within the maximum overdue period" | accepted | | overdue | "intends to fully mitigate or remediate but has not or will not do so within the time frames recommended or required by FedRAMP" | vendor dependency, or a late patch | | partially mitigated | "likelihood or Potential Agency Impact N-rating has been reduced from the original evaluation but the risk of exploitation still exists" | interim mitigation | | fully mitigated | "reduced from the original evaluation until either are negligible, but the vulnerability is still detected" | interim mitigation, scanner still reports it | | remediated | "neutralized or eliminated and is no longer detected" | Closes the row: rescan evidence | | false positive | "not actually present in an exploitable state in the information resource" | Closes the row: evidence of why | An interim mitigation row therefore carries partially mitigated or fully mitigated, never remediated. A remediated or fully mitigated vulnerability cannot also be a false positive; FedRAMP's definition of false positive says so. When vulnerability_state reaches remediated or false positive, exception_state becomes closed. The false positive route is in [vulnerability false positive handling](https://remediationvsmitigation.com/guides/false-positive-handling/). ## The monthly review routine 1. Sort by expiry_date. Every row expiring within 30 days moves to renewal due; the approver renews in writing or the row moves on. 2. Sort by next_review_date and clear every row that is due. NIST SP 800-40 Rev. 4 asks that exceptions be "reviewed regularly to determine if the maintenance plan can be implemented now" (Section 3.5.5). 3. Re-check each control_ids entry. NIST SP 800-40 Rev. 4 step 3d says to make sure no one "deactivates the additional security controls" (Section 2.2). 4. Look up each finding_ref on the CISA KEV catalog. A KEV listing reopens the row; for a federal civilian agency the BOD 26-04 date then applies and no waiver exists. 5. Read vendor_last_checkin. The legacy FedRAMP playbook asks for a check "at least once a month". 6. Close rows only with closure_evidence. A row marked remediated needs a rescan that no longer detects the finding, because that is what FedRAMP's definition requires. In the filled row, expiry on 12 Dec 2026 put EX-2026-031 into renewal due on 12 Nov 2026, 30 days earlier, whether or not any milestone was met. At the 19 Nov review step 1 lists it. If the patch lands on the 26 Nov milestone, the row closes with the rescan as closure_evidence; if not, the approver signs a renewal as a new row before 12 Dec or the row lapses. Step 4 applies from day one, because the CVE was already on KEV when the row was signed on 28 Sep. ## Accepted rows stay on the register The legacy FedRAMP playbook says "Approved ORs are still considered open risks" and that they must be "periodically reassessed". Expiry does not remove a row: it moves to closed, lapsed or escalated, never to deleted. Two other pages explain the neighbors: [risk acceptance versus risk exception](https://remediationvsmitigation.com/guides/risk-acceptance-vs-risk-exception/) for why a register holds both kinds of row, and [risk register versus exception register](https://remediationvsmitigation.com/guides/risk-register-vs-exception-register/) for how the two link. Expiry rules are in [risk exception expiry](https://remediationvsmitigation.com/guides/exception-expiry-and-renewal/), rows waiting on a fix in [the POA&M guide](https://remediationvsmitigation.com/guides/poam-plan-of-action-and-milestones/), and the monthly control check in [verifying a mitigation](https://remediationvsmitigation.com/guides/verifying-a-mitigation-works/). ## Sources - FedRAMP Consolidated Rules for 2026, definitions: accepted, overdue, partially mitigated, fully mitigated, remediated, false positive: https://www.fedramp.gov/2026/definitions/ (read 19 Nov 2026) - NIST SP 800-40 Rev. 4, Sections 2.2 and 3.5.5: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf (read 19 Nov 2026) - FedRAMP legacy POA&M playbook, General POA&M Guidance (reference only after 24 Jun 2026): https://www.fedramp.gov/legacy/playbook/csp/authorization/poam/ (read 19 Nov 2026) - CISA, BOD 26-04 implementation guidance, FAQ on waivers: https://www.cisa.gov/news-events/directives/bod-26-04-implementation-guidance-prioritizing-security-updates-based-risk (read 19 Nov 2026) - CISA, KEV JSON feed, catalogVersion 2026.11.18, entry CVE-2026-65660: https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) - Microsoft MSRC, August 2026 security updates, CVE-2026-65660: https://api.msrc.microsoft.com/cvrf/v3.0/cvrf/2026-Aug (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Risk Acceptance Form: Free Template With 15 Fields](https://remediationvsmitigation.com/risk-acceptance-form/index.md) - [Risk Acceptance vs Risk Exception in Cyber Security](https://remediationvsmitigation.com/guides/risk-acceptance-vs-risk-exception/index.md) - [Maximum Duration of a Risk Acceptance: 9 Sources](https://remediationvsmitigation.com/risk-acceptance-form/maximum-durations/index.md) - [Risk Register vs Exception Register: How They Link](https://remediationvsmitigation.com/guides/risk-register-vs-exception-register/index.md) - [Risk Exception Expiry: What Happens and How to Renew](https://remediationvsmitigation.com/guides/exception-expiry-and-renewal/index.md) - [POA&M Example: A Filled Plan of Action and Milestones Entry](https://remediationvsmitigation.com/guides/poam-plan-of-action-and-milestones/index.md) - [Vulnerability False Positive: Close It With Evidence](https://remediationvsmitigation.com/guides/false-positive-handling/index.md) - [How to Verify a Mitigation Works: Checks for 5 Real CVEs](https://remediationvsmitigation.com/guides/verifying-a-mitigation-works/index.md) --- # Remediation and mitigation glossary Canonical: https://remediationvsmitigation.com/glossary/ Thirty-one terms for remediating, mitigating, accepting and excepting security findings, each defined in the issuer's own words with a sourced example or note. Definitions come from NIST, FedRAMP, CISA, PCI SSC, OWASP and Microsoft, all read on 19 Nov 2026, and the entries run A to Z. ## Accepted vulnerability FedRAMP defines it as "A vulnerability that the provider does not intend to fully mitigate or remediate, OR that has not or will not be fully mitigated or remediated within the maximum overdue period in FedRAMP Vulnerability Detection and Response rules" ([definitions](https://www.fedramp.gov/2026/definitions/)). Example: the definition refers to a "maximum overdue period"; the figure is set in rule VER-TFR-MAV of the Vulnerability Evaluation and Reporting rules: 192 days from evaluation ([FedRAMP](https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-evaluation-and-reporting/)). A flaw evaluated on 15 Oct 2026 and still unfixed becomes accepted on 25 Apr 2027. See [maximum durations](https://remediationvsmitigation.com/risk-acceptance-form/maximum-durations/). ## Authorization termination date The date on which an authorization expires. NIST SP 800-37 Rev. 2 says it "is established by the authorizing official and indicates when the authorization expires". Example: It is the model for field 14 of the [Risk Acceptance Form](https://remediationvsmitigation.com/risk-acceptance-form/). Source: [SP 800-37 Rev. 2, Task R-4](https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-37r2.pdf). ## Authorizing official NIST: "A senior Federal official or executive with the authority to authorize (i.e., assume responsibility for) the operation of an information system" ([glossary](https://csrc.nist.gov/glossary/term/authorizing_official)). Also: Acceptance of risk "cannot be delegated to other officials within the organization" ([SP 800-37 Rev. 2, Task R-4](https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-37r2.pdf)). See [who may sign](https://remediationvsmitigation.com/risk-acceptance-form/who-may-sign/). ## Compensating control NIST: "The security and privacy controls employed in lieu of the controls in the baselines described in NIST Special Publication 800-53B that provide equivalent or comparable protection for a system or organization" ([glossary](https://csrc.nist.gov/glossary/term/compensating_controls)). Example: PCI SSC treats a compensating control as one way to address a vulnerability that is not resolved, for example "with a compensating control or by disabling a vulnerable service" ([FAQ 1597](https://www.pcisecuritystandards.org/faqs/1597)). ## Decommission Retiring a system or device so the vulnerable component no longer exists in the environment. CISA BOD 26-04 counts it as a way to remediate: elimination "through patching, decommissioning the system, or another action". Example: BOD 26-02 requires agencies to "Decommission all identified EOS edge devices from agency networks" ([BOD 26-02](https://www.cisa.gov/news-events/directives/bod-26-02-mitigating-risk-end-support-edge-devices)). ## Exception (to a maintenance plan) An asset that cannot follow its normal patch plan. NIST SP 800-40 Rev. 4 says organizations "should closely track and monitor all exceptions to maintenance plans". Also: The same section says "Assets with similar long-term exceptions might need to be moved to a separate maintenance group with its own maintenance plan" ([Section 3.5.5](https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf)). See [acceptance versus exception](https://remediationvsmitigation.com/guides/risk-acceptance-vs-risk-exception/). ## False positive vulnerability FedRAMP: "A detected vulnerability that is not actually present in an exploitable state in the information resource" ([definitions](https://www.fedramp.gov/2026/definitions/)). Also: the definition covers vulnerable code that exists on a machine but "not loaded, running, or otherwise in an operating state required for exploitation", and excludes anything remediated or fully mitigated. See [false positive handling](https://remediationvsmitigation.com/guides/false-positive-handling/). ## Fully mitigated vulnerability FedRAMP: "A vulnerability where the likelihood of exploitation or Potential Agency Impact N-rating has been reduced from the original evaluation until either are negligible, but the vulnerability is still detected" ([definitions](https://www.fedramp.gov/2026/definitions/)). Also: FedRAMP adds that a fully mitigated vulnerability still exists until it is remediated ([VDR-CSO-RES](https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-detection-and-response/)), and still expects KEVs to be remediated by the CISA date. ## Internet-reachable vulnerability FedRAMP: "A vulnerability in a machine-based information resource that might be exploited or otherwise triggered by a payload originating from a source on the public internet" ([definitions](https://www.fedramp.gov/2026/definitions/)). Also: A resource with "no direct route to/from the internet" still counts if it receives payloads or acts on internet activity. This is field 4 of the [form](https://remediationvsmitigation.com/risk-acceptance-form/). ## Known exploited vulnerability (KEV) FedRAMP, citing CISA: "any vulnerability identified in CISA's Known Exploited Vulnerabilities catalog" ([definitions](https://www.fedramp.gov/2026/definitions/)). Example: Catalog version 2026.11.18 held 1786 entries, read from the [CISA feed](https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json) on 19 Nov 2026. See the [KEV rows of the Decision Table](https://remediationvsmitigation.com/decision-table/kev-patch-available-exposed/). ## Likely exploitable vulnerability FedRAMP: a vulnerability that is not fully mitigated and is reachable by a likely threat actor who "would likely gain unauthorized access, cause harm, disrupt operations, or otherwise have an undesired adverse impact" ([definitions](https://www.fedramp.gov/2026/definitions/)). Also: The note sets a floor: "any vulnerability that an automated unauthenticated system can exploit over the internet is a likely exploitable vulnerability". ## Mitigation NIST SP 800-216: "The temporary reduction or lessening of the impact of a vulnerability or the likelihood of its exploitation" ([glossary](https://csrc.nist.gov/glossary/term/mitigation)). Also: FedRAMP separates it from remediation: "Mitigation is the process of reducing the risk and impact of a vulnerability" ([VDR-CSO-RES](https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-detection-and-response/)). ## Ongoing authorization NIST SP 800-37 Rev. 2, Task R-4: an organization may eliminate the authorization termination date "if the system is operating under an ongoing authorization", where continuous monitoring gives the authorizing official "the needed information to conduct ongoing risk determination and risk acceptance" ([SP 800-37 Rev. 2](https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-37r2.pdf)). ## Overdue vulnerability FedRAMP: "A vulnerability that the provider intends to fully mitigate or remediate but has not or will not do so within the time frames recommended or required by FedRAMP" ([definitions](https://www.fedramp.gov/2026/definitions/)). Also: It differs from an accepted vulnerability by intent: the provider still plans to fix it. The two get separate statuses in the [exception register](https://remediationvsmitigation.com/exception-register/). ## Partially mitigated vulnerability FedRAMP: a vulnerability whose likelihood or impact level "has been reduced from the original evaluation but the risk of exploitation still exists" ([definitions](https://www.fedramp.gov/2026/definitions/)). Also: Where full mitigation or remediation is not possible, providers "SHOULD instead partially mitigate vulnerabilities promptly, progressively, and persistently" ([VDR-CSO-RES](https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-detection-and-response/)). ## Plan of action and milestones (POA&M) NIST: "A document that identifies tasks that need to be accomplished. It details resources required to accomplish the elements of the plan, milestones for meeting the tasks, and the scheduled completion dates for the milestones" ([glossary](https://csrc.nist.gov/glossary/term/plan_of_action_and_milestones)). Example: FedRAMP tells agencies to create a POA&M "only when there are actions the agency needs to take, track, fund, manage, or accept" ([agency POA&M page](https://www.fedramp.gov/2026/agencies/use/ongoing/poams/)). ## Potential Agency Impact (PAIN) FedRAMP, which lists PAIN as the short form of Potential Agency Impact N-rating: "The estimated cumulative effect of unauthorized access, disruption, harm, or other adverse impacts to all agencies using the cloud service" ([definitions](https://www.fedramp.gov/2026/definitions/)). Example: it drives the timeframes in VDR-TFR-PVR. A Class D provider has 12 hours for a PAIN-5 flaw that is likely exploitable and internet-reachable, and 192 days for a PAIN-2 flaw that is not ([FedRAMP](https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-detection-and-response/)). ## Qualified Security Assessor (QSA) In PCI DSS compliance, the assessor who judges compensating controls. PCI SSC: "QSAs are responsible to determine whether a compensating control is sufficient to meet the intent of a requirement" ([FAQ 1046](https://www.pcisecuritystandards.org/faqs/1046)). Example: when a critical security patch was not installed within 30 days of release, PCI SSC says an assessor can find the requirement "In Place" only with assurance of four things: a repeatable documented process, an exceptional circumstance, the cause addressed and recurrence prevented. Otherwise the assessor can consider "Not in Place" ([FAQ 1572](https://www.pcisecuritystandards.org/faqs/1572), June 2025). ## Remediated vulnerability FedRAMP: "A vulnerability that has been neutralized or eliminated and is no longer detected" ([definitions](https://www.fedramp.gov/2026/definitions/)). Example: The phrase "no longer detected" makes a rescan the closing evidence; the register column for it is described on the [exception register page](https://remediationvsmitigation.com/exception-register/). ## Remediation NIST SP 800-216: "The neutralization or elimination of a vulnerability or the likelihood of its exploitation"; CNSSI 4009-2022 also lists "The act of mitigating a vulnerability or a threat" ([glossary](https://csrc.nist.gov/glossary/term/remediation)). Also: CISA BOD 26-04 uses the narrower meaning: "the vulnerability is eliminated through patching, decommissioning the system, or another action" ([BOD 26-04](https://www.cisa.gov/news-events/directives/bod-26-04-prioritizing-security-updates-based-risk)). ## Residual risk The risk that remains after a risk response. NIST SP 800-37 Rev. 2: "regardless of the risk response, there remains a degree of residual risk". Also: Organizations "determine acceptable degrees of residual risk based on organizational risk tolerance" ([SP 800-37 Rev. 2, Task R-3](https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-37r2.pdf)). SP 800-39, Task 3-3, adds "and the specific risk tolerances of particular decision makers" ([SP 800-39](https://nvlpubs.nist.gov/nistpubs/Legacy/SP/nistspecialpublication800-39.pdf)). It is field 11 of the [form](https://remediationvsmitigation.com/risk-acceptance-form/). ## Risk acceptance NIST SP 800-39: "Risk acceptance is the appropriate risk response when the identified risk is within the organizational risk tolerance" ([Task 3-1](https://nvlpubs.nist.gov/nistpubs/Legacy/SP/nistspecialpublication800-39.pdf)). Also: SP 800-40 Rev. 4 adds that "By default, an organization accepts the risk posed by using its software" ([Section 2.1](https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf)). ## Risk avoidance NIST SP 800-40 Rev. 4: "Ensure that the risk does not occur by eliminating the attack surface, such as by uninstalling the vulnerable software, decommissioning assets with the vulnerabilities, or disabling computing capabilities in assets that can function without them" ([Section 2.1](https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf)). Also: SP 800-39 sets the trigger: avoidance "may be the appropriate risk response when the identified risk exceeds the organizational risk tolerance". ## Risk mitigation NIST SP 800-30 Rev. 1: "Prioritizing, evaluating, and implementing the appropriate risk-reducing controls/countermeasures recommended from the risk management process. A subset of Risk Response" ([glossary](https://csrc.nist.gov/glossary/term/risk_mitigation)). Also: SP 800-40 Rev. 4 includes patching in its mitigate response, which is why the word is not used the same way across publications. ## Risk response NIST: "Accepting, avoiding, mitigating, sharing, or transferring risk to organizational operations, organizational assets, individuals, other organizations, or the Nation" ([glossary](https://csrc.nist.gov/glossary/term/risk_response)). Also: SP 800-40 Rev. 4 names four for software vulnerabilities (accept, mitigate, transfer, avoid) and does not list sharing separately ([Section 2.1](https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf)). ## Risk tolerance NIST: "The level of risk or the degree of uncertainty that is acceptable to an organization" ([glossary](https://csrc.nist.gov/glossary/term/risk_tolerance)). Also: SP 800-39 notes "There is no correct level of organizational risk tolerance" ([SP 800-39](https://nvlpubs.nist.gov/nistpubs/Legacy/SP/nistspecialpublication800-39.pdf)). See [risk acceptance criteria](https://remediationvsmitigation.com/guides/risk-acceptance-criteria/). ## Risk transfer NIST SP 800-39: risk transfer "shifts the entire risk responsibility or liability from one organization to another organization", for example by insurance. Also: The same text warns that "risk transfer reduces neither the likelihood of harmful events occurring nor the consequences" ([Task 3-1](https://nvlpubs.nist.gov/nistpubs/Legacy/SP/nistspecialpublication800-39.pdf)). ## Vendor dependency In the legacy FedRAMP playbook, a finding whose fix waits on a vendor patch that does not yet exist. The playbook is marked reference only after 24 Jun 2026. Also: It required that providers "check in with the vendor at least once a month" and said high-risk vendor dependencies must be mitigated to a Moderate level within 30 days ([playbook](https://www.fedramp.gov/legacy/playbook/csp/authorization/poam/)). ## Virtual patch OWASP: "A security policy enforcement layer that detects and blocks matching exploitation attempts against a known vulnerability without changing the vulnerable code" ([cheat sheet](https://cheatsheetseries.owasp.org/cheatsheets/Virtual_Patching_Cheat_Sheet.html)). Example: Fastly offers a virtual patch for CVE-2026-75650 in Adobe Commerce, and Sansec says to install Adobe's hotfix anyway, so the fix is still owed ([Fastly](https://www.fastly.com/blog/what-is-style-smuggler-cve-2026-75650)). OWASP says a virtual patch leaves the underlying vulnerability in place. ## Waiver A release from a required action. CISA states that it "does not issue waivers or exceptions for actions required in cyber directives" ([BOD 26-04 guidance](https://www.cisa.gov/news-events/directives/bod-26-04-implementation-guidance-prioritizing-security-updates-based-risk)). Example: A missed KEV date therefore cannot be waived; see the KEV row of the [Decision Table](https://remediationvsmitigation.com/decision-table/). ## Workaround Microsoft's 2010 bulletin MS10-046 defines it as "a setting or configuration change that does not correct the underlying vulnerability but would help block known attack vectors before you apply the update" ([bulletin](https://learn.microsoft.com/en-us/security-updates/securitybulletins/2010/ms10-046)). Also: the same bulletin uses "mitigation" for a setting or common configuration "existing in a default state" that could reduce severity. See [workaround versus mitigation](https://remediationvsmitigation.com/guides/workaround-vs-mitigation/). Longer explanations of terms used above: [risk tolerance and risk acceptance](https://remediationvsmitigation.com/guides/risk-acceptance-vs-risk-tolerance/), [residual risk](https://remediationvsmitigation.com/guides/residual-risk-after-mitigation/), [inherent versus residual risk](https://remediationvsmitigation.com/guides/inherent-vs-residual-risk/), [risk owner](https://remediationvsmitigation.com/guides/risk-owner/), [plan of action and milestones](https://remediationvsmitigation.com/guides/poam-plan-of-action-and-milestones/) and [false positive](https://remediationvsmitigation.com/guides/false-positive-handling/). ## Sources - FedRAMP Consolidated Rules for 2026, definitions: https://www.fedramp.gov/2026/definitions/ (read 19 Nov 2026) - FedRAMP 2026, Vulnerability Detection and Response: https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-detection-and-response/ (read 19 Nov 2026) - FedRAMP 2026, Vulnerability Evaluation and Reporting: https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-evaluation-and-reporting/ (read 19 Nov 2026) - FedRAMP 2026, Managing POA&Ms: https://www.fedramp.gov/2026/agencies/use/ongoing/poams/ (read 19 Nov 2026) - FedRAMP legacy POA&M playbook: https://www.fedramp.gov/legacy/playbook/csp/authorization/poam/ (read 19 Nov 2026) - NIST CSRC glossary, remediation: https://csrc.nist.gov/glossary/term/remediation (read 19 Nov 2026) - NIST CSRC glossary, mitigation: https://csrc.nist.gov/glossary/term/mitigation (read 19 Nov 2026) - NIST CSRC glossary, risk mitigation: https://csrc.nist.gov/glossary/term/risk_mitigation (read 19 Nov 2026) - NIST CSRC glossary, risk response: https://csrc.nist.gov/glossary/term/risk_response (read 19 Nov 2026) - NIST CSRC glossary, risk tolerance: https://csrc.nist.gov/glossary/term/risk_tolerance (read 19 Nov 2026) - NIST CSRC glossary, compensating controls: https://csrc.nist.gov/glossary/term/compensating_controls (read 19 Nov 2026) - NIST CSRC glossary, plan of action and milestones: https://csrc.nist.gov/glossary/term/plan_of_action_and_milestones (read 19 Nov 2026) - NIST CSRC glossary, authorizing official: https://csrc.nist.gov/glossary/term/authorizing_official (read 19 Nov 2026) - NIST SP 800-37 Rev. 2: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-37r2.pdf (read 19 Nov 2026) - NIST SP 800-39: https://nvlpubs.nist.gov/nistpubs/Legacy/SP/nistspecialpublication800-39.pdf (read 19 Nov 2026) - NIST SP 800-40 Rev. 4: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf (read 19 Nov 2026) - CISA BOD 26-04: https://www.cisa.gov/news-events/directives/bod-26-04-prioritizing-security-updates-based-risk (read 19 Nov 2026) - CISA BOD 26-04 implementation guidance: https://www.cisa.gov/news-events/directives/bod-26-04-implementation-guidance-prioritizing-security-updates-based-risk (read 19 Nov 2026) - CISA BOD 26-02: https://www.cisa.gov/news-events/directives/bod-26-02-mitigating-risk-end-support-edge-devices (read 19 Nov 2026) - PCI SSC FAQ 1046: https://www.pcisecuritystandards.org/faqs/1046 (read 19 Nov 2026) - PCI SSC FAQ 1597: https://www.pcisecuritystandards.org/faqs/1597 (read 19 Nov 2026) - OWASP Virtual Patching Cheat Sheet: https://cheatsheetseries.owasp.org/cheatsheets/Virtual_Patching_Cheat_Sheet.html (read 19 Nov 2026) - Microsoft, Security Bulletin MS10-046 (2010): https://learn.microsoft.com/en-us/security-updates/securitybulletins/2010/ms10-046 (read 19 Nov 2026) - CISA KEV catalog JSON feed, version 2026.11.18: https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) - PCI SSC FAQ 1572: https://www.pcisecuritystandards.org/faqs/1572 (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Risk Acceptance Form: Free Template With 15 Fields](https://remediationvsmitigation.com/risk-acceptance-form/index.md) - [Exception Register: 15 Columns, States and Blank CSV](https://remediationvsmitigation.com/exception-register/index.md) - [Risk Acceptance vs Risk Exception in Cyber Security](https://remediationvsmitigation.com/guides/risk-acceptance-vs-risk-exception/index.md) - [Risk Acceptance vs Risk Tolerance: Who Sets the Limit](https://remediationvsmitigation.com/guides/risk-acceptance-vs-risk-tolerance/index.md) - [Residual Risk After Mitigation: Define, Estimate, Record](https://remediationvsmitigation.com/guides/residual-risk-after-mitigation/index.md) - [Inherent vs Residual Risk on a Security Decision Record](https://remediationvsmitigation.com/guides/inherent-vs-residual-risk/index.md) - [Who Is the Risk Owner for a Security Finding?](https://remediationvsmitigation.com/guides/risk-owner/index.md) - [POA&M Example: A Filled Plan of Action and Milestones Entry](https://remediationvsmitigation.com/guides/poam-plan-of-action-and-milestones/index.md) - [Vulnerability False Positive: Close It With Evidence](https://remediationvsmitigation.com/guides/false-positive-handling/index.md) --- # End of life software risk acceptance Canonical: https://remediationvsmitigation.com/guides/accepting-risk-on-end-of-life-software/ Accept end-of-life software only for a stated period, with isolation or paid support in place and a trigger that reopens the record. NIST SP 800-53 SA-22 asks for "documented approvals (including justification)". In CISA's feed read on 19 Nov 2026, 83 of 1,786 required actions use end-of-life wording. ## Where the decision row stops and this guide starts The [end-of-life row of the decision table](https://remediationvsmitigation.com/decision-table/end-of-life-software/) gives the answer (replace, remove, or isolate) and the federal decommissioning dates. This guide covers the case where replacement is blocked and a signed exception is needed: what NIST allows, the steps and records, one case worked with dates, and what CISA's feed shows. ## What SA-22 allows SA-22 gives two options. The organization must "Replace system components when support for the components is no longer available from the developer, vendor, or manufacturer", or "Provide the following options for alternative sources for continued support for unsupported components". Its guidance names the two cases where replacement may wait: "systems that provide critical mission or business capabilities where newer technologies are not available or where the systems are so isolated that installing replacement components is not an option." Alternative support can be "customized patches for critical software components" built in-house or "external providers who provide ongoing support" under contract. The added risk "can be mitigated, for example, by prohibiting the connection of such components to public or uncontrolled networks, or implementing other forms of isolation." The assessment objects to examine include "documented approvals (including justification) for the continued use of unsupported system components". An exception is therefore a record of three things: why replacement is blocked, who provides support or what isolates the component, and who approved. The PCI SSC says the same about unsupported operating systems in its FAQ 1130 (June 2013): compensating controls "must protect the system from vulnerabilities that may lead to exploit of the unsupported code", and "should be considered a temporary solution only" with "an active migration plan". It names three controls that may be combined: active monitoring of system logs and network traffic, application allowlisting, and isolating the system, and adds that "these examples alone would not provide sufficient mitigation." ## Six steps, who does them and what they leave Steps are this page's synthesis of SA-22, PCI SSC FAQ 1130 and NIST SP 800-40 Rev. 4 Section 3.5.5 (exceptions are tracked closely and reviewed regularly). | Step | Who | Record left | How it usually fails | |---|---|---|---| | 1. Confirm the end-of-support date from the vendor's own lifecycle page | Product owner | Product, version, vendor date, page address and date read | The date is copied from a scanner label, which can be wrong or late | | 2. Write why replacement is blocked and what it needs | System owner | Reason, replacement project, cost and target date | "No budget" is written with no date, so the exception has no end | | 3. Choose the route: paid support, in-house patches, or isolation | Security lead | Route, the controls, how each was verified | Isolation is claimed in a diagram and never tested | | 4. Set the end date and the reopen triggers | Risk owner | Expiry date and trigger list on the [risk acceptance form](https://remediationvsmitigation.com/risk-acceptance-form/) | The expiry equals "when the project finishes", with no date | | 5. Check the KEV catalog and vendor advisories each month | Vulnerability management | Date of each check and the catalog version read | Checks name the product but not its bundled components | | 6. Close, renew or escalate on the expiry date | Approver | New signature, or closure with the decommission record | Renewals repeat the same text with a new date and no new evidence | ## A worked case: Node.js 20.x, 30 Apr to 10 Nov 2026 The organization and its exception are invented. The end-of-life date, the CVE and the KEV dates are real. An internal service decompresses data with Node.js 20.x, which the Node.js release schedule ends on 30 Apr 2026. A supported Node.js line exists, so neither SA-22 ground holds and the exception below is out of policy, as the [end-of-life row](https://remediationvsmitigation.com/decision-table/end-of-life-software/) says of this service. The case shows what the reopen triggers do to such an exception; use SA-22 only where replacement is truly blocked. Sources: Node.js Release schedule.json, Node.js October 2026 security releases, CISA KEV feed 2026.11.18, all read 19 Nov 2026. | Date | Event | Record | |---|---|---| | 30 Apr 2026 | Node.js 20 reaches end of life (Node.js Release schedule, "end": 2026-04-30) | Step 1 date confirmed | | 12 May 2026 | Exception signed, 12 days after end of life. Route: isolation from the internet plus allowlisted callers. Expiry 30 Sep 2026, 141 days later. Reopen triggers: any KEV listing for Node.js, or any Node.js advisory naming 20.x | Form with expiry and triggers | | 28 Sep 2026 | Renewal due since 31 Aug. The isolation is retested on 25 Sep, and the approver signs a new exception 2 days before the old one expires on 30 Sep. The old row is closed "renewed as" the new one. New expiry 31 Dec 2026, same triggers | Retest record, new form, old row closed | | 21 Oct 2026 | Node.js ships 22.23.4, 24.22.1 and 26.12.1 for CVE-2026-111497. The advisory says "End-of-Life versions are always affected when a security release occurs." Node.js 20.x gets no fix | Trigger 2 fires: the renewed exception is reopened | | 27 Oct 2026 | CISA adds CVE-2026-111497 to KEV (CVSS 3.1 base score 9.8) with a due date of 10 Nov 2026, 14 days later | Trigger 1 fires | | Before 10 Nov 2026 | Options recorded: move the service to Node.js 22.23.4 (the real fix), or as a stopgap reject inbound requests with Content-Encoding br at the reverse proxy, which the researchers at Aikido and ELLIO suggest and which does not cover Brotli responses the service fetches from others | Decision with the stopgap's cost written beside it | The case shows why the end date needs a trigger and not only a calendar. The renewed exception ran to 31 Dec, yet a vendor release on 21 Oct made 20.x vulnerable with no fix possible, and 6 days later CISA's catalog set a 14-day clock. Had nobody signed by 30 Sep, the row would have moved to lapsed in the [exception register](https://remediationvsmitigation.com/exception-register/) and gone to the escalation owner, and the service would have been out of policy from 1 Oct. Isolation from the internet is not enough on its own: the flaw needs attacker-influenced compressed input, which an internal caller or an upstream response can also supply. Because no patch can exist for an end-of-life line, the only closing evidence is the move to a supported line and a version check after it. ## What federal policy adds CISA's BOD 26-02 quotes OMB Circular A-130: "unsupported information systems and system components are phased out as rapidly as possible, and planning and budgeting activities for all IT systems and services incorporate migration planning and resourcing to accomplish this requirement." Its dated steps cover edge devices only, and the directive "does not apply to Operational Technology" (the dates are in the [end-of-life row](https://remediationvsmitigation.com/decision-table/end-of-life-software/)). For OT, NIST SP 800-82 Rev. 3 says "many legacy systems may contain hardware and software that are no longer supported by vendors and cannot be patched or updated to protect against new vulnerabilities", and asks for "a complete and convincing rationale for how the selected compensating controls provide an equivalent security capability or level of protection". The OT row is [OT and medical devices that cannot be patched](https://remediationvsmitigation.com/decision-table/ot-medical-cannot-patch/). ## What the KEV feed shows about end-of-life products Counted by script on 19 Nov 2026 from the CISA KEV JSON feed, catalogVersion 2026.11.18: 83 of 1,786 required actions contain the words end-of-life (any case, with or without hyphens). They cover 19 vendor names, 36 product names and 37 vendor and product pairs; 17 are marked as used in ransomware campaigns. Of the 83 texts, 52 read "The impacted product is end-of-life and should be disconnected if still in use." Four others hedge: "The impacted product could be end-of-life (EoL) and/or end-of-service (EoS)." Entries by dateAdded year. The CVE years on the 83 entries run from 2009 to 2025, so an old flaw can be listed years after its CVE was assigned. | Year added to KEV | Entries with end-of-life wording | |---|---| | 2021 | 1 | | 2022 | 54 | | 2023 | 4 | | 2024 | 18 | | 2025 | 6 | | 2026 | 0 | Vendor names as written in the feed, in alphabetical order. Counts are entries, not products or customers. | Vendor in the feed | Entries | Products named | |---|---|---| | Adobe | 36 | Acrobat and Reader, Flash Player; Flash Player; Flash Player and AIR | | Checkbox | 1 | Checkbox Survey | | D-Link | 14 | DCS-930L Devices; DIR-300 Router; DIR-600 Router; DIR-605 Router; DIR-610 Devices; DIR-645 Router; DIR-820 Router; DIR-820L; DNS-320 Storage Device; Multiple NAS Devices; Multiple Routers | | D-Link and TRENDnet | 1 | Multiple Devices | | Dasan | 2 | Gigabit Passive Optical Network (GPON) Routers | | Delta Electronics | 1 | DOPSoft 2 | | IBM | 1 | InfoSphere BigInsights | | Ivanti | 4 | Cloud Services Appliance; Cloud Services Appliance (CSA) | | Kaseya | 1 | Virtual System/Server Administrator (VSA) | | Linux | 2 | Kernel | | Microsoft | 8 | Internet Explorer; SharePoint; Silverlight | | NETGEAR | 2 | DGN2200 Devices; Multiple Devices | | NUUO | 2 | NVRmini Devices; NVRmini2 Devices | | Reolink | 2 | Multiple IP Cameras; RLC-410W IP Camera | | Ruckus Wireless | 1 | Multiple Products | | Schneider Electric | 1 | U.motion Builder | | SonicWall | 1 | Secure Remote Access (SRA) | | Sophos | 1 | CyberoamOS | | Zyxel | 2 | DSL CPE Devices | Repeat listings arrive in clusters. Ivanti Cloud Services Appliance 4.6.x was added on 13 Sep 2024 (CVE-2024-8190), 19 Sep 2024 (CVE-2024-8963) and twice on 9 Oct 2024 (CVE-2024-9379 and CVE-2024-9380): four entries in 26 days. Its required action said "future vulnerabilities on the 4.6.x version of CSA are unlikely to receive security updates." An acceptance that rests on the product never having been attacked has no support in this data; the trigger in step 4 should be a listing for the product or for the vendor line. ## Limits to write into the exception - An end date tied to the replacement or decommission date ([decommission scheduled](https://remediationvsmitigation.com/decision-table/decommission-scheduled/)). - The isolation in place and the test that confirmed it, with its date. - A trigger that reopens the record: a KEV listing for the product, or a vendor release that names the end-of-life line. - For federal edge devices in scope of BOD 26-02, no exception path is stated; CISA's BOD 26-04 implementation guidance says CISA "does not issue waivers or exceptions for actions required in cyber directives". ## Sources - NIST SP 800-53 Rev. 5.2.0, control SA-22 (NIST OSCAL catalog): https://raw.githubusercontent.com/usnistgov/oscal-content/main/nist.gov/SP800-53/rev5/json/NIST_SP-800-53_rev5_catalog.json (read 19 Nov 2026) - PCI SSC FAQ 1130, unsupported operating systems (June 2013): https://www.pcisecuritystandards.org/faqs/1130 (read 19 Nov 2026) - CISA, BOD 26-02: https://www.cisa.gov/news-events/directives/bod-26-02-mitigating-risk-end-support-edge-devices (read 19 Nov 2026) - NIST SP 800-82 Rev. 3, legacy systems and compensating controls: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-82r3.pdf (read 19 Nov 2026) - NIST SP 800-40 Rev. 4, Section 3.5.5: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf (read 19 Nov 2026) - CISA, KEV JSON feed, catalogVersion 2026.11.18: https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) - CISA, BOD 26-04 implementation guidance: https://www.cisa.gov/news-events/directives/bod-26-04-implementation-guidance-prioritizing-security-updates-based-risk (read 19 Nov 2026) - Node.js Release working group, schedule.json (v20 end 2026-04-30): https://raw.githubusercontent.com/nodejs/Release/main/schedule.json (read 19 Nov 2026) - Node.js, October 2026 security releases: https://nodejs.org/en/blog/vulnerability/october-2026-security-releases (read 19 Nov 2026) - NVD, CVE-2026-111497: https://services.nvd.nist.gov/rest/json/cves/2.0?cveId=CVE-2026-111497 (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [End-of-Life Software: Replace, Isolate or Accept With Dates](https://remediationvsmitigation.com/decision-table/end-of-life-software/index.md) - [Decommission Date Set: An Exception That Ends on the Date](https://remediationvsmitigation.com/decision-table/decommission-scheduled/index.md) - [Risk Acceptance Form: Free Template With 15 Fields](https://remediationvsmitigation.com/risk-acceptance-form/index.md) - [OT and Medical Devices You Cannot Patch: Mitigate, Review](https://remediationvsmitigation.com/decision-table/ot-medical-cannot-patch/index.md) - [Exception Register: 15 Columns, States and Blank CSV](https://remediationvsmitigation.com/exception-register/index.md) - [KEV Required Action: Mitigate or Discontinue Use](https://remediationvsmitigation.com/guides/kev-required-actions-mitigate-or-discontinue/index.md) --- # Compensating controls in audit: what an auditor checks Canonical: https://remediationvsmitigation.com/guides/compensating-controls-in-audit/ An assessor decides whether a compensating control meets the intent of the requirement; PCI SSC says it cannot approve one itself (FAQ 1046). FAQ 1572, June 2025, adds that a missed periodic action cannot be repaired afterward, and lists four conditions for an In Place finding. Read 19 Nov 2026. ## Who decides PCI SSC states: "QSAs are responsible to determine whether a compensating control is sufficient to meet the intent of a requirement during their review of all other controls in place to satisfy PCI DSS requirements." (FAQ 1046, April 2012). The Council "is not able to approve specific configurations or compensating controls" because it is not onsite. The evidence has to persuade one assessor looking at the whole environment, not match a form. Under NIST SP 800-53 the decision to accept residual risk sits with the authorizing official. Control CA-6 describes authorizations as management decisions that "explicitly accept the risk to organizational operations and assets" based on "the implementation of agreed-upon controls", and says nonfederal organizations may have "similar processes". Whoever plays that role in your organization signs the [exception](https://remediationvsmitigation.com/exception-register/) behind the control. ## What the control itself has to show PCI SSC FAQ 1093 (June 2025) sends readers to Appendices B and C of PCI DSS v4.0.1 for compensating controls. The assessor evaluates each control in every annual assessment, and processes must keep it effective after the assessment. Four PCI SSC FAQs then state what an assessor looks for in the record: PCI SSC FAQs read 19 Nov 2026. | FAQ | What it says | What to put in the record | |---|---|---| | 1093 (June 2025, mainframes and Requirement 3.5.1) | "Compensating controls must address the additional risk introduced by not meeting the original requirement." | The risk the missing requirement leaves, stated, and the control matched to it | | 1466 (May 2019) | Document the approach and "demonstrate how the risk is mitigated and how the intent of the requirement is met"; recommendations are not to be viewed "in isolation" | The intent of the requirement in your words, and the effect of the change on neighboring settings | | 1130 (June 2013) | For unsupported systems, controls "must protect the system from vulnerabilities that may lead to exploit"; log monitoring, allowlisting and isolation "alone would not provide sufficient mitigation"; it is "a temporary solution only" | A set of controls, a migration plan and its date | | 1597 (May 2025) | "Addressed" means the entity "determines whether to resolve the vulnerability or to mitigate the risk by addressing the vulnerability in another way (e.g., with a compensating control or by disabling a vulnerable service)" | The risk level the entity assigned under 6.3.1 and the targeted risk analysis behind the timeframe | ## A missed deadline cannot be repaired afterward FAQ 1572 (June 2025) asks whether a compensating control can cover a requirement with a periodic frequency that was missed. It cannot: "there is no way for them to perform the control retroactively or backdate a later occurrence of the control to an earlier period." Its examples include a missed external ASV scan, "not installing a critical security patch within 30 days of release" and not reviewing network security control configurations every six months. An assessor can still mark the requirement In Place if the entity has implemented corrective actions and performed the control as required, and the assessor has assurance of four things: a repeatable and documented process; that the activity was missed due to an exceptional circumstance ("poor security practices and recurring failures are not 'exceptional circumstances'"); that the cause was addressed; and that steps now prevent recurrence. Otherwise the assessor "can consider whether a 'Not in Place' finding would be the appropriate result". ## A patch deadline worked with dates The payment API below belongs to an invented entity. Dates are real, from the Node.js October 2026 security releases and the CISA KEV feed (2026.11.18). CVE-2026-111497 is an out-of-bounds write in the Brotli decoder bundled with Node.js, reachable when an application decompresses attacker-supplied data with node:zlib. Node.js fixed it in 22.23.4, 24.22.1 and 26.12.1 on 21 Oct 2026, and says end-of-life lines are always affected by a security release. NVD lists as affected every Node.js version below 22.23.4, from 23.0.0 below 24.22.1, and from 25.0.0 below 26.12.1. FAQ 1597 says PCI DSS Requirement 6.3.3 requires patches for critical vulnerabilities to be "resolved within one month of release", and that the entity assigns the risk level under 6.3.1 rather than accepting an outside rating. FAQ 1572 words the same limit as "30 days of release". From 21 Oct, 30 days ends on 20 Nov and one month on 21 Nov, so the entity plans to 20 Nov, the earlier date. Sources: Node.js October 2026 security releases, CISA KEV feed 2026.11.18, PCI SSC FAQ 1597 and FAQ 1572, read 19 Nov 2026. | Date | Event | What the assessor can later see | |---|---|---| | 21 Oct 2026 | Node.js releases 22.23.4, 24.22.1 and 26.12.1 for CVE-2026-111497 (CVSS 3.1 base score 9.8 in NVD) | Release date: the one-month clock starts | | 27 Oct 2026 | CISA adds the CVE to KEV; the entity rates it critical under 6.3.1 and records why | The rating and its reason, dated | | 10 Nov 2026 | KEV due date, 14 days after listing | Whether the entity treated it as an internal target | | 20 Nov 2026 | 30 days from release (FAQ 1572's wording) ends here, the planning deadline under 6.3.3. One month under FAQ 1597 ends a day later, on 21 Nov | Patch evidence dated on or before 20 Nov meets both readings | If the service is on 22.x, 24.x or 26.x and moved to the matching fixed build (22.23.4, 24.22.1 or 26.12.1) by 20 Nov, the patch record is the evidence and no compensating control is needed. If the patch lands on 24 Nov, FAQ 1572 applies: the 20 Nov deadline was missed, and a control put in place now cannot be backdated. The entity needs the four assurances above, and the packet below shows the corrective actions. If the service runs Node.js 20.x, an end-of-life line, the October releases do not cover it; that is the unsupported-software case in FAQ 1130, and the control must be in place before the deadline with a migration plan attached. The step-by-step version is in [accepting risk on end-of-life software](https://remediationvsmitigation.com/guides/accepting-risk-on-end-of-life-software/). ## Federal justification text FedRAMP's legacy playbook, reference only since 24 Jun 2026, set the bar for a risk adjustment: "To justify a RA, CSPs must describe mitigating factors or compensating controls in place that reduce likelihood and/or impact of exploitation." It also required that high-risk vendor dependencies be "mitigated to a Moderate level through compensating controls within thirty (30) days." The 2026 rules replace that scheme with a 192-day limit: a vulnerability not fully mitigated or remediated within 192 days of evaluation is categorized as accepted (VER-TFR-MAV). A compensating control buys time on a clock that ends, which is the subject of [maximum duration of a risk acceptance](https://remediationvsmitigation.com/risk-acceptance-form/maximum-durations/). ## Evidence packet checklist Compensating control evidence packet (one per finding) 1. Finding: identifier, asset, date found. 2. Requirement or timeline that cannot be met as written, with the clause number and the date it fell due. 3. Why it cannot be met (vendor patch not available, outage window, end of life) and the risk the gap leaves. 4. The intent of the requirement, in your words. 5. Control used, from the compensating controls register, and what it changes. 6. What the control does not fix, stated in the entry. 7. Verification: test record with date and tester; configuration or rule export with date. 8. Evidence the control continues: monitoring, drift alert, review dates. 9. Approver (name and role) and date of approval. 10. End date or review date, the migration or fix plan, and the action that closes the finding. 11. If a periodic action was missed: the four FAQ 1572 assurances, each with its document. 12. Link to the dated exception in the exception register. The packet below filled in for the Node.js 20.x service from the case above. The entity, finding number, dates after 21 Oct 2026 and the migration date are invented to show the layout; the vulnerability, the fixed lines and the stopgap are real. Invented entity and dates after 21 Oct 2026; vulnerability facts from the Node.js October 2026 security releases, Aikido and ELLIO, read 19 Nov 2026. | Item | Entry for the Node.js 20.x payment API | |---|---| | 1 Finding | F-2041, payment API, CVE-2026-111497, found 27 Oct 2026 when CISA added it to KEV | | 2 Requirement | 6.3.3, critical patch within one month of release (21 Oct 2026), so due 20 Nov 2026 on the planning date used above | | 3 Why it cannot be met | Node.js 20.x is end of life and the October releases cover only 22.x, 24.x and 26.x; the gap is that the Brotli decoder in the service stays unpatched | | 4 Intent | Keep attacker-supplied compressed data from reaching the vulnerable decoder | | 5 Control | Gateway rule rejecting inbound requests with Content-Encoding br, the stopgap described by Aikido and ELLIO; it changes likelihood on request bodies only | | 6 Not fixed | Brotli responses the service fetches from other servers; the service still runs the vulnerable code | | 7 Verification | Dated test by a person other than the rule author: a request with Content-Encoding br to the production path is refused; gateway rule export saved | | 8 Continues | Weekly count of rejected requests; rule export compared at each monthly review; alert if the rule is disabled | | 9 Approver | Head of Payments Engineering, approved 12 Nov 2026 | | 10 End date | Move to 22.23.4 by 15 Jan 2027, then remove the rule; plan in the POA&M | | 11 Missed periodic action or deadline | Not a FAQ 1572 case while the control is approved (12 Nov) before the 20 Nov date. If 20 Nov passes with the patch not installed, 6.3.3 rests on the control alone: FAQ 1597 says a critical flaw is "resolved" when fixed, which a gateway rule is not, so the assessor decides under FAQ 1046 whether the control meets the intent | | 12 Exception | EX-2041 in the exception register, expiring 15 Jan 2027 | The 15 Jan 2027 end date in item 10 is 56 days after the 20 Nov date in item 2, so the control stands in for the patch for that whole period. FAQ 1130 words the limit for unsupported systems: the controls are "a temporary solution only", which is why item 10 carries the migration date and the exception expires on it. The system owner assembles the packet, with the owner of the control supplying items 5 to 8 and the approver in item 9 signing; the assessor reviews it and does not write it. Items 7 and 8 are the two NIST SP 800-40 Rev. 4 checks: the control is "functioning as intended" and the risk response "continues to be in place", which means no one "deactivates the additional security controls". A reviewer can only repeat those checks if the record names exactly what to repeat. For item 7, write the test so a second person can run it and get the same result: the path, the input, the expected rejection and where it is logged. For item 8, attach dated output from the monitoring (for example, a count of rejected requests per week) and the rule export taken at each review, so a sample of dates can be compared with the original setting. The compensating controls register holds the control entries, [how to verify a mitigation](https://remediationvsmitigation.com/guides/verifying-a-mitigation-works/) the test, and [the POA&M guide](https://remediationvsmitigation.com/guides/poam-plan-of-action-and-milestones/) the plan that tracks the fix. ## Sources - PCI SSC FAQ 1046 (April 2012), approval of compensating controls: https://www.pcisecuritystandards.org/faqs/1046 (read 19 Nov 2026) - PCI SSC FAQ 1093 (June 2025), mainframes storing cardholder data (Requirement 3.5.1); the compensating-control sentence refers to "this or any other requirement" and points to PCI DSS Appendices B and C: https://www.pcisecuritystandards.org/faqs/1093 (read 19 Nov 2026) - PCI SSC FAQ 1130 (June 2013), unsupported operating systems: https://www.pcisecuritystandards.org/faqs/1130 (read 19 Nov 2026) - PCI SSC FAQ 1466 (May 2019), alternative password methods as compensating controls: https://www.pcisecuritystandards.org/faqs/1466 (read 19 Nov 2026) - PCI SSC FAQ 1572 (June 2025), compensating controls and missed periodic requirements: https://www.pcisecuritystandards.org/faqs/1572 (read 19 Nov 2026) - PCI SSC FAQ 1597 (May 2025), assigning risk levels, resolving or addressing vulnerabilities: https://www.pcisecuritystandards.org/faqs/1597 (read 19 Nov 2026) - NIST SP 800-53 Rev. 5.2.0, control CA-6 (NIST OSCAL catalog): https://raw.githubusercontent.com/usnistgov/oscal-content/main/nist.gov/SP800-53/rev5/json/NIST_SP-800-53_rev5_catalog.json (read 19 Nov 2026) - NIST SP 800-40 Rev. 4, Guide to Enterprise Patch Management Planning, Section 2.2: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf (read 19 Nov 2026) - FedRAMP legacy playbook, POA&M, General POA&M Guidance: https://www.fedramp.gov/legacy/playbook/csp/authorization/poam/ (read 19 Nov 2026) - FedRAMP 2026 Vulnerability Evaluation and Reporting, VER-TFR-MAV: https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-evaluation-and-reporting/ (read 19 Nov 2026) - Node.js, October 2026 security releases: https://nodejs.org/en/blog/vulnerability/october-2026-security-releases (read 19 Nov 2026) - CISA, KEV JSON feed, catalogVersion 2026.11.18: https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) - NVD record for CVE-2026-111497 (affected Node.js ranges, CVSS 3.1 base score): https://services.nvd.nist.gov/rest/json/cves/2.0?cveId=CVE-2026-111497 (read 19 Nov 2026) - Aikido analysis of CVE-2026-111497: https://www.aikido.dev/blog/node2shell-nodejs-brotli-rce-cve-2026-111497 (read 19 Nov 2026) - ELLIO analysis of CVE-2026-111497: https://ellio.tech/en/blog/node2shell-cve-2026-111497-rce-on-stock-nodejs-builds/ (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Exception Register: 15 Columns, States and Blank CSV](https://remediationvsmitigation.com/exception-register/index.md) - [How to Verify a Mitigation Works: Checks for 5 Real CVEs](https://remediationvsmitigation.com/guides/verifying-a-mitigation-works/index.md) - [POA&M Example: A Filled Plan of Action and Milestones Entry](https://remediationvsmitigation.com/guides/poam-plan-of-action-and-milestones/index.md) - [Accepting Risk on End-of-Life Software: Limits and Record](https://remediationvsmitigation.com/guides/accepting-risk-on-end-of-life-software/index.md) --- # Compensating controls vs mitigating controls Canonical: https://remediationvsmitigation.com/guides/compensating-controls-vs-mitigating-controls/ A mitigating control lowers the risk of a vulnerability; a compensating control meets a stated requirement another way. NIST defines them separately, and PCI SSC lists a compensating control as one way to address a finding. A worked PCI DSS example below uses CVE-2026-0322 and a 13 Nov 2026 planning date. Read 19 Nov 2026. ## Two definitions, two jobs NIST SP 800-160 Vol. 2 Rev. 1 defines mitigation as "A decision, action, or practice intended to reduce the level of risk associated with one or more threat events, threat scenarios, or vulnerabilities." The object is a risk or a vulnerability. SP 800-53 Rev. 5 defines compensating controls as the controls "employed in lieu of the controls in the baselines described in NIST Special Publication 800-53B that provide equivalent or comparable protection for a system or organization" ([NIST CSRC glossary](https://csrc.nist.gov/glossary/term/compensating_controls)). The object is a required control. A mitigating control answers "what lowers this risk?" A compensating control answers "what meets this requirement another way?" The same action can be both. Limiting a firewall's management interface to one jump box lowers the risk of a flaw in that interface (mitigating) and, if a standard requires the patch within a stated window, may also be offered as the substitute for the patch (compensating). The label depends on which question the record answers. The glossary carries a second mitigation definition, from SP 800-216: "The temporary reduction or lessening of the impact of a vulnerability or the likelihood of its exploitation." This page uses the SP 800-160 wording because the comparison is about risk and requirements. The [difference between remediation and mitigation](https://remediationvsmitigation.com/guides/difference-between-remediation-and-mitigation/) uses the SP 800-216 wording because that question is about whether the flaw is still there. ## How standards treat a compensating control Wording from the NIST CSRC glossary, PCI SSC FAQs 1597 and 1046 and the FedRAMP legacy POA&M playbook, read 19 Nov 2026. | Publisher | Wording read on 19 Nov 2026 | What it shows | |---|---|---| | NIST CSRC glossary (SP 800-53 Rev. 5) | Controls "employed in lieu of the controls in the baselines ... that provide equivalent or comparable protection" | Tied to a baseline requirement | | PCI SSC FAQ 1597 (May 2025) | "Addressed - the entity determines whether to resolve the vulnerability or to mitigate the risk by addressing the vulnerability in another way (e.g., with a compensating control or by disabling a vulnerable service)." | A compensating control is one way to address a finding that is not resolved | | PCI SSC FAQ 1046 (April 2012) | "QSAs are responsible to determine whether a compensating control is sufficient to meet the intent of a requirement" | The assessor decides, not the Council | | FedRAMP legacy playbook (reference only after 24 Jun 2026) | "High-risk VDs must be mitigated to a Moderate level through compensating controls within thirty (30) days." | Compensating controls used as the means of mitigation | PCI SSC and the FedRAMP playbook use a compensating control as a means of mitigation, while NIST ties the term to a baseline requirement. Quote any one of them with its publisher named. How FedRAMP's 2026 rules and CISA word the remediation side is in the [difference guide](https://remediationvsmitigation.com/guides/difference-between-remediation-and-mitigation/). ## Worked example: one flaw, two record lines An invented merchant runs a PAN-OS firewall in its card data environment. Palo Alto Networks published fixed builds for CVE-2026-0322 on 14 Oct 2026 ([advisory](https://security.paloaltonetworks.com/CVE-2026-0322)), and CISA added the flaw to its Known Exploited Vulnerabilities catalog the same day with a due date of 17 Oct 2026 ([KEV feed, version 2026.11.18](https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json)). The merchant rates the flaw critical for its own environment, which PCI DSS Requirement 6.3.1 lets it do: FAQ 1597 says the requirement does not oblige an entity to accept a rating assigned by an outside source. Requirement 6.3.3 gives critical patches one month from release (FAQ 1597), which ends on 14 Nov 2026. FAQ 1572 words the same requirement as "30 days of release"; 30 days from 14 Oct is 13 Nov 2026. The merchant plans to the earlier date, 13 Nov 2026, and uses it below. A change freeze blocks installs until 30 Nov. On 15 Oct the network team limits the management interface to a single jump box, the step Palo Alto describes: the vendor rates the flaw CVSS 4.0 8.5 with the jump box against 8.6 without it. **Line 1, mitigation (answers: what lowers this risk?)** - Finding: CVE-2026-0322 on fw-cde-01 (sample name). - Control: management interface reachable only from jump box JB-02, in force from 15 Oct 2026. - What it changes: the network path to the interface. What stays: the flaw, and any account that can reach JB-02. - Vendor rating: CVSS 4.0 8.5 with the jump box, 8.6 without. - Ends: when a fixed build for the firewall's release train is installed and `show system info` shows it. **Line 2, compensating (answers: what meets the requirement another way?)** - Requirement: PCI DSS 6.3.3, critical patch within one month of release (14 Nov 2026; 30 days is 13 Nov). Planning date 13 Nov 2026. - Why it cannot be met as written: change freeze until 30 Nov 2026. - Substitute: the controls offered, and the reasoning that they give equivalent or comparable protection. - Accepted by: the QSA, name and date, before 13 Nov 2026. - Ends: the day the fixed build is installed. The jump box alone is a thin basis for line 2. The vendor's rating moves by 0.1, and the flaw stays reachable by anyone who holds a session on the jump box. The QSA decides whether a control meets the intent of the requirement (FAQ 1046), so line 2 needs more than line 1, for example controls that detect use of the interface. The date matters as much as the content. FAQ 1572 names "not installing a critical security patch within 30 days of release" among the periodic activities that cannot be performed retroactively, and says an assessor may still mark the requirement In Place only if the entity has corrected course and the assessor has assurance of four things, including that the miss was an "exceptional circumstance". The substitute is therefore agreed before 13 Nov, not written after it. ## A control does not close the finding FedRAMP's 2026 rules say providers "SHOULD remediate Known Exploited Vulnerabilities according to the due dates in the CISA Known Exploited Vulnerabilities Catalog (even if the vulnerability has been fully mitigated)" ([VDR-TFR-KEV](https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-detection-and-response/)). Neither kind of control closes a finding on its own. The finding closes when the flaw is removed, or when a dated acceptance is recorded in the [exception register](https://remediationvsmitigation.com/exception-register/). ## Sources - NIST CSRC glossary, mitigation (SP 800-160 Vol. 2 Rev. 1; SP 800-216): https://csrc.nist.gov/glossary/term/mitigation (read 19 Nov 2026) - NIST CSRC glossary, compensating controls (SP 800-53 Rev. 5): https://csrc.nist.gov/glossary/term/compensating_controls (read 19 Nov 2026) - PCI SSC FAQ 1597 (May 2025): https://www.pcisecuritystandards.org/faqs/1597 (read 19 Nov 2026) - PCI SSC FAQ 1572 (June 2025): https://www.pcisecuritystandards.org/faqs/1572 (read 19 Nov 2026) - PCI SSC FAQ 1046 (April 2012): https://www.pcisecuritystandards.org/faqs/1046 (read 19 Nov 2026) - FedRAMP 2026 Vulnerability Detection and Response rules, VDR-TFR-KEV: https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-detection-and-response/ (read 19 Nov 2026) - FedRAMP legacy playbook, POA&M: https://www.fedramp.gov/legacy/playbook/csp/authorization/poam/ (read 19 Nov 2026) - Palo Alto Networks advisory CVE-2026-0322: https://security.paloaltonetworks.com/CVE-2026-0322 (read 19 Nov 2026) - CISA KEV feed, catalog version 2026.11.18: https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Difference Between Remediation and Mitigation in IT Security](https://remediationvsmitigation.com/guides/difference-between-remediation-and-mitigation/index.md) - [Compensating Controls in Audit: What Assessors Check](https://remediationvsmitigation.com/guides/compensating-controls-in-audit/index.md) - [Exception Register: 15 Columns, States and Blank CSV](https://remediationvsmitigation.com/exception-register/index.md) - [Workaround vs Mitigation in Real Vendor Advisories](https://remediationvsmitigation.com/guides/workaround-vs-mitigation/index.md) --- # Difference between remediation and mitigation Canonical: https://remediationvsmitigation.com/guides/difference-between-remediation-and-mitigation/ Remediation eliminates a vulnerability; mitigation lowers the chance or impact of exploiting it while it remains. NIST SP 800-216 and CISA BOD 26-04 use the words this way, and FedRAMP's 2026 rules insist on it. NIST SP 800-40 Rev. 4 still counts patching as a form of mitigating risk. ## What each word means in the sources The NIST glossary lists several definitions for each word, taken from different publications. Two of them settle the common meaning. SP 800-216 defines remediation as "The neutralization or elimination of a vulnerability or the likelihood of its exploitation" and mitigation as "The temporary reduction or lessening of the impact of a vulnerability or the likelihood of its exploitation" ([remediation](https://csrc.nist.gov/glossary/term/remediation), [mitigation](https://csrc.nist.gov/glossary/term/mitigation)). The glossary also holds a definition that blurs the line: CNSSI 4009-2022 gives remediation as "The act of mitigating a vulnerability or a threat." A reader who finds that entry first will think the words are the same. A reader who finds SP 800-216 will think they are opposites. Both are in NIST's own glossary. Five uses of the pair. Read 19 Nov 2026. Sources: NIST CSRC glossary, BOD 26-04, FedRAMP VDR rules, PCI SSC FAQ 1597. | Source | Treats remediation as | Treats mitigation as | |---|---|---| | NIST SP 800-216 | Elimination of the vulnerability | Temporary reduction of impact or likelihood | | CNSSI 4009-2022 | Correcting known deficiencies; also "the act of mitigating" | Not defined in the CNSSI remediation entry | | CISA BOD 26-04 | Elimination through patching, decommissioning or another action | A valid action that can change the timeline, such as removing the system from the internet | | FedRAMP 2026 rules | The end state; required for KEVs by the CISA due date | Leaves the vulnerability in place "with negligible risk" | | PCI SSC FAQ 1597 | "Resolved" | Part of "addressed", for example a compensating control or disabling a vulnerable service | ## What each action changes Five actions and the word each source uses. Read 19 Nov 2026. | Action | What changes | What the sources call it | |---|---|---| | Install the vendor patch | The vulnerable code is replaced | Remediation in BOD 26-04; "Mitigate" in SP 800-40 Rev. 4 | | Decommission the asset | The asset is gone | Remediation in BOD 26-04; "Avoid" in SP 800-40 Rev. 4 | | Remove the system from the internet | The exposure value moves from Yes to No; the flaw stays | A "valid mitigation" in BOD 26-04 | | Add a WAF rule | Matching requests are blocked | OWASP: the underlying vulnerability stays | | Disable the vulnerable feature | The code path is not reachable; the software stays installed | Listed under Mitigate in SP 800-40 Rev. 4 | The test that works in every row: after the action, does a scan or test of the asset still find the vulnerability? If not, it was remediated. If it does, it was mitigated, and the record needs a control, a test and an end date. ## Worked example: CVE-2026-0322 on a firewall Palo Alto Networks published its advisory for the PAN-OS management interface flaw CVE-2026-0322 on 14 Oct 2026, with fixed builds for every affected train (for the 11.2 train, 11.2.13-h3 or 11.2.16 among them). The advisory states "No known workarounds exist for this issue" and also says "You can reduce the risk of exploitation by restricting access to a jump box that is the only system allowed to access the management interface" ([Palo Alto Networks](https://security.paloaltonetworks.com/CVE-2026-0322)). The vendor's own wording keeps the two apart: no workaround that removes the flaw, one step that lowers the risk. CISA added the CVE to the KEV catalog on 14 Oct 2026 with a due date of 17 Oct 2026, three days later ([KEV feed, version 2026.11.18](https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json)). Its required action reads "Apply mitigations in accordance with vendor instructions", although BOD 26-04 defines remediation as elimination, so a catalog entry may say mitigations where the directive's timeline means a fix. The table follows one invented firewall, fw-01, on build 11.2.13-h2. Invented firewall, real CVE and dates. Palo Alto advisory and CISA KEV entry read 19 Nov 2026. | Date | What happens | Word to record | |---|---|---| | 14 Oct 2026 | Advisory published and KEV entry added; due date 17 Oct 2026 | Detection; clock starts | | 15 Oct 2026 | Management interface limited to one jump box; vendor rating 8.5 against 8.6; a scan of fw-01 still reports the flaw | Mitigation, with the control, a test and an end date | | 16 Oct 2026 | fw-01 upgraded to 11.2.13-h3; `show system info` shows the new build; the next scan no longer reports the flaw | Remediation, 2 days after the KEV entry and 1 day before the due date | Suppose instead the upgrade waits for a change window on 24 Oct 2026. The mitigation from 15 Oct stays valid, but the remediation is 7 days past the 17 Oct due date. A FedRAMP provider is told to remediate KEVs by the CISA due date "even if the vulnerability has been fully mitigated". For a federal agency, BOD 26-04 names one case where mitigation moves the timeline: removing the system from the internet changes the exposure value. A narrower restriction such as a jump box is not that example, so record which exposure value was used and why. ## Where NIST SP 800-40 Rev. 4 differs SP 800-40 Rev. 4 names four responses to software vulnerabilities: accept, mitigate, transfer and avoid. Under Mitigate it lists "patching the vulnerable software, disabling a vulnerable feature, or upgrading to a newer software version". So in that publication patching is a way of mitigating risk, and a patch is one of "the only forms of risk response that can completely eliminate the vulnerabilities without removing functionality" ([NIST SP 800-40 Rev. 4](https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf), Section 2.1). Use the narrower meaning when a deadline is involved. FedRAMP's 2026 rules say "A fully mitigated vulnerability will still exist (with negligible risk) until it has been remediated" ([FedRAMP VDR](https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-detection-and-response/)). The [decision tool](https://remediationvsmitigation.com/) applies that meaning, and the [Decision Table](https://remediationvsmitigation.com/decision-table/) shows it by situation. See also [remediation in cybersecurity](https://remediationvsmitigation.com/guides/remediation-in-cybersecurity/). Vendors use the two words with their own senses. What is left after a mitigation is covered in [residual risk after mitigation](https://remediationvsmitigation.com/guides/residual-risk-after-mitigation/), and the line between a mitigation and a requirement substitute is on [compensating controls vs mitigating controls](https://remediationvsmitigation.com/guides/compensating-controls-vs-mitigating-controls/). ## Sources - NIST CSRC glossary, remediation: https://csrc.nist.gov/glossary/term/remediation (read 19 Nov 2026) - NIST CSRC glossary, mitigation: https://csrc.nist.gov/glossary/term/mitigation (read 19 Nov 2026) - NIST SP 800-40 Rev. 4, Section 2.1: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf (read 19 Nov 2026) - CISA, BOD 26-04, Appendix A and footnote 8: https://www.cisa.gov/news-events/directives/bod-26-04-prioritizing-security-updates-based-risk (read 19 Nov 2026) - FedRAMP, VDR rules: https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-detection-and-response/ (read 19 Nov 2026) - PCI SSC FAQ 1597: https://www.pcisecuritystandards.org/faqs/1597 (read 19 Nov 2026) - OWASP Virtual Patching Cheat Sheet: https://cheatsheetseries.owasp.org/cheatsheets/Virtual_Patching_Cheat_Sheet.html (read 19 Nov 2026) - Palo Alto Networks advisory CVE-2026-0322: https://security.paloaltonetworks.com/CVE-2026-0322 (read 19 Nov 2026) - CISA KEV feed, catalog version 2026.11.18: https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [What Is Remediation in Cyber Security? Steps and Proof](https://remediationvsmitigation.com/guides/remediation-in-cybersecurity/index.md) - [Workaround vs Mitigation in Real Vendor Advisories](https://remediationvsmitigation.com/guides/workaround-vs-mitigation/index.md) - [Decision Table: Remediate, Mitigate or Accept by Situation](https://remediationvsmitigation.com/decision-table/index.md) - [Mitigation vs Remediation Wording in Vendor Advisories](https://remediationvsmitigation.com/guides/mitigation-vs-remediation-in-vendor-advisories/index.md) - [Residual Risk After Mitigation: Define, Estimate, Record](https://remediationvsmitigation.com/guides/residual-risk-after-mitigation/index.md) - [Compensating Controls vs Mitigating Controls](https://remediationvsmitigation.com/guides/compensating-controls-vs-mitigating-controls/index.md) --- # Risk exception expiry and renewal Canonical: https://remediationvsmitigation.com/guides/exception-expiry-and-renewal/ An exception that reaches its end date has not removed the risk. NIST SP 800-37 Rev. 2 lets the authorizing official set a termination date, and at review the official either acknowledges continued acceptance or requires further response. NIST sets no renewal limit. Read 19 Nov 2026. ## What the date means in NIST SP 800-37 SP 800-37 Rev. 2 says "The authorization termination date is established by the authorizing official and indicates when the authorization expires." It also says the date "can be adjusted at any time by the authorizing official to reflect an increased level of concern regarding the security and privacy posture of the system." Organizations "may eliminate the authorization termination date if the system is operating under an ongoing authorization", where a time-driven review frequency replaces it. At that review: "If the risk remains acceptable, the authorizing official acknowledges the acceptance in accordance with organizational processes. If not, the authorizing official indicates that the risk is no longer acceptable and requires further risk response or a full denial of the authorization." The text leaves the formality to the organization: it "determines the level of formality for the process of communicating and acknowledging continued risk acceptance." A renewal is therefore a fresh acknowledgment of the same fields, not a change of a date in a spreadsheet. The fields are on the [risk acceptance form](https://remediationvsmitigation.com/risk-acceptance-form/), and who may sign is on [who may sign a risk acceptance](https://remediationvsmitigation.com/risk-acceptance-form/who-may-sign/). ## Triggers that end an exception early - Event-driven review. SP 800-37 Rev. 2 describes a review the authorizing official may require "in addition to the time-driven frequency review". - Change monitoring. SP 800-39 lists as one purpose of monitoring to "identify risk-impacting changes to organizational information systems and the environments in which the systems operate". - Changed facts. CISA's BOD 26-04 says "The timelines for each asset are dynamic as facts change", so an exception written for an internal asset lapses in substance if the asset becomes exposed or the CVE is added to the KEV catalog. - Open maintenance exceptions. SP 800-40 Rev. 4 (Section 3.5.5) asks that exceptions to maintenance plans be "reviewed regularly to determine if the maintenance plan can be implemented now." ## A worked case: CVE-2026-41709 on ESX hosts The organization is invented; the CVE, the advisory and the KEV dates are real. Broadcom published advisory VMSA-2026-0006 on 29 Jul 2026 with fixed ESX builds, and NVD recorded CVE-2026-41709 on 30 Jul as Low (CVSS 3.1 base 2.7): an administrator on the host can perform some operations that leave no log entry. Broadcom lists no workaround. On Monday 24 Aug 2026 an operations team with 40 ESX 8.0 hosts below build ESXi80U3j-25429389 signs a 90-day acceptance. The reasons: the rating is Low, the flaw needs ESX administrator rights, and the hosts are patched in the November cluster window. The controls: administrator rights limited to named accounts, management interface reachable only from a jump host, and host syslog forwarded to a collector. One reopen trigger is written on the record: the CVE appears on the KEV catalog. Day 90 is 22 Nov 2026. Vendor and CISA dates read 19 Nov 2026 (Broadcom VMSA-2026-0006, NVD, KEV feed 2026.11.18). Signing date, estate and decisions are illustrative; day counts worked out by this site. | Date | Day of term | What happens | |---|---|---| | 24 Aug 2026 | 0 | Approver signs; expiry set to 22 Nov 2026 | | 23 Sep 2026 | 30 | First review: facts unchanged, controls retested, exception continues | | 23 Oct 2026 | 60 | Second review: facts unchanged, exception continues | | 28 Oct 2026 | 65 | CISA adds the CVE to KEV with a due date of 11 Nov 2026. Its SSVC entry in NVD lists exploitation as active. The reopen trigger fires and the approver holds an event review the same day | | 28 Oct 2026 | 65 | Decision: close the exception, patch all 40 hosts to ESXi80U3j-25429389 or later by 11 Nov, and review the administrator-account list and host logs against vCenter task history | | 11 Nov 2026 | 79 | KEV due date. Closure evidence: build number per host from `esxcli system version get`. Exception closed | | 22 Nov 2026 | 90 | Original expiry. Nothing is left to lapse | The scheduled reviews were never going to catch this. The 23 Oct review came five days before the listing, and the end date of 22 Nov fell 11 days after the KEV due date. An exception that waited for its expiry would have kept hosts unpatched through a listing for known exploitation. Only the written trigger and an event review moved the decision to 28 Oct. A federal civilian agency would have had no choice at all: CISA states that it "does not issue waivers or exceptions for actions required in cyber directives". ## What the approver is given at renewal 1. List what changed since the last signature. Owner: the exception owner. Record: KEV status, vendor fix status, exposure, whether the scanner finding is still open. In the case above this list, dated 28 Oct, is what ended the exception. 2. Retest every control named in the exception. Owner: the control owner. Record: a test result with date and tester. SP 800-40 Rev. 4 asks that additional controls be confirmed "functioning as intended" and that the response "continues to be in place". This is where renewals usually fail: the control was never rechecked. 3. Confirm the risk owner. NIST IR 8286C suggests reviewing monthly that the risk owner is still the appropriate designee, so confirm the owner before the choice, not after. 4. Choose renew, close or escalate (table below). Owner: the risk owner proposes, the approver decides. Record: the choice with its reason in one sentence. 5. Obtain a new acknowledgment from the approver, with a new date and a new end date. Owner: the approver named in the original record, or that role's successor. Record: a signature and date on a new version of the record, linked to the old one. Editing the old end date leaves no evidence that anyone decided. Outcomes drawn from the NIST texts named, read 19 Nov 2026. | Outcome | What the sources require | Evidence | |---|---|---| | Renew | A fresh acknowledgment by the signer (SP 800-37 Rev. 2) | New signature and date, current residual risk | | Close | Verify the response: patch installed, controls "functioning as intended", or asset decommissioned (SP 800-40 Rev. 4) | Rescan result, change ticket or decommission record | | Escalate | Risk elevation to a more senior stakeholder (IR 8286B) | Named new owner and decision date | How the register marks a row as renewal due, lapsed or closed, and the monthly review that opens a renewal 30 days before the end date, are on the [exception register](https://remediationvsmitigation.com/exception-register/). ## When renewal is the wrong answer Renewal is wrong when the facts that justified the exception are gone. Close instead if the vendor fix exists, escalate if the CVE is now on the KEV catalog, and stop if a control failed its retest. A KEV listing also shortens the clock: BOD 26-04 starts its timeline on the earlier of the catalog addition and the agency's own identification of the flaw. Repetition is a signal too. NIST IR 8286B gives an escalation example for risk left without treatment: "a risk that has remained through more than two fiscal periods without adequate treatment might be flagged for additional scrutiny." Neither IR 8286B nor IR 8286C sets an exception lifetime. For FedRAMP providers the 2026 rules do set a ceiling of a kind. "Providers MUST categorize any vulnerability that is not or will not be fully mitigated or remediated within 192 days of evaluation as an accepted vulnerability" (VER-TFR-MAV), and an accepted vulnerability is one the provider "does not intend to fully mitigate or remediate, OR that has not or will not be fully mitigated or remediated within the maximum overdue period" ([FedRAMP definitions](https://www.fedramp.gov/2026/definitions/)). It is a reporting category, not a permission; the dates are on [maximum durations](https://remediationvsmitigation.com/risk-acceptance-form/maximum-durations/). The legacy playbook said approved operational requirements "are still considered open risks" that must be "periodically reassessed", and its deviation form listed Pending, Approved, Rejected and Withdrawn as statuses. Where the exception covers unsupported software, see [accepting risk on end-of-life software](https://remediationvsmitigation.com/guides/accepting-risk-on-end-of-life-software/). ## Sources - NIST SP 800-37 Rev. 2, Task R-4 and Appendix F: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-37r2.pdf (read 19 Nov 2026) - NIST SP 800-39, risk monitoring: https://nvlpubs.nist.gov/nistpubs/Legacy/SP/nistspecialpublication800-39.pdf (read 19 Nov 2026) - NIST SP 800-40 Rev. 4, Sections 2.2 and 3.5.5: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf (read 19 Nov 2026) - NIST IR 8286B, Section 2.4.3 (escalation and elevation): https://nvlpubs.nist.gov/nistpubs/ir/2022/NIST.IR.8286B.pdf (read 19 Nov 2026) - NIST IR 8286C (periodic review of the risk owner): https://nvlpubs.nist.gov/nistpubs/ir/2022/NIST.IR.8286C.pdf (read 19 Nov 2026) - FedRAMP legacy playbook, POA&M: https://www.fedramp.gov/legacy/playbook/csp/authorization/poam/ (read 19 Nov 2026) - FedRAMP legacy Vulnerability Deviation Request Form, status list: https://www.fedramp.gov/legacy/assets/LEGACY%20FedRAMP-Vulnerability-Deviation-Request-Form.xlsx (read 19 Nov 2026) - FedRAMP 2026 definitions, Accepted Vulnerability: https://www.fedramp.gov/2026/definitions/ (read 19 Nov 2026) - FedRAMP 2026, VER-TFR-MAV: https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-evaluation-and-reporting/ (read 19 Nov 2026) - CISA BOD 26-04, Appendix A: https://www.cisa.gov/news-events/directives/bod-26-04-prioritizing-security-updates-based-risk (read 19 Nov 2026) - Broadcom advisory VMSA-2026-0006 (CVE-2026-41709): https://support.broadcom.com/web/ecx/support-content-notification/-/external/content/SecurityAdvisories/0/38017 (read 19 Nov 2026) - NVD record, CVE-2026-41709: https://services.nvd.nist.gov/rest/json/cves/2.0?cveId=CVE-2026-41709 (read 19 Nov 2026) - CISA KEV feed, catalogVersion 2026.11.18, CVE-2026-41709: https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) - CISA BOD 26-04 implementation guidance, FAQ on waivers: https://www.cisa.gov/news-events/directives/bod-26-04-implementation-guidance-prioritizing-security-updates-based-risk (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Maximum Duration of a Risk Acceptance: 9 Sources](https://remediationvsmitigation.com/risk-acceptance-form/maximum-durations/index.md) - [Risk Acceptance Form: Free Template With 15 Fields](https://remediationvsmitigation.com/risk-acceptance-form/index.md) - [Exception Register: 15 Columns, States and Blank CSV](https://remediationvsmitigation.com/exception-register/index.md) - [Who Is the Risk Owner for a Security Finding?](https://remediationvsmitigation.com/guides/risk-owner/index.md) - [Accepting Risk on End-of-Life Software: Limits and Record](https://remediationvsmitigation.com/guides/accepting-risk-on-end-of-life-software/index.md) --- # Vulnerability false positive Canonical: https://remediationvsmitigation.com/guides/false-positive-handling/ A vulnerability false positive is a detected vulnerability that is not present in an exploitable state. FedRAMP's 2026 definition excludes any flaw that was ever present, so a mitigated or fixed one does not qualify. Close it with evidence from the system itself, not a second scan. Read 19 Nov 2026. ## Three definitions Glossary and FedRAMP definition text, read 19 Nov 2026. | Source | Wording | Difference | |---|---|---| | NIST SP 800-115 | "An alert that incorrectly indicates that a vulnerability is present." | Alert-based | | NIST SP 800-83 Rev. 1 | "An instance in which a security tool incorrectly classifies benign content as malicious." | Malware context, not vulnerabilities | | FedRAMP 2026 | "A detected vulnerability that is not actually present in an exploitable state in the information resource" | Includes code that is present but not in an operating state | FedRAMP's definition has two notes. The first covers vulnerable software "not loaded, running, or otherwise in an operating state required for exploitation". The second limits it: "This only applies if the vulnerability is not and was not present; a remediated vulnerability or a fully mitigated vulnerability cannot also be a false positive vulnerability." A flaw blocked by a control is mitigated; it is not a false positive, and it stays on the [exception register](https://remediationvsmitigation.com/exception-register/). ## What looks like a false positive and is not Distinctions drawn from FedRAMP's definition, VER-EVA-ELX and PCI SSC FAQ 1597, read 19 Nov 2026. | Looks like a false positive when | Actually | Where it goes | |---|---|---| | The flaw exists but exploitation needs conditions that never occur | A "not likely exploitable" evaluation | [Not reachable, not on the KEV](https://remediationvsmitigation.com/decision-table/not-reachable-low-risk/) | | The flaw exists and a control blocks the path | A mitigated flaw | [Control already in front](https://remediationvsmitigation.com/decision-table/compensating-control-in-front/) | | You disagree with the severity the scanner assigned | A change of risk assessment, not an absent flaw | PCI SSC FAQ 1597: under Requirement 6.3.1 the entity may rate a vulnerability for its own environment instead of accepting an outside rating | | The code is on disk but the vulnerable feature is not loaded or running | False positive under FedRAMP's first note, if it cannot be loaded either | This page; if it can be loaded, it is present and needs an exposure check | | The condition is truly absent on this system | False positive | This page | FedRAMP's rule VER-EVA-ELX gives the neighboring warning: "The simple reality is that most traditional vulnerabilities discovered by scanners or during assessment are not likely to be exploitable", but providers "who regularly evaluate vulnerabilities as not likely exploitable without careful consideration are more likely to suffer from an adverse impact". Declaring a false positive deserves the same care. A flaw that is present but unlikely to be exploited is evaluated and reported; a flaw that is absent is closed. SP 800-53 RA-5 keeps the order: "Analyze vulnerability scan reports and results from vulnerability monitoring" comes before "Remediate legitimate vulnerabilities". ## Why a second tool is weak evidence NIST SP 800-115 says "Automated tools can produce a significant number of findings, but these findings often need to be validated to isolate false positives. Assessors may validate vulnerabilities by manually examining the vulnerable system or by using a second automated tool and comparing the results." It warns that although a second tool is quick, "these comparison tools can often produce similar results", which include the same false positives. It adds that manual examination "typically provides more accurate results than comparing results from multiple tools, but it also has the potential to be time-consuming." The same publication says "vulnerability scanners can have a high false positive error rate (i.e., reporting vulnerabilities when none exist)". ## Worked example: a version-only scan on a patched Red Hat kernel CVE-2026-46300 is a Linux kernel flaw, named Fragnesia, in the code that merges network receive buffers. The function skb_try_coalesce() used to drop the marker that flags page fragments as shared. IPsec ESP input code trusts that marker: without it, ESP decrypts in place over fragments that belong to the page cache, so a local low-privilege user can alter the cached copy of a read-only file such as a setuid binary ([Red Hat Security Data API](https://access.redhat.com/hydra/rest/securitydata/cve/CVE-2026-46300.json)). The flaw is in socket buffer coalescing, and it is reachable only where the ESP code is loaded or can be loaded. NVD lists several affected mainline ranges; the one containing 5.14 runs from 5.11 up to but excluding 5.15.208. Red Hat fixed it in RHEL 9 by backport: errata RHSA-2026:19568 of 20 May 2026 ships kernel 5.14.0-687.10.1.el9_8. CISA added the CVE to the KEV catalog on 28 Oct 2026 with a due date of 31 Oct 2026 ([KEV feed](https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json)). Suppose a scanner that compares version numbers reports the invented host rhel9-app-04 on 29 Oct 2026, because the kernel number 5.14 falls inside the mainline range. The scanner never saw the backport. Two commands settle it: `uname -r` shows the running kernel, and `rpm -q --changelog kernel | grep CVE-2026-46300` shows whether the installed package carries the fix. Distribution kernels carry backports, so the version number alone can mislead. Outcomes for the invented host. Red Hat errata and CISA due date read 19 Nov 2026. | What the host shows | Finding is | Next step | |---|---|---| | `uname -r` shows 5.14.0-687.10.1.el9_8 or a later build, and the changelog lists CVE-2026-46300 | False positive | Close with the two outputs attached, 29 Oct 2026, two days before the 31 Oct due date | | The package is installed but `uname -r` shows an older kernel | Present: the host runs the vulnerable code until it reboots | Reboot, or apply one of Red Hat's two mitigations (below) and record it | | Older kernel, no changelog line | Present | Remediate by the 31 Oct due date | The second row is the common mistake: an installed fix that is not running is not a false positive. FedRAMP's note on code that is not loaded does not rescue it either, because the ESP code can be loaded and the running kernel is the vulnerable one. - Block the IPsec modules (Red Hat's broader method, bulletin RHSB-2026-003). Run `lsmod | grep -E 'esp4|esp6'` first. Red Hat treats IPsec as not in active use when neither module is loaded, and warns that blocking the modules breaks IPsec where it is configured. Write `install esp4 /bin/false` and `install esp6 /bin/false` as two lines into /etc/modprobe.d/dirtyfrag.conf, then run `rmmod esp4 esp6`. The rmmod unloads the modules at once (it refuses if they are in use) and the file stops them loading again, including after a reboot. A host that starts IPsec later loses it then, so confirm with the network team that no tunnel is planned. Deleting the file reverts it. - Disable unprivileged user namespaces (Red Hat's targeted method, for hosts where IPsec must stay): write `user.max_user_namespaces=0` into /etc/sysctl.d/dirtyfrag.conf and run `sysctl --system`. Rootless containers, sandboxed browsers and Flatpak can stop working. With either in place the host is mitigated, not a false positive: FedRAMP's second note says a fully mitigated vulnerability "cannot also be a false positive". It stays on the [exception register](https://remediationvsmitigation.com/exception-register/) until the reboot. ## Steps that leave a record 1. Log the detection. Owner: whoever raised the finding. Record: finding ID, tool, signature, and the time and source of the detection. 2. Examine the system, not the report. Owner: the system's administrator. Record: installed and running version, whether the component is loaded or running, the setting involved, with date and command output (SP 800-115, manual examination). 3. Have someone other than the person who raised or will close the finding validate it. Owner: a second engineer or the assessor. Record: validator name and date. FedRAMP's legacy playbook marked a false positive Yes and moved it to the Closed tab when the third-party assessor validated it, because "validated FPs are not considered open risks"; one not validated stayed Pending until the agency authorizing official approved it. 4. Tune or suppress the signature, with the reason. Owner: the scanner owner. Record: the rule change, so the same alert does not return next scan. 5. Close, and write the trigger that reopens it. Owner: the finding owner. Record: closing date, and the reopen triggers: a change of build or configuration, or a new scanner signature. Evidence is true for one build and one configuration; see [verifying a mitigation](https://remediationvsmitigation.com/guides/verifying-a-mitigation-works/) for drift checks. If verification shows the flaw is present after all, the finding moves to the row of the [Decision Table](https://remediationvsmitigation.com/decision-table/) that fits, and the clock does not restart at verification. BOD 26-04 says the timelines begin when CISA adds the vulnerability to the KEV catalog or the agency identifies it on an asset, and "Whichever event occurs first starts the remediation timeline"; see [patch available and exposed](https://remediationvsmitigation.com/decision-table/kev-patch-available-exposed/). A real finding is recorded on the [risk acceptance form](https://remediationvsmitigation.com/risk-acceptance-form/) if it will be accepted. For the legacy deviation request types that preceded the 2026 rules, which also covered false positives, see [FedRAMP deviation requests](https://remediationvsmitigation.com/guides/fedramp-deviation-requests/). ## Sources - NIST CSRC glossary, false positive: https://csrc.nist.gov/glossary/term/false_positive (read 19 Nov 2026) - NIST SP 800-115, Section 7.3 and glossary: https://nvlpubs.nist.gov/nistpubs/Legacy/SP/nistspecialpublication800-115.pdf (read 19 Nov 2026) - FedRAMP 2026 definitions, False Positive Vulnerability: https://www.fedramp.gov/2026/definitions/ (read 19 Nov 2026) - FedRAMP 2026, VER-EVA-EFP, VER-EVA-ELX, VER-RPT-VDT and VER-RPT-AVI: https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-evaluation-and-reporting/ (read 19 Nov 2026) - FedRAMP legacy playbook, POA&M (reference only after 24 Jun 2026): https://www.fedramp.gov/legacy/playbook/csp/authorization/poam/ (read 19 Nov 2026) - PCI SSC FAQ 1597: https://www.pcisecuritystandards.org/faqs/1597 (read 19 Nov 2026) - CISA BOD 26-04, Appendix A: https://www.cisa.gov/news-events/directives/bod-26-04-prioritizing-security-updates-based-risk (read 19 Nov 2026) - NIST SP 800-53 Rev. 5, RA-5: https://csrc.nist.gov/projects/cprt/catalog#/cprt/framework/version/SP_800_53_5_2_0/home/RA-5 (read 19 Nov 2026) - Red Hat Security Data API, CVE-2026-46300: https://access.redhat.com/hydra/rest/securitydata/cve/CVE-2026-46300.json (read 19 Nov 2026) - CISA KEV feed, catalog version 2026.11.18: https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) - Red Hat security bulletin RHSB-2026-003, Dirty Frag and Fragnesia: https://access.redhat.com/security/vulnerabilities/RHSB-2026-003 (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Decision Table: Remediate, Mitigate or Accept by Situation](https://remediationvsmitigation.com/decision-table/index.md) - [Not Exposed, Not on the KEV, Not Automatable: Accept](https://remediationvsmitigation.com/decision-table/not-reachable-low-risk/index.md) - [A Compensating Control Is in Front: Why a Fix Date Stays](https://remediationvsmitigation.com/decision-table/compensating-control-in-front/index.md) - [FedRAMP Deviation Requests and Their 2026 Replacements](https://remediationvsmitigation.com/guides/fedramp-deviation-requests/index.md) - [Exception Register: 15 Columns, States and Blank CSV](https://remediationvsmitigation.com/exception-register/index.md) - [How to Verify a Mitigation Works: Checks for 5 Real CVEs](https://remediationvsmitigation.com/guides/verifying-a-mitigation-works/index.md) - [Risk Acceptance Form: Free Template With 15 Fields](https://remediationvsmitigation.com/risk-acceptance-form/index.md) --- # FedRAMP deviation requests Canonical: https://remediationvsmitigation.com/guides/fedramp-deviation-requests/ FedRAMP's legacy deviation request form covered false positives, risk adjustments and operational requirements, and FedRAMP would not approve an operational requirement for a high-risk vulnerability. The 2026 Rev5 rules, required for certification from 7 Dec 2026, use none of the three terms: a vulnerability not fixed within 192 days of evaluation is reported as accepted. ## The three types in the legacy form The form's own notice reads "June 23, 2026" and calls it a legacy document "being replaced by the FedRAMP Consolidated Rules for 2026". FedRAMP's legacy documentation site marks every page reference only from 24 Jun 2026, the day the 2026 rules launched. From the FedRAMP Vulnerability Deviation Request Form (legacy), read 19 Nov 2026. | Type | Form definition | Limit or evidence | |---|---|---| | False positive (FP) | "A finding that incorrectly indicates a given condition exists when it does not." | "It must be justified through documented explanation and evidence." | | Risk adjustment (RA) | "A reduction in the scanner-cited risk level of a finding." Done through compensating controls that reduce likelihood and impact | Rationale, evidence and CVSS environmental metrics | | Operational requirement (OR) | "A finding that cannot be remediated, often because the system will not function as intended." | "FedRAMP will not approve an OR for a high-risk vulnerability"; the provider may request an RA with it | | RA and OR | One request can justify both | Both sections completed | ## What the form asked for One POA&M item per row, with a fixed set of fields: tool-provided description and recommended action, initial risk level, original detection date, requested risk level, "DR Rationale", "Evidence Description" and "List of Evidence Attachments". An operational requirement asks the provider to "Explain the limitations that prevent the vulnerability from being fixed. Include negative operational impacts of remediation." A Moderate vulnerability not reduced to Low needs an explanation of "why the authorizing official should accept the risk without mitigating it." A risk adjustment is argued with the CVSS environmental metrics: attack vector, complexity, privileges, user interaction, and impact on confidentiality, integrity and availability, plus a remediation level ("Official fix", "Temporary fix", "Workaround" or "Unavailable"). Each metric needs a sentence on the control that changes it, such as a management network reachable only through a jump host. ## Pending is not approved The form says "FedRAMP adjudicates each DR individually". A provider may mark the matching POA&M item Pending after submitting, "while waiting for FedRAMP adjudication; however, they may only treat the vulnerability differently after FedRAMP approves the DR." Status values are Pending, Approved, Rejected and Withdrawn. The [legacy POA&M playbook](https://www.fedramp.gov/legacy/playbook/csp/authorization/poam/) adds that "Approved ORs are still considered open risks" and must be reassessed periodically. Vendor dependencies were handled beside the types. The form says a high-risk one "must be risk-adjusted to a lower risk rating and an RA DR must be submitted"; the playbook says VDs "are not considered deviation requests and do not require approval" and that high-risk VDs must be mitigated to a Moderate level within 30 days. Read them together: the DR is needed only for the risk adjustment, not for the dependency. ## What the 2026 rules use instead On 19 Nov 2026 the phrases "deviation request", "risk adjustment" and "operational requirement" had no match in FedRAMP's Vulnerability Detection and Response rules, Vulnerability Evaluation and Reporting rules or 2026 definitions, and neither rule page contains the word "approve". The provider evaluates, reports a state and does not wait for a decision. The pairing below is this site's reading, not a FedRAMP mapping. Mapping drawn by this site from the 2026 rule pages, read 19 Nov 2026. The rules are required for Rev5 initial and ongoing certification from 7 Dec 2026; the grace period ends 7 Mar 2027. | Legacy type | 2026 mechanism | Rule | |---|---|---| | False positive | Evaluate for false positives (SHOULD) | VER-EVA-EFP | | Risk adjustment | Partial or full mitigation, with each reduction in the Potential Agency Impact N-rating reported | VDR-TFR-PVR, VER-RPT-VDT | | Operational requirement | Accepted vulnerability: not fully mitigated or remediated within 192 days of evaluation, with an explanation | VER-TFR-MAV, VER-RPT-AVI | | Vendor dependency | No separate term found on 19 Nov 2026 | none | FedRAMP's definition of an accepted vulnerability (FRD-ACV) is "A vulnerability that the provider does not intend to fully mitigate or remediate, OR that has not or will not be fully mitigated or remediated within the maximum overdue period". What closes each state is on the [exception register](https://remediationvsmitigation.com/exception-register/), and the longest periods are on [maximum durations](https://remediationvsmitigation.com/risk-acceptance-form/maximum-durations/). ## Worked example: an operational requirement under the 2026 rules The provider, Class C certification and asset are invented for illustration. The CVE, the Apache advisory and the KEV dates are real. Apache announced CVE-2026-90436 on 6 Nov 2026: Axis 1.x through 1.4 is end of life, "we recommend you migrate to a different SOAP engine", and "The Apache Axis project does not expect to create an Axis 1.x release fixing this problem." CISA added it to the [KEV catalog](https://remediationvsmitigation.com/guides/kev-required-actions-mitigate-or-discontinue/) on 18 Nov 2026 with a due date of 21 Nov 2026. The legacy POA&M playbook describes the case as an OR where "a vendor explicitly indicated it does not intend to offer a fix". Day counts from the Class C table in VDR-TFR-PVR and the 192-day rule in VER-TFR-MAV, read 19 Nov 2026. Dates counted from 19 Nov 2026. | Date | Event | Rule and result | |---|---|---| | 18 Nov 2026 | Scanner finds Axis 1.4 on the partner SOAP gateway | Detection time and source recorded (VER-RPT-VDT) | | 19 Nov 2026 | Evaluation complete: internet-reachable, likely exploitable (it is on the KEV list), exploitation assumed automatable, rated N4 | VER-EVA-EIR, VER-EVA-ELX, VER-EVA-AIA, VER-EVA-EPA | | 20 Nov 2026 | Gateway moved behind the partner VPN; no longer internet-reachable; re-rated N3 | Reduction reported with its time and rating (VER-RPT-VDT) | | 23 Nov 2026 | End of the Class C window for N4 with LEV and IRV (4 days from evaluation); the 20 Nov reduction came first | VDR-TFR-PVR (SHOULD partially mitigate, fully mitigate or remediate) | | 21 Dec 2026 | Class C window for N3 with LEV and NIRV is 32 days from evaluation | VDR-TFR-PVR | | 30 May 2027 | 192 days from evaluation: the last day before the item must be marked accepted if still open | VER-TFR-MAV | Two outcomes follow. If the replacement SOAP engine goes live on 15 Mar 2027 (116 days after evaluation), the item is remediated and never becomes an accepted vulnerability; the monthly report carries its reductions and the estimated time of the next one. If cutover slips past 30 May 2027, or the provider decides earlier that it will not migrate, the item is accepted and reported with the record below. The old form would have asked FedRAMP to approve an OR and an RA together; now the provider writes the explanation itself and reports it. Accepted vulnerability record (fields of VER-RPT-AVI, illustrative values) 1. Tracking identifier: SOAP-GW-0412 2. Time and source of detection: 18 Nov 2026, authenticated scan of the partner SOAP gateway 3. Time of completed evaluation: 19 Nov 2026 4. Internet-reachable: No, since 20 Nov 2026 (partner VPN only) 5. Likely exploitable: Yes (CVE-2026-90436, KEV entry added 18 Nov 2026) 6. Currently estimated Potential Agency Impact N-rating: N3 7. Explanation of why this is an accepted vulnerability: Apache has said no Axis 1.x fix will be released; cutover to the replacement engine was not complete 192 days after evaluation (30 May 2027). Cutover now planned for 30 Sep 2027. 8. Supplementary information: VPN-only access, WAF rule on the gateway, monitoring of the gateway's outbound connections, review on the first of each month. The KEV catalog entry still asks for action by 21 Nov 2026, and VDR-TFR-KEV says providers SHOULD remediate KEVs by the catalog due dates "even if the vulnerability has been fully mitigated". Where the fix does not exist, the record above is where that gap is written down. The same gap in a POA&M is covered in the [POA&M guide](https://remediationvsmitigation.com/guides/poam-plan-of-action-and-milestones/), and a scanner finding that is simply wrong goes through [false positive handling](https://remediationvsmitigation.com/guides/false-positive-handling/). ## Sources - FedRAMP legacy Vulnerability Deviation Request Form (xlsx), notice dated 23 Jun 2026: https://www.fedramp.gov/legacy/assets/LEGACY%20FedRAMP-Vulnerability-Deviation-Request-Form.xlsx (read 19 Nov 2026) - FedRAMP legacy playbook, POA&M: https://www.fedramp.gov/legacy/playbook/csp/authorization/poam/ (read 19 Nov 2026) - FedRAMP 2026, Vulnerability Evaluation and Reporting rules (VER): https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-evaluation-and-reporting/ (read 19 Nov 2026) - FedRAMP 2026, Vulnerability Detection and Response rules (VDR): https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-detection-and-response/ (read 19 Nov 2026) - FedRAMP 2026 definitions: https://www.fedramp.gov/2026/definitions/ (read 19 Nov 2026) - Apache Axis project announcement of CVE-2026-90436, 6 Nov 2026: https://lists.apache.org/thread/8slf0j9gz35z56t0glt5h0ncwlbn4bdz (read 19 Nov 2026) - CISA KEV JSON feed, catalogVersion 2026.11.18: https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [POA&M Example: A Filled Plan of Action and Milestones Entry](https://remediationvsmitigation.com/guides/poam-plan-of-action-and-milestones/index.md) - [Vulnerability False Positive: Close It With Evidence](https://remediationvsmitigation.com/guides/false-positive-handling/index.md) - [Maximum Duration of a Risk Acceptance: 9 Sources](https://remediationvsmitigation.com/risk-acceptance-form/maximum-durations/index.md) - [Exception Register: 15 Columns, States and Blank CSV](https://remediationvsmitigation.com/exception-register/index.md) - [KEV Required Action: Mitigate or Discontinue Use](https://remediationvsmitigation.com/guides/kev-required-actions-mitigate-or-discontinue/index.md) --- # How virtual patching works Canonical: https://remediationvsmitigation.com/guides/how-virtual-patching-works/ A virtual patch is a rule in a WAF, IPS or web server module that blocks matching traffic to a known flaw without changing the vulnerable code. OWASP's method has six phases and starts rules in log-only mode. This page carries one rule through them for CVE-2026-40217, KEV due 2 Dec 2026. ## What counts as a virtual patch OWASP defines it as "A security policy enforcement layer that detects and blocks matching exploitation attempts against a known vulnerability without changing the vulnerable code." It adds that a virtual patch "does not remove the underlying vulnerability" and reduces exposure only for traffic the layer inspects. OWASP names three places for the layer: an intermediary device such as a WAF or IPS appliance, a web server plugin such as ModSecurity, or an application layer filter ([OWASP Virtual Patching Cheat Sheet](https://cheatsheetseries.owasp.org/cheatsheets/Virtual_Patching_Cheat_Sheet.html)). ## The six phases, with owner and record Phases and aims from the OWASP Virtual Patching Cheat Sheet, read 19 Nov 2026. The owner and failure columns are this site's summary of where the work lands. | Phase | Who usually does it | What is done | Record left | Where it usually fails | |---|---|---|---|---| | 1. Preparation | Security operations with the change board | Vendor alert subscriptions, a pre-authorized expedited rollout, the tool installed before it is needed, audit logging configured | Standing change authorization; logging standard | Approval sought during the incident, so the rule waits days | | 2. Identification | Vulnerability management | Flaw found by assessment or code review (proactive), or by vendor notice, public disclosure or an incident (reactive) | Ticket with source and date | Notice arrives with the application team, not the WAF team | | 3. Analysis | Application owner and WAF engineer | Confirm the CVE or internal ID, affected versions and the exact route or parameter; decide whether a rule can express it (OWASP: ideally suited to injection-type flaws) | Affected-asset list; rule design note | Flaw is in logic or authorization, which a pattern cannot see | | 4. Rule creation | WAF engineer | Write the rule. OWASP's two aims: no false positives, no false negatives. Prefer an allow-list on the route to a copy of one sample request | Rule text and rule ID in the ticket | Rule written for one exploit sample, which OWASP says has low long-term value | | 5. Implementation and testing | WAF engineer, then the assessor or a second engineer | Deploy in log-only mode first, review hits, switch to blocking, request a retest; if the retest finds evasions, return to analysis | Test log; retest result; date blocking began | Tested with one request, not alternate encodings or a direct path that skips the WAF | | 6. Recovery and follow-up | Application owner | Record the change, review periodically, remove the rule after the code fix, run reports on when the rule fired | Removal date and reason; hit report | No removal date, so the rule outlives its reason | ## Worked example: LiteLLM CVE-2026-40217 The flaw and the vendor workaround are real; the team, host names and test results are invented for illustration. The BerriAI advisory GHSA-wxxx-gvqv-xp7p, published 7 May 2026, says the POST /guardrails/test_custom_code endpoint runs user-supplied code in a sandbox that can be escaped, that reaching the endpoint "requires a proxy-admin credential in default configurations", and that the fix is to upgrade to 1.83.11 or later. Its workaround reads: "If upgrading is not immediately possible, block POST /guardrails/test_custom_code at your reverse proxy or API gateway." CISA added the CVE to the KEV catalog on 18 Nov 2026 with a due date of 2 Dec 2026 (14 days). A rule for that route carries no payload, so it is a route block and not an exploit-specific pattern. Step one, log only. It matches the path after URL decoding and normalization, so encoded and dotted variants of the path are seen, and it logs without blocking: # Rule 9100001, CVE-2026-40217 LiteLLM, stage 1: log only SecRule REQUEST_FILENAME "@rx ^/guardrails/test_custom_code/*$" \ "id:9100001,phase:1,t:none,t:urlDecodeUni,t:normalizePath,t:lowercase,pass,log,auditlog,msg:'VP CVE-2026-40217 LiteLLM test_custom_code, log only'" # Stage 2, enforcement: same rule with the action changed # ...,deny,status:403,log,auditlog,msg:'VP CVE-2026-40217 LiteLLM test_custom_code, blocked' Advisory and KEV dates read 19 Nov 2026; every other row is a worked plan, not an observed result. | Date | Step | Record | |---|---|---| | 19 Nov 2026 | Stage 1 deployed on the reverse proxy in front of the LiteLLM proxy | Change ticket with rule ID 9100001; removal review set for 8 Dec 2026 | | 19 to 23 Nov 2026 | Hits reviewed daily. Expected: none, since the route needs a proxy-admin credential. Any hit is matched to a named admin session | Hit report attached to the ticket | | 24 Nov 2026 | Stage 2 enabled. Retest by a second engineer: the path with a trailing slash, with percent-encoded characters and with doubled slashes returns 403; a request straight to the service port from the internal network is not blocked | Retest result. The direct-port result is the rule's gap, so the service port is closed to all but the proxy | | 1 Dec 2026 | Upgrade to 1.83.11 completed, one day before the KEV due date | Version before and after in the ticket | | 8 Dec 2026 | Rule removed after a week of no hits on the upgraded service | Removal date and reason closed in the ticket | The rule matches all methods, not only POST. That is wider than the vendor text and costs nothing unless admins use the route; if they do, restrict the route by source network instead. The vendor's advisory is a workaround for the period before the upgrade, which is why the plan ends in the fix and a removal date. ## What the log should hold OWASP asks for the rule ID, timestamp, affected endpoint, enforcement outcome and a correlation identifier. It warns against capturing query strings, headers, cookies and bodies indiscriminately because "WAF alerts can expose passwords and payment data"; sensitive fields should be excluded or redacted. The test record doubles as the evidence for the [WAF rule entry in the register](https://remediationvsmitigation.com/compensating-controls/waf-rule/), and the same checks apply to an [IPS signature](https://remediationvsmitigation.com/compensating-controls/ips-signature/). ## Where a rule is silently off - Cloudflare's documentation says "Some rules in the Cloudflare Managed Ruleset are disabled by default" ([vendor documentation](https://developers.cloudflare.com/waf/managed-rules/reference/cloudflare-managed-ruleset/), vendor claim). Check that the rule for your flaw is on. - Check Point's IPS page describes an optional detect-only mode that "sets all protections to only detect" ([vendor page](https://www.checkpoint.com/quantum/intrusion-prevention-system-ips/), vendor claim). A detect-only sensor logs and does not block. - ModSecurity's engine setting has a DetectionOnly value, which the v3 reference manual describes as passive: "the intent is to be totally passive and not take any disruptive actions". A deny action in a rule then does nothing ([reference manual](https://github.com/owasp-modsecurity/ModSecurity/wiki/Reference-Manual-(v3.x))). OWASP: "A passing test for one payload does not establish complete protection." Test alternate encodings and every other path that reaches the same component. ## Keeping it tied to a fix OWASP's author notes that many teams leave virtual patches in place after the code fix because the logging is better than the application's own. That is a choice to record, not a default. A virtual patch with no removal date is the pattern the [maximum durations page](https://remediationvsmitigation.com/risk-acceptance-form/maximum-durations/) warns about. For checking that a rule stays in force see [verifying a mitigation](https://remediationvsmitigation.com/guides/verifying-a-mitigation-works/). ## Sources - OWASP Virtual Patching Cheat Sheet: https://cheatsheetseries.owasp.org/cheatsheets/Virtual_Patching_Cheat_Sheet.html (read 19 Nov 2026) - BerriAI LiteLLM advisory GHSA-wxxx-gvqv-xp7p, published 7 May 2026: https://github.com/BerriAI/litellm/security/advisories/GHSA-wxxx-gvqv-xp7p (read 19 Nov 2026) - CISA KEV JSON feed, catalogVersion 2026.11.18, entry CVE-2026-40217: https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) - ModSecurity v3 Reference Manual, transformations and SecRuleEngine: https://github.com/owasp-modsecurity/ModSecurity/wiki/Reference-Manual-(v3.x) (read 19 Nov 2026) - Cloudflare, Managed Ruleset reference (vendor): https://developers.cloudflare.com/waf/managed-rules/reference/cloudflare-managed-ruleset/ (read 19 Nov 2026) - Check Point, IPS page (vendor): https://www.checkpoint.com/quantum/intrusion-prevention-system-ips/ (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [When Virtual Patching Is Enough, and When It Is Not](https://remediationvsmitigation.com/guides/virtual-patching-and-when-it-is-enough/index.md) - [WAF Rule as a Virtual Patch: Atlassian Case and Limits](https://remediationvsmitigation.com/compensating-controls/waf-rule/index.md) - [IPS Signature as a Mitigation: When It Counts](https://remediationvsmitigation.com/compensating-controls/ips-signature/index.md) - [Maximum Duration of a Risk Acceptance: 9 Sources](https://remediationvsmitigation.com/risk-acceptance-form/maximum-durations/index.md) - [How to Verify a Mitigation Works: Checks for 5 Real CVEs](https://remediationvsmitigation.com/guides/verifying-a-mitigation-works/index.md) - [Mitigation vs Remediation Wording in Vendor Advisories](https://remediationvsmitigation.com/guides/mitigation-vs-remediation-in-vendor-advisories/index.md) --- # Inherent vs residual risk Canonical: https://remediationvsmitigation.com/guides/inherent-vs-residual-risk/ Inherent risk is the risk before any management action; residual risk is the exposure left after action. NIST IR 8286 quotes COSO and OMB for both, then records "current risk" instead. The words "inherent risk" appear 0 times in SP 800-37 Rev. 2, SP 800-39 and SP 800-40 Rev. 4 (read 19 Nov 2026). ## Definitions as published Quotations as printed in NIST IR 8286, Section 3.2, read 19 Nov 2026. | Term | Wording | Source | |---|---|---| | Inherent risk | "The risk to an entity in the absence of any direct or focused actions by management to alter its severity." | COSO ERM, via NIST IR 8286 and the NIST glossary | | Inherent risk | "conditions in the absence of risk management actions" | OMB Circular A-123, via NIST IR 8286 | | Residual risk | "the exposure remaining from an inherent risk after action has been taken to manage it, using the same assessment standards as the inherent assessment" | OMB Circular A-123, via NIST IR 8286 | | Target residual risk | "the amount of risk that an entity prefers to assume in the pursuit of its strategy and business objectives" | COSO ERM, via NIST IR 8286 | | Actual residual risk | "the risk remaining after management has taken action to alter its severity. Actual residual risk should be equal to or less than the target residual risk." | COSO ERM, via NIST IR 8286 | NIST IR 8286 states that for federal agencies, OMB Circular A-123 requires the enterprise risk register to "consider both inherent and residual risk". Its own register uses a third state: "this publication typically refers to current risk (rather than inherent risk) that represents a baseline risk posture." Current risk is measured before the response in question and may already include existing controls, which is why it is the practical starting figure for a vulnerability on a running system. ## Which texts use which word Counts on 19 Nov 2026. Method: text extracted from each NIST PDF with pypdf 7.1.1, hyphenated line breaks joined, whitespace collapsed, case-insensitive count of the exact phrase (the figure for IR 8286 includes its glossary and footnotes). SP 800-53 was counted in the NIST OSCAL JSON of Rev. 5.2.0, control prose only. FedRAMP and BOD figures are from the page text of the VER and VDR rules, the definitions page and the directive page. | Document | "inherent risk" | "residual risk" | |---|---|---| | NIST SP 800-37 Rev. 2 | 0 | 6 | | NIST SP 800-39 | 0 | 6 | | NIST SP 800-30 Rev. 1 | 0 | 1 | | NIST SP 800-40 Rev. 4 | 0 | 0 | | NIST SP 800-82 Rev. 3 | 0 | 6 | | NIST SP 800-53 Rev. 5.2.0 control text | 1 (discussion of MA-5(2), maintenance personnel) | 2 (discussion of SI-19, de-identification) | | NIST IR 8286 | 6 | 25 | | FedRAMP 2026 rules and definitions | 0 | 0 | | CISA BOD 26-04 | 0 | 0 | The federal publications that govern authorization use only the second word, and SP 800-53 uses both only inside guidance text. The BOD 26-04 clocks and FedRAMP's partial and full mitigation definitions use neither; they describe the asset as it is now. BOD 26-04 says "The timelines for each asset are dynamic as facts change". A decision record built on those rules therefore stores the current facts and the target, not a reconstruction of an unmanaged past. ## Worked example: one record with three states The environment is invented; the CVE, the advisory and the dates are real. A team runs a Synology DiskStation Manager NAS with a port forward from the internet. CVE-2024-10441 (CVSS 3.1 base score 9.8, network, no privileges) is in Synology advisory SA-24:20, published 5 Nov 2024 with fixed releases and, in the vendor's words, "Mitigation None". CISA added it to the KEV catalog on 18 Nov 2026 with a due date of 21 Nov 2026 and forensic triage marked Yes, which is why the window is 3 days. The scale is an illustrative 1 to 5 for likelihood and for impact; exposure is their product, as IR 8286 defines exposure as the combination of likelihood and impact. Record layout follows the IR 8286 register elements (current assessment, response, residual, target residual). Scale, values and owners are illustrative; 21 Nov 2026 is a Saturday. | State | Date | Likelihood | Impact | Exposure | What changed | Set by | |---|---|---|---|---|---|---| | Current risk | 18 Nov 2026 | 5 | 4 | 20 | Port forward open; the entry was added to the KEV list that day | Vulnerability analyst proposes; head of IT operations, the risk owner, confirms | | Target residual risk | 18 Nov 2026 | | | 8 or lower | Appetite for an internal file server; a KEV entry may not stay above it past the due date | Head of IT operations (risk owner) | | Residual risk, first response | 19 Nov 2026 | 2 | 4 | 8 | Port forward removed; NAS reachable only from the office network. No vendor mitigation exists, so this is the owner's own control | NAS administrator measures; risk owner accepts | | Residual risk, after fix (planned) | 21 Nov 2026 | 1 | 4 | 4 | DSM upgraded in the Saturday maintenance window to a fixed release listed in the advisory (for DSM 7.2.2: 7.2.2-72806-1 or above) | NAS administrator upgrades; a second administrator verifies the build number and signs | The 19 Nov row is the one a single-number record cannot show. Exposure fell from 20 to 8 and met the target of 8, yet the KEV date of 21 Nov 2026 still asks for the vendor's action, so the record also keeps the response and its date: hold the interim control for two days, then remediate on 21 Nov, the due date itself. If the window slipped past 21 Nov, the gap would need a signed, dated exception that names the interim control and the new fix date; the due date would still be missed. The actual-below-target test is the tolerance test in [risk acceptance versus risk tolerance](https://remediationvsmitigation.com/guides/risk-acceptance-vs-risk-tolerance/), and how the second number is estimated is in [residual risk after mitigation](https://remediationvsmitigation.com/guides/residual-risk-after-mitigation/). ## Where each state appears on the record The response is accept, transfer, mitigate or avoid (IR 8286 Table 3), with its date and the risk owner, defined in IR 8286 as "the designated party responsible and accountable for ensuring that the risk is maintained in accordance with enterprise requirements". After it, IR 8286 says "the results of a post-response assessment could be reflected in the register as the Residual Risk", and an organization "might even document a desired risk state based on risk appetite/tolerance, the Target Residual Risk". The pre-response fields sit on the [risk acceptance form](https://remediationvsmitigation.com/risk-acceptance-form/). The register and the [exception register](https://remediationvsmitigation.com/exception-register/) hold the same finding at different grain; how they link is on [risk register versus exception register](https://remediationvsmitigation.com/guides/risk-register-vs-exception-register/). ## When the inherent framing is the wrong tool - A vulnerability on a running asset has no unmanaged state to measure. The firewall, login and monitoring already exist, so current risk is the honest starting number. Inherent risk fits a design decision, such as whether to build a service at all. - A KEV due date or a FedRAMP timeframe turns on facts now (internet exposure, exploitation status, impact rating), not on a before-and-after pair. Record the facts and the clock, then the residual. - A residual figure counts only for a control that has been checked. Until then the record holds the current figure and a planned residual, labeled as planned. ## Sources - NIST CSRC glossary, inherent risk: https://csrc.nist.gov/glossary/term/inherent_risk (read 19 Nov 2026) - NIST CSRC glossary, residual risk: https://csrc.nist.gov/glossary/term/residual_risk (read 19 Nov 2026) - NIST IR 8286, Sections 2 and 3.2 and glossary: https://nvlpubs.nist.gov/nistpubs/ir/2020/NIST.IR.8286.pdf (read 19 Nov 2026) - NIST SP 800-37 Rev. 2: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-37r2.pdf (read 19 Nov 2026) - NIST SP 800-39: https://nvlpubs.nist.gov/nistpubs/Legacy/SP/nistspecialpublication800-39.pdf (read 19 Nov 2026) - NIST SP 800-30 Rev. 1: https://nvlpubs.nist.gov/nistpubs/Legacy/SP/nistspecialpublication800-30r1.pdf (read 19 Nov 2026) - NIST SP 800-40 Rev. 4: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf (read 19 Nov 2026) - NIST SP 800-82 Rev. 3: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-82r3.pdf (read 19 Nov 2026) - NIST SP 800-53 Rev. 5.2.0 catalog (OSCAL JSON): https://raw.githubusercontent.com/usnistgov/oscal-content/main/nist.gov/SP800-53/rev5/json/NIST_SP-800-53_rev5_catalog.json (read 19 Nov 2026) - CISA BOD 26-04: https://www.cisa.gov/news-events/directives/bod-26-04-prioritizing-security-updates-based-risk (read 19 Nov 2026) - FedRAMP 2026 definitions: https://www.fedramp.gov/2026/definitions/ (read 19 Nov 2026) - Synology advisory Synology-SA-24:20: https://www.synology.com/en-us/security/advisory/Synology_SA_24_20 (read 19 Nov 2026) - CISA KEV JSON feed, catalogVersion 2026.11.18: https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Residual Risk After Mitigation: Define, Estimate, Record](https://remediationvsmitigation.com/guides/residual-risk-after-mitigation/index.md) - [Risk Acceptance Form: Free Template With 15 Fields](https://remediationvsmitigation.com/risk-acceptance-form/index.md) - [Remediation and Mitigation Glossary: 31 Terms, Sourced](https://remediationvsmitigation.com/glossary/index.md) - [Risk Register vs Exception Register: How They Link](https://remediationvsmitigation.com/guides/risk-register-vs-exception-register/index.md) - [Risk Acceptance vs Risk Tolerance: Who Sets the Limit](https://remediationvsmitigation.com/guides/risk-acceptance-vs-risk-tolerance/index.md) --- # Apply mitigations per vendor instructions or discontinue use of the product Canonical: https://remediationvsmitigation.com/guides/kev-required-actions-mitigate-or-discontinue/ In CISA's KEV feed of 19 Nov 2026 (catalog 2026.11.18), 893 of 1,786 entries say only to apply updates, 762 say to apply mitigations or discontinue the product, 83 name an end-of-life product to disconnect, and 48 use other wording. All 302 entries added in 2026 carry a discontinue clause. ## Seven wordings in the feed Counted by script from the CISA KEV JSON feed (catalogVersion 2026.11.18, released 18 Nov 2026) on 19 Nov 2026. Each of the 1,786 entries is in one row, sorted by its requiredAction text in the order of the rows. Rows 2 to 4 are the standard mitigate-or-discontinue strings: 762 entries. Vendor counts are per row and overlap between rows. | Required action | Entries | Added between | Vendors | Days from added to due date (entries) | |---|---|---|---|---| | "Apply updates per vendor instructions." and nothing else | 893 | 3 Nov 2021 and 23 Jun 2023 | 150 | 21 (441); 181 (235); 14 (192); 182 (11); 184 (8); 31 (4); 24 or 25 (2) | | "Apply mitigations per vendor instructions or discontinue use of the product if mitigations are unavailable." | 264 | 17 Jul 2023 and 24 Feb 2025 | 86 | 21 (245); 7 (8); 4 (5); 8 (2); 12 (2); 2 (1); 43 (1) | | The same, with "follow applicable BOD 22-01 guidance for cloud services" | 329 | 10 Oct 2023 and 9 Jun 2026 | 124 | 21 (235); 14 (55); 3 (27); 7 (9); 1, 2 and 5 (1 each) | | "Apply mitigations in accordance with vendor instructions, ensuring compliance with CISA's BOD 26-04" and forensic triage | 169 | 11 Jun 2026 and 18 Nov 2026 | 79 | 3 (131); 14 (38) | | End-of-life product: disconnect, discontinue, remove or retire | 83 | 3 Nov 2021 and 22 Jul 2025 | 19 | 21 (71); 14 (6); 1 (3); 181 (3) | | Older form: "updates ... or discontinue use of the product if updates are unavailable" (also "remediations or mitigations") | 28 | 29 Jun 2023 and 8 Oct 2024 | 10 | 21 (28) | | Other: emergency directive instructions, CISA instructions, contact the vendor, disable a service | 20 | 10 Dec 2021 and 14 May 2026 | 9 | 21 (5); 3 (4); 7 (3); 4 (3); 1 (2); 2 (2); 14 (1) | The word "discontinue" appears in 815 entries: all 762 in rows 2 to 4, 28 in row 6, 15 in row 5 and 10 in row 7. Of the 547 entries added since 1 Jan 2025, 531 are in rows 2 to 4. ## The wording changed with the year From the same feed, read 19 Nov 2026. Other wording combines rows 6 and 7. | Year added | Entries | Updates only | Mitigate or discontinue (rows 2 to 4) | End-of-life | Other wording | Contain "discontinue" | |---|---|---|---|---|---|---| | 2021 | 311 | 309 | 0 | 1 | 1 | 0 | | 2022 | 555 | 499 | 0 | 54 | 2 | 0 | | 2023 | 187 | 85 | 69 | 4 | 29 | 95 | | 2024 | 186 | 0 | 162 | 18 | 6 | 175 | | 2025 | 245 | 0 | 235 | 6 | 4 | 243 | | 2026 | 302 | 0 | 296 | 0 | 6 | 302 | Through 23 Jun 2023 the standard action was an update. The first discontinue clauses were added on 29 Jun 2023, still worded around updates, and the standard mitigations string starts on 17 Jul 2023. A reader treating KEV as a patch list sees only part of the instruction. ## Due dates follow the wording Across the 815 entries with a discontinue clause, 522 are due 21 days after they were added, 162 after 3 days, 93 after 14 days, 17 after 7 days, 6 after 1 day, 5 after 4 days, 4 after 2 days, 2 each after 8 and 12 days, and 1 each after 5 and 43 days. The newest wording, from 11 Jun 2026, names BOD 26-04 and CISA's forensic triage guidance. Its feed field forensicTriage is "Yes" on 108 entries, every one due in 3 days; of the 61 marked "No", 23 are due in 3 days and 38 in 14 days. The date logic for those entries is in [patch available and exposed](https://remediationvsmitigation.com/decision-table/kev-patch-available-exposed/). ## Two entries, one clause, opposite instructions Both entries were added on 18 Nov 2026 with the BOD 26-04 wording, which ends: "Follow applicable BOD 26-04 guidance for cloud services or discontinue use of the product if mitigations are unavailable." Both are due 21 Nov 2026, 3 days later, and the feed marks both forensicTriage "Yes". What the vendors say differs. Entries from the KEV feed and vendor pages, read 19 Nov 2026. | | Synology DiskStation Manager, CVE-2024-10441 | Apache Axis 1.x, CVE-2026-90436 | |---|---|---| | Vendor source in the entry's notes | Synology-SA-24:20, published 5 Nov 2024 | Apache Axis announcement on the java-user list, 6 Nov 2026 | | Fix | Fixed releases listed, for example DSM 7.2.2-72806-1 or above | "The Apache Axis project does not expect to create an Axis 1.x release fixing this problem." A source patch (a commit) exists | | Vendor mitigation | "Mitigation None" | None stated; recommends migrating to another SOAP engine | | What "mitigations per vendor instructions" comes to | Apply the vendor's update | Migrate, or build Axis with the published patch | | When discontinuing applies | Only if the update cannot be installed | The product is end of life, so for most users it is the route | The clause is a template. For Synology, the vendor's only instruction is an update, so the three-day clock is an update clock. For Axis there is no vendor release, so the choice is between a patched build you own and retiring the component. Neither entry offers a mitigation in the sense of a control that leaves the flaw in place. ## Choosing, and who decides 1. Open the source in the entry's notes field. Every one of the 815 entries carries text there: the vendor advisory or CISA guidance the action refers to. 2. Find your version in that source. If it lists a fixed release or a vendor workaround, that is the action, and the due date is the date to meet. Record the version before and after. 3. If no fix or workaround exists for your version, choose between a control you own with a dated fix plan (mitigation when no patch exists, [no patch yet](https://remediationvsmitigation.com/decision-table/kev-no-patch-yet/)) and removing the product from service. Discontinuing is closer to the avoid response in NIST SP 800-40 Rev. 4: "uninstalling the vulnerable software, decommissioning assets with the vulnerabilities, or disabling computing capabilities". 4. If the date cannot be met, the route is to remove the asset from the network; write the exception down. BOD 26-04 requires each agency's policy to "Assign roles and responsibilities for executing agency actions" and to "Establish internal validation and enforcement procedures". The deciding person is therefore named by the organization's own policy, usually the system owner with the risk owner signing any interim control. The record left behind holds the entry's date added and due date, the source read, the version, the action taken and its date, or the exception and its end date. ## Sources - CISA KEV JSON feed, catalogVersion 2026.11.18: https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) - CISA BOD 26-04, Required Actions: https://www.cisa.gov/news-events/directives/bod-26-04-prioritizing-security-updates-based-risk (read 19 Nov 2026) - Synology advisory Synology-SA-24:20: https://www.synology.com/en-us/security/advisory/Synology_SA_24_20 (read 19 Nov 2026) - Apache Axis project announcement of CVE-2026-90436, 6 Nov 2026: https://lists.apache.org/thread/8slf0j9gz35z56t0glt5h0ncwlbn4bdz (read 19 Nov 2026) - NIST SP 800-40 Rev. 4, Section 2.1: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Exploited Flaw, No Patch Yet: Mitigate and Set Two Dates](https://remediationvsmitigation.com/decision-table/kev-no-patch-yet/index.md) - [KEV Entry, Patch Out, Internet-Exposed: 3 or 14 Days](https://remediationvsmitigation.com/decision-table/kev-patch-available-exposed/index.md) --- # Mitigate vs manage risk Canonical: https://remediationvsmitigation.com/guides/mitigate-vs-manage-risk/ Managing risk is the process NIST describes as framing, assessing, responding to and monitoring risk. Mitigating is one response inside it, beside accepting, avoiding, sharing and transferring. SP 800-40 Rev. 4 uses four responses for software flaws. One Cisco ISE flaw, CVE-2026-76460 (16 Sep 2026), is carried through every step below. ## Managing is the process, mitigating is one response NIST SP 800-39 names the process by its four parts: "frame, assess, respond to, and monitor risk". The glossary entry for risk response lists what the respond part can choose: "Accepting, avoiding, mitigating, sharing, or transferring risk to organizational operations, organizational assets, individuals, other organizations, or the Nation." Risk mitigation is defined in SP 800-30 Rev. 1 as "Prioritizing, evaluating, and implementing the appropriate risk-reducing controls/countermeasures recommended from the risk management process. A subset of Risk Response." ([risk response](https://csrc.nist.gov/glossary/term/risk_response), [risk mitigation](https://csrc.nist.gov/glossary/term/risk_mitigation)) So the two words sit at different levels. "Manage" covers all four steps and every response. "Mitigate" is a choice made in step three. A register that has a column called "risk management" next to a column called "mitigation" is mixing a process with one of its outputs. ## The four responses for a software vulnerability SP 800-40 Rev. 4, Section 2.1, reduces the five to four. It also says that "By default, an organization accepts the risk posed by using its software." NIST SP 800-40 Rev. 4, Section 2.1, https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf, read 19 Nov 2026. | Response | SP 800-40 Rev. 4 wording | What it looks like on a finding | |---|---|---| | Accept | "Accept the risk from vulnerable software as is, such as by relying on existing security controls to prevent vulnerability exploitation or by determining that the potential impact is low enough that no additional action is needed" | A signed, dated [exception](https://remediationvsmitigation.com/risk-acceptance-form/) | | Mitigate | "Reduce the risk by eliminating the vulnerabilities (e.g., patching the vulnerable software, disabling a vulnerable feature, or upgrading to a newer software version without the vulnerabilities) and/or deploying additional security controls to reduce vulnerability exploitation" | A patch, a disabled feature or a control placed in front | | Transfer | "Reduce the risk by sharing some of the consequences with another party, such as by purchasing cybersecurity insurance or by replacing conventional software installations with software-as-a-service (SaaS) usage where the SaaS vendor/managed service provider takes care of patching" | Insurance, or a move to a hosted service; see [risk transfer vs risk acceptance](https://remediationvsmitigation.com/guides/risk-acceptance-vs-risk-transfer/) | | Avoid | "Ensure that the risk does not occur by eliminating the attack surface, such as by uninstalling the vulnerable software, decommissioning assets with the vulnerabilities, or disabling computing capabilities in assets that can function without them" | Uninstall or decommission; see the [decommission row](https://remediationvsmitigation.com/decision-table/decommission-scheduled/) | Inside this frame, patching is mitigation of the strongest kind: the publication says installing a patch or upgrading "are the only forms of risk response that can completely eliminate the vulnerabilities without removing functionality". ## One finding taken through all four steps On 16 Sep 2026 Cisco published advisory cisco-sa-ISE-ABP-VNSW7Tn5 for CVE-2026-76460, an authentication bypass in an Identity Services Engine (ISE) API. Cisco rates it CVSS base 10.0 and says its PSIRT is aware of active exploitation. CISA added it to KEV the same day with a due date of 19 Sep 2026. The advisory header reads "No workarounds available", but its Workarounds section adds: "However, there is a mitigation", which is to use infrastructure access control lists (iACLs) "to allow only required management and control plane traffic that is destined to the affected device". The estate below is invented: four ISE nodes, two on 3.3 Patch 11 and two on 3.4 Patch 6. Both are affected, one patch below the fixed releases in Cisco's advisory (3.3 Patch 12 and 3.4 Patch 7), and NVD lists them as the highest affected patches. The vendor and CISA dates are real. Dates for Cisco and CISA from the advisory and the KEV entry (CISA feed version 2026.11.18), read 19 Nov 2026. Estate and owners are illustrative. | Date | Step | Owner | Record it leaves | How it usually fails | |---|---|---|---|---| | Before 16 Sep | Frame: the tolerance that the later decision uses | Risk owner | [Acceptance criteria](https://remediationvsmitigation.com/guides/risk-acceptance-criteria/) approved earlier: known-exploited flaws on a network-reachable system are never accepted without a compensating control | Tolerance written after the finding, so it is fitted to the answer wanted | | 16 Sep | Assess: which nodes, how exposed | Vulnerability analyst | Four-node list with build, and the advisory ID. Every unpatched node is affected whatever its configuration, so the question is who can reach the API | Version list built but reachability left unanswered | | 16 Sep | Respond, decide: mitigate in two parts, iACL now and patch by 19 Sep. Accept is out (exploited, a due date exists). Avoid is out (the nodes enforce network access). Transfer is out (no hosted service can replace them in three days) | Risk owner signs | Decision record: response, control, two dates, signer | Choice made in a chat with no signer or end date | | 16 Sep | Respond, implement the iACL | Network engineer | Change ticket with the ACL text; a test from a network not on the list showing the API no longer answers. Cost: any admin station or integration not listed loses access | ACL written, never tested, or it blocks an integration that was never inventoried | | 17 to 18 Sep | Respond, implement the fix: 3.3 Patch 12 and 3.4 Patch 7, two nodes at a time | ISE administrator | Build output per node after restart | One node missed because it sat in a lab VLAN | | 18 Sep | Monitor: look for signs of compromise using Cisco's log-review steps | Security operations | Review note; if compromise is suspected Cisco recommends re-imaging the node | Patching treated as proof the node was clean | | 19 Sep | Monitor: close the remediation, keep the ACL as hardening | Risk owner | Register line closed with four build numbers; residual risk noted on the [residual risk page](https://remediationvsmitigation.com/guides/residual-risk-after-mitigation/) | Closing on the change ticket, not on the build output | Cisco states the policy that makes this two-part: "Cisco considers any workarounds and mitigations (if applicable) to be temporary solutions until an upgrade to a fixed software release is available." The iACL is therefore a bridge with its own end date, not a closing state. See [verifying a mitigation works](https://remediationvsmitigation.com/guides/verifying-a-mitigation-works/) for the test in the fourth row. ## Writing the word in a register SP 800-40 counts patching as mitigation; BOD 26-04 and FedRAMP's 2026 rules keep the words apart. In a register that auditors in either regime will read, put "remediate" on the line for a fix and "mitigate" on the line for a control, and never mark a finding closed on a mitigation alone. The quotations are in the [difference guide](https://remediationvsmitigation.com/guides/difference-between-remediation-and-mitigation/). ## Tactical and strategic SP 800-39 adds a time scale. "Some risk response measures are tactical in nature (e.g., applying patches to identified vulnerabilities in organizational information systems) and may be implemented rather quickly." Lower-priority risks are not ignored, but "fewer resources might be directed" to them or they might be "addressed at a later time". The ISE case is tactical. The tolerance set in step one is the strategic part, and no single finding settles it. Inherent and residual risk on the record are covered in [inherent versus residual risk](https://remediationvsmitigation.com/guides/inherent-vs-residual-risk/). ## Sources - NIST CSRC glossary, risk response: https://csrc.nist.gov/glossary/term/risk_response (read 19 Nov 2026) - NIST CSRC glossary, risk mitigation (SP 800-30 Rev. 1): https://csrc.nist.gov/glossary/term/risk_mitigation (read 19 Nov 2026) - NIST SP 800-40 Rev. 4, Section 2.1: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf (read 19 Nov 2026) - NIST SP 800-39, Tasks 3-3 and 3-4: https://nvlpubs.nist.gov/nistpubs/Legacy/SP/nistspecialpublication800-39.pdf (read 19 Nov 2026) - Cisco advisory cisco-sa-ISE-ABP-VNSW7Tn5 (CVE-2026-76460): https://sec.cloudapps.cisco.com/security/center/content/CiscoSecurityAdvisory/cisco-sa-ISE-ABP-VNSW7Tn5 (read 19 Nov 2026) - CISA KEV entry for CVE-2026-76460: https://www.cisa.gov/known-exploited-vulnerabilities-catalog?field_cve=CVE-2026-76460 (read 19 Nov 2026) - NVD record, CVE-2026-76460 (affected patch levels): https://services.nvd.nist.gov/rest/json/cves/2.0?cveId=CVE-2026-76460 (read 19 Nov 2026) - CISA KEV feed, catalogVersion 2026.11.18: https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Difference Between Remediation and Mitigation in IT Security](https://remediationvsmitigation.com/guides/difference-between-remediation-and-mitigation/index.md) - [Risk Acceptance Criteria: 7 Tests From NIST and CISA](https://remediationvsmitigation.com/guides/risk-acceptance-criteria/index.md) - [Decision Table: Remediate, Mitigate or Accept by Situation](https://remediationvsmitigation.com/decision-table/index.md) - [Risk Acceptance Form: Free Template With 15 Fields](https://remediationvsmitigation.com/risk-acceptance-form/index.md) - [Residual Risk After Mitigation: Define, Estimate, Record](https://remediationvsmitigation.com/guides/residual-risk-after-mitigation/index.md) - [Inherent vs Residual Risk on a Security Decision Record](https://remediationvsmitigation.com/guides/inherent-vs-residual-risk/index.md) - [Risk Transfer vs Risk Acceptance for Security Findings](https://remediationvsmitigation.com/guides/risk-acceptance-vs-risk-transfer/index.md) --- # Vulnerability mitigation in cloud and SaaS Canonical: https://remediationvsmitigation.com/guides/mitigation-in-cloud-and-saas/ In SaaS the vendor patches; the customer keeps data, endpoints, accounts, access and application configuration. In Microsoft's October 2026 release, 13 of 3,873 CVE records state that the issue was already fully mitigated and customers need take no action. For tenant settings, CISA BOD 25-01 made baselines mandatory for federal agencies by 20 Jun 2025. ## What stays with the customer Microsoft's shared responsibility page describes SaaS as "You use ready-made applications. Examples include Microsoft 365, Dynamics 365, and other cloud applications." It lists four responsibilities the customer "always" keeps: data, endpoints, accounts, and access management, including "role-based access control (RBAC), multifactor authentication, and conditional access policies". For applications: "In PaaS and SaaS, Microsoft manages parts of the application stack, but you're responsible for application configuration, code security, and access controls." For networks: "In SaaS, Microsoft manages network security." When a SaaS flaw has no customer-side patch, those are the levers: who can reach the tenant, which accounts hold privilege and which features are on. The [SaaS row of the decision table](https://remediationvsmitigation.com/decision-table/saas-cannot-patch/) lists the duties by party. The controls themselves are described under [access restriction](https://remediationvsmitigation.com/compensating-controls/access-restriction/) and [configuration change](https://remediationvsmitigation.com/compensating-controls/configuration-change/). ## Five steps for a customer Procedure written for this page from the sources listed; owners are illustrative. | Step | Who | Record it leaves | How it usually fails | |---|---|---|---| | 1. Keep a list of the SaaS services in scope and the vendor's advisory feed for each | Security lead | List with feed URL and a named reader | Scanners never see a vendor's own service, so with no watcher a cloud CVE reaches nobody | | 2. Read the vendor's customer-action field for each new record | Vulnerability analyst | Register line quoting the field and the date | Closing on the title alone. In Microsoft's October 2026 release the field reads Yes on 3,860 records, so the title cannot tell you | | 3. If no action: close with the vendor's statement. If action: apply the customer-side lever | Analyst, then identity or tenant administrator | Closure quoting the statement; or a change ticket naming the setting | Copying the vendor's word "mitigated" into the register as if it were your control | | 4. Compare tenant settings with a baseline | Tenant administrator | Assessment output with date | Running the tool once, not on a schedule | | 5. Put each deviation on the exception register with an approver and an end date | Risk owner | [Exception](https://remediationvsmitigation.com/exception-register/) line | Deviations tolerated with no end date | ## When the vendor has already fixed it: the 13 records Microsoft's CVRF document for the October 2026 release (3,873 vulnerability records; document first released 13 Oct 2026, current revision 18 Nov 2026) holds 13 records with the FAQ: "This vulnerability has already been fully mitigated by Microsoft. There is no action for users of this service to take. The purpose of this CVE is to provide further transparency." Each of the 13 carries the field Customer Action Required: No and the exploitability line Exploited:No. They are the only records of the 3,873 with that field set to No; the other 3,860 read Yes. The release is large because 1,646 of those records list Azure Linux packages. Microsoft MSRC CVRF API, October 2026 security updates, read 19 Nov 2026. CVSS values are the base scores in the records; the date is the Information published entry of each record's revision history. | CVE | Service and flaw type | CVSS 3.1 base | Record published | |---|---|---|---| | CVE-2026-105014 | Azure Pipelines, remote code execution | 9.1 | 22 Oct 2026 | | CVE-2026-108456 | Azure AI Foundry, elevation of privilege | 7.7 | 22 Oct 2026 | | CVE-2026-110063 | VM Inspector for Azure Virtual Machines, elevation of privilege | 10.0 | 22 Oct 2026 | | CVE-2026-69435 | Azure SRE Agent, elevation of privilege | 9.6 | 8 Oct 2026 | | CVE-2026-77900 | Azure App Service, remote code execution | 9.8 | 8 Oct 2026 | | CVE-2026-83938 | Azure Kubernetes Configuration Service, elevation of privilege | 10.0 | 22 Oct 2026 | | CVE-2026-83943 | Azure API Center, information disclosure | 8.7 | 8 Oct 2026 | | CVE-2026-83947 | Azure Event Grid, spoofing | 7.7 | 8 Oct 2026 | | CVE-2026-88131 | Microsoft Dataverse, remote code execution | 9.8 | 8 Oct 2026 | | CVE-2026-94510 | Microsoft Bookings, elevation of privilege | 9.9 | 8 Oct 2026 | | CVE-2026-96185 | Microsoft Exchange Online, remote code execution | 7.5 | 22 Oct 2026 | | CVE-2026-96207 | Microsoft Partner Center, elevation of privilege | 10.0 | 8 Oct 2026 | | CVE-2026-97785 | AAD App Proxy, elevation of privilege | 10.0 | 22 Oct 2026 | Cloud records are published on their own dates and are collected into the monthly document. Seven of the 13 carry a publication date of 8 Oct 2026 and six of 22 Oct 2026, so the dates fall before and after the document's 13 Oct release. Several are rated 10.0 and none was exploited, which shows why the severity alone cannot set the customer's work: for these records the work is a closure with evidence. Note the vendors' own words differ. Microsoft says "mitigated" and does not say whether the flaw was eliminated or contained. Fortinet's FortiMail advisory FG-IR-26-175 says it "has remediated this issue" on 5 Oct 2026 in its cloud service. FedRAMP's rules keep the words apart: "A fully mitigated vulnerability will still exist (with negligible risk) until it has been remediated." That sentence is about a provider tracking its own service. A customer holds no control in these 13 cases, so no residual risk of the customer's own exists to track. Record the line as vendor-resolved, quote the vendor, and do not enter a customer mitigation that was never applied. ### Register entry: vendor-resolved cloud vulnerability - Record: CVE-2026-96185, Microsoft Exchange Online, remote code execution, CVSS 3.1 base 7.5 - Source: Microsoft MSRC CVRF, October 2026 release, first revision 22 Oct 2026 - Vendor statement (quoted): "This vulnerability has already been fully mitigated by Microsoft. There is no action for users of this service to take." - Vendor fields: Customer Action Required: No. Exploited: No. Publicly Disclosed: No. - Our exposure: tenant uses Exchange Online (yes/no), checked by: [name], date: [date] - Customer-side control in play: none - Status: Closed, vendor-resolved, no customer action - Reviewer: [name], date: [date] - Reopen if: Microsoft revises the record, or the field Customer Action Required changes ## When the customer must act: a record that reads Yes Microsoft's October 2026 release also holds CVE-2026-100207, an OS command injection in Azure CycleCloud (CVSS 3.1 base 7.3, Important, first released 13 Oct 2026, Exploited: No, Publicly Disclosed: No). The record says Customer Action Required: Yes and lists CycleCloud 8.9.1 as affected with the security update 8.9.2. Microsoft's FAQ gives the conditions: the attacker must be authenticated to CycleCloud with permission to create or manage a cluster and to configure a Windows node's credentials, and a different user must then connect to that node with the CycleCloud CLI from a macOS or Linux workstation. The customer runs this software, so the vendor's fix does nothing until the customer installs it. Here the work is a version check, an upgrade and, until then, a look at who holds cluster-management permission. ### Register entry: customer-action cloud vulnerability - Record: CVE-2026-100207, Azure CycleCloud, OS command injection, CVSS 3.1 base 7.3 - Source: Microsoft MSRC CVRF, October 2026 release, first released 13 Oct 2026 - Vendor fields: Customer Action Required: Yes. Exploited: No. Publicly Disclosed: No. Fixed in: 8.9.2 - Our exposure: CycleCloud installed (yes/no), version [version], checked by: [name], date: [date] - Who holds permission to create or manage clusters: [list], reviewed by: [name], date: [date] - Customer-side control until upgraded: [none, or the permission change made] - Response: Remediate, upgrade to 8.9.2 by [date]; owner: [name] - Closure evidence: installed version after the upgrade, change ticket - Reopen if: a new revision of the record, or the entry appears on the KEV catalog ## Configuration as a mandated mitigation CISA's BOD 25-01 treats a configuration baseline as the control for SaaS. Its background says "Outdated security configurations expose systems to exploits that can be easily mitigated by recommended and mandatory security configurations." It covers production tenants with a finalized SCuBA Secure Configuration Baseline; at issuance only Microsoft Office 365 had one. Agencies had to deploy the assessment tools by 25 Apr 2025 and implement all mandatory policies by 20 Jun 2025. Authorizing officials "may accept risk for deviations from the mandatory SCuBA policies to account for operational needs", and agencies "shall identify and explain deviations in the output of the SCuBA assessment tools". CISA describes ScubaGear as "a no-cost assessment tool that verifies M365 tenant configuration alignment to the policies described in SCuBA's secure configuration baselines", with ScubaGoggles as the Google Workspace counterpart. Both check a tenant you administer. A deviation is an exception in the sense of step five, so it goes on the [exception register](https://remediationvsmitigation.com/exception-register/) with the authorizing official as approver and an end date. ## KEV wording for hosted products In the KEV feed (catalogVersion 2026.11.18, read 19 Nov 2026), 329 entries carry the mitigate-or-discontinue string that tells readers to "follow applicable BOD 22-01 guidance for cloud services", and 11 more say "Adhere to the applicable BOD 22-01 guidance for cloud services". The 169 newest say "Follow applicable BOD 26-04 guidance for cloud services or discontinue use of the product if mitigations are unavailable". Counts by wording are on the [KEV mitigate-or-discontinue guide](https://remediationvsmitigation.com/guides/kev-required-actions-mitigate-or-discontinue/). For a customer, discontinuing may mean switching off one integration rather than leaving the vendor. ## Sources - Microsoft Learn, shared responsibility in the cloud: https://learn.microsoft.com/en-us/azure/security/fundamentals/shared-responsibility (read 19 Nov 2026) - Microsoft MSRC CVRF API, October 2026 security updates: https://api.msrc.microsoft.com/cvrf/v3.0/cvrf/2026-Oct (read 19 Nov 2026) - CISA BOD 25-01: https://www.cisa.gov/news-events/directives/bod-25-01-implementing-secure-practices-cloud-services (read 19 Nov 2026) - CISA SCuBA project: https://www.cisa.gov/resources-tools/services/secure-cloud-business-applications-scuba-project (read 19 Nov 2026) - FedRAMP 2026 rules, VDR-CSO-RES notes: https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-detection-and-response/ (read 19 Nov 2026) - Fortinet PSIRT advisory FG-IR-26-175: https://fortiguard.fortinet.com/psirt/FG-IR-26-175 (read 19 Nov 2026) - CISA KEV JSON feed, catalogVersion 2026.11.18: https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [SaaS You Cannot Patch: Mitigate and Track the Vendor](https://remediationvsmitigation.com/decision-table/saas-cannot-patch/index.md) - [Access Restriction as a Compensating Control: PAN-OS](https://remediationvsmitigation.com/compensating-controls/access-restriction/index.md) - [Configuration Change as a Mitigation: Cisco ASA Case](https://remediationvsmitigation.com/compensating-controls/configuration-change/index.md) - [Mitigation vs Remediation Wording in Vendor Advisories](https://remediationvsmitigation.com/guides/mitigation-vs-remediation-in-vendor-advisories/index.md) - [Exception Register: 15 Columns, States and Blank CSV](https://remediationvsmitigation.com/exception-register/index.md) - [KEV Required Action: Mitigate or Discontinue Use](https://remediationvsmitigation.com/guides/kev-required-actions-mitigate-or-discontinue/index.md) - [Risk Transfer vs Risk Acceptance for Security Findings](https://remediationvsmitigation.com/guides/risk-acceptance-vs-risk-transfer/index.md) --- # Mitigation vs remediation in vendor advisories Canonical: https://remediationvsmitigation.com/guides/mitigation-vs-remediation-in-vendor-advisories/ Vendors use the same words for different things. OASIS CSAF 2.0 defines five remediation categories. Of 50 Fortinet advisories read on 19 Nov 2026, 47 list a vendor fix and 1 also lists a mitigation. Fifty Red Hat errata files for a single CVE all list a fix and a workaround, and none uses the mitigation category. ## The shared vocabulary: CSAF 2.0 OASIS CSAF Version 2.0, Section 3.2.3.12.1, gives machine-readable advisories five remediation categories: workaround, mitigation, vendor_fix, none_available and no_fix_planned. Each is defined below in the standard's words. OASIS CSAF Version 2.0, Section 3.2.3.12.1, https://docs.oasis-open.org/csaf/csaf/v2.0/os/csaf-v2.0-os.html, read 19 Nov 2026. | Category | CSAF 2.0 meaning | |---|---| | workaround | "a configuration or specific deployment scenario that can be used to avoid exposure to the vulnerability"; typically the "first line of defense" before a mitigation or fix | | mitigation | "helps to reduce the risk of the vulnerability but that does not resolve the vulnerability on the affected product"; may use "devices or access controls external to the affected product" | | vendor_fix | "an official fix that is issued by the original author of the affected product", assumed to fully resolve it | | none_available | "currently no fix or other remediation available" | | no_fix_planned | "no fix for the vulnerability and it is not planned to provide one at any time", often because the product is "orphaned, declared end-of-life, or otherwise deprecated" | The specification makes vendor_fix exclusive with none_available and no_fix_planned for the same product. An advisory that lists both a fix and a mitigation for one product is telling the reader what to do until the fix is installed. The prose side, how twelve advisories word workaround and mitigation, is on [workaround versus mitigation](https://remediationvsmitigation.com/guides/workaround-vs-mitigation/). This page covers the machine-readable categories and what to record from them. ## Advisories where the label and the action differ The rows below ask who acts and what a label tells the reader to record. Advisories and Microsoft's October 2026 CVRF file read 19 Nov 2026; sources are listed below. Dates are the vendors' publication dates, except the Fortinet cloud date, which is the date the advisory gives for the vendor's own change. | Vendor and record | Wording in the advisory | What the label tells the reader | |---|---|---| | Cisco ISE, CVE-2026-76460 (16 Sep 2026) | Header: "No workarounds available". Body: "There are no workarounds that address this vulnerability. However, there is a mitigation." | The header counts only steps that avoid exposure; the infrastructure access control list that limits who can reach the flaw is filed as a mitigation. The case with owners and dates is on [mitigate versus manage risk](https://remediationvsmitigation.com/guides/mitigate-vs-manage-risk/) | | Fortinet FortiMail, FG-IR-26-175 (1 Oct 2026) | "Fortinet has remediated this issue on oct 05, 2026 in FortiMail Cloud", so "FortiMail Cloud customers need not perform any action"; self-managed units get a workaround | Remediated is the vendor's own act on its hosted service; the workaround is the customer's step elsewhere | | Microsoft Azure Pipelines, CVE-2026-105014 (published 22 Oct 2026) | "This vulnerability has already been fully mitigated by Microsoft. There is no action for users of this service to take." The entry carries no remediation items. 13 of the 3,876 entries in the October 2026 CVRF file carry the same sentence. | Vendor-side resolution with no customer step and no fix to install; see [mitigation in cloud and SaaS](https://remediationvsmitigation.com/guides/mitigation-in-cloud-and-saas/) | Cisco's Fixed Software section states its own policy: "Cisco considers any workarounds and mitigations (if applicable) to be temporary solutions until an upgrade to a fixed software release is available." Control first, fix on a date, control removed. ## What 50 Fortinet advisories show Fortinet publishes a CSAF file per advisory, linked from its PSIRT RSS feed. The 50 advisories in the feed on 19 Nov 2026 split as follows by remediation category. Fortinet PSIRT RSS feed https://filestore.fortinet.com/fortiguard/rss/ir.xml and the 50 linked CSAF files, read 19 Nov 2026. | Remediation categories in the file | Advisories | |---|---| | vendor_fix only | 46 | | vendor_fix and mitigation | 1 | | No vulnerabilities block | 3 | | Total | 50 | So 47 advisories list a vendor_fix, and the one with a mitigation is among them. It is FG-IR-26-273, a FortiOS flaw in high availability traffic. Its mitigation entry reads "A virtual patch is available" and names an IPS signature package, while the vendor_fix entry lists FortiOS 8.0.1, 7.6.8, 7.4.13 and 7.2.14 and says "FortiOS 7.0: Migrate to a fixed release". A signature that blocks traffic is filed as a mitigation, not a fix, which matches the [IPS signature entry](https://remediationvsmitigation.com/compensating-controls/ips-signature/). ## What 50 Red Hat errata for one CVE show Red Hat publishes one CSAF file per security advisory. The 50 files counted here are RHSA errata first released between 22 Oct and 13 Nov 2026, and every one of them carries the same CVE, CVE-2026-116591 (a zstd decoder flaw, shipped in many images). They show how Red Hat files one flaw, not its mix across flaws, so no comparison with the Fortinet count is drawn. Red Hat CSAF files for CVE-2026-116591 listed in https://security.access.redhat.com/data/csaf/v2/advisories/changes.csv, read 19 Nov 2026. | Remediation categories in the file | Advisories | |---|---| | vendor_fix and workaround | 50 | | mitigation | 0 | | Total | 50 | The two vendors file steps under different categories. Fortinet puts a virtual patch under mitigation. Red Hat's files use workaround for text such as "Mitigation for this issue is either not available or the currently available options do not meet the Red Hat Product Security criteria comprising ease of use and deployment, applicability to widespread installation base, or stability", and for configuration steps that begin "To mitigate this issue". A script that counts only the mitigation category finds one Fortinet advisory and none of the Red Hat files, so read the text of every remediation entry. ## One advisory worked into a record RHSA-2026:88185 (Red Hat Hardened Images RPMs, first released 22 Oct 2026, rated Important) covers CVE-2026-116591, a denial of service in the klauspost/compress zstd decoder (CVSS 3.1 base score 7.5 in Red Hat's file). Its CSAF file carries a vendor_fix entry pointing to the errata and a workaround entry. The workaround entry says: "Reject or filter untrusted Zstandard-compressed payloads at ingress proxies, or restrict decompression endpoints to authenticated, trusted clients." The entry also carries guidance dated 3 Nov 2026 whose label says it was generated with AI tools and that customers should "independently evaluate and test these temporary mitigations". Using the reading rule below, the record for a team that runs an affected image reads: Advisory facts from Red Hat's CSAF file for RHSA-2026:88185, read 19 Nov 2026. The control, test and removal rows are the record a team would write. | Record field | Entry | |---|---| | Vendor labels | vendor_fix (RHSA-2026:88185) and workaround | | Treat as | Mitigate now, fix on a date, because both categories are present | | Control | Ingress proxy rejects or filters untrusted zstd-compressed request bodies; the vendor text does not name a product setting, so the team writes its own | | What it leaves open | Applications that decompress zstd data fetched from elsewhere, and any path that does not pass the proxy | | Test | Dated test by a second person that an untrusted zstd request is refused on the production path | | Fix | Install the fixed Red Hat build from RHSA-2026:88185 and record the version and date; Red Hat's text names klauspost/compress 1.20.2 or later as the upstream fix | | Removal | Delete the proxy rule on the day the fixed build is confirmed in production | ## Reading rule for a decision record Mapping drawn up for this page; the labels are the vendors'. For industrial products see ICS vulnerability mitigation. | Vendor label | Treat as | Record | |---|---|---| | vendor_fix or Vendor fix | Remediate | Version installed and date | | mitigation or workaround | Mitigate now, fix on a date | Control, tester and the date the fix is due | | none_available | Mitigate, track the vendor | Last vendor contact date | | "Migrate to a fixed release" | Remediate by moving to the fixed branch | Target version and migration date | | no_fix_planned | Replace, remove or accept with a dated exception | Replacement date or signed exception | ## Sources - OASIS CSAF Version 2.0, Section 3.2.3.12.1: https://docs.oasis-open.org/csaf/csaf/v2.0/os/csaf-v2.0-os.html (read 19 Nov 2026) - Cisco advisory cisco-sa-ISE-ABP-VNSW7Tn5: https://sec.cloudapps.cisco.com/security/center/content/CiscoSecurityAdvisory/cisco-sa-ISE-ABP-VNSW7Tn5 (read 19 Nov 2026) - Fortinet PSIRT advisory FG-IR-26-175 (published 1 Oct 2026; cloud remediation 5 Oct 2026): https://fortiguard.fortinet.com/psirt/FG-IR-26-175 (read 19 Nov 2026) - Fortinet CVRF file for FG-IR-26-175: https://fortiguard.fortinet.com/psirt/cvrf/FG-IR-26-175 (read 19 Nov 2026) - Fortinet PSIRT RSS feed and 50 linked CSAF files: https://filestore.fortinet.com/fortiguard/rss/ir.xml (read 19 Nov 2026) - Microsoft MSRC CVRF API, October 2026: https://api.msrc.microsoft.com/cvrf/v3.0/cvrf/2026-Oct (read 19 Nov 2026) - Red Hat CSAF changes.csv and the 50 first-listed RHSA CSAF files: https://security.access.redhat.com/data/csaf/v2/advisories/changes.csv (read 19 Nov 2026) - Red Hat CSAF file for RHSA-2026:88185 (CVE-2026-116591): https://security.access.redhat.com/data/csaf/v2/advisories/2026/rhsa-2026_88185.json (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Workaround vs Mitigation in Real Vendor Advisories](https://remediationvsmitigation.com/guides/workaround-vs-mitigation/index.md) - [IPS Signature as a Mitigation: When It Counts](https://remediationvsmitigation.com/compensating-controls/ips-signature/index.md) - [SaaS Vulnerability Mitigation: Steps for the Customer](https://remediationvsmitigation.com/guides/mitigation-in-cloud-and-saas/index.md) - [Mitigating Risk vs Managing Risk, With a Worked Case](https://remediationvsmitigation.com/guides/mitigate-vs-manage-risk/index.md) --- # Plan of action and milestones (POA&M) Canonical: https://remediationvsmitigation.com/guides/poam-plan-of-action-and-milestones/ A plan of action and milestones entry is required when a mitigation cannot be completed immediately (NIST SP 800-53 RA-7), and it holds tasks, resources, milestones and scheduled completion dates (SP 800-37 Task A-6). The filled entry tracks CVE-2026-19868, on CISA's KEV catalog since 6 Nov 2026 with a 20 Nov due date, fixed in SonicOS 7.3.4-7013. ## When a finding gets an entry NIST SP 800-53 CA-5 asks the organization to "Develop a plan of action and milestones for the system to document the planned remediation actions of the organization to correct weaknesses or deficiencies noted during the assessment of the controls and to reduce or eliminate known vulnerabilities in the system". Its guidance adds that the plans "are required in authorization packages and subject to federal reporting requirements established by OMB". PM-4 notes that "While plans of action and milestones are required for federal organizations, other types of organizations can help reduce risk by documenting and tracking planned remediations." Quotations from NIST SP 800-53 Rev. 5.2.0, SP 800-37 Rev. 2 and FedRAMP's 2026 agency page, read 19 Nov 2026. | Situation | Entry? | Where the finding lives instead | Text | |---|---|---|---| | Fixed or mitigated the same day | No | The ticket and the closed finding | RA-7: the mitigation may be done "immediately so that a plan of action and milestones entry is not needed" | | Mitigation cannot be completed immediately | Yes | The plan | RA-7: "a plan of action and milestones entry is generated" | | Interim control in place, fix waits on the vendor or a change window | Yes | The plan, with the exception record if the gap is accepted | CA-5: planned remediation of weaknesses | | Authorizing official accepts the deficiency as residual risk | Not necessary | The assessment report and the exception record | SP 800-37 Task A-6: entries "are not necessary when deficiencies are accepted by the authorizing official as residual risk" | | Provider-reported vulnerability and the agency has nothing to do | No | The provider's own vulnerability records | FedRAMP 2026: "Agencies should not copy every provider-reported vulnerability into an agency POA&M." | The [exception register](https://remediationvsmitigation.com/exception-register/) holds the accepted rows and the plan holds the rows waiting for a fix, which is why the two documents are kept apart in the [risk register comparison](https://remediationvsmitigation.com/guides/risk-register-vs-exception-register/). ## The fields of an entry SP 800-37 Task A-6 names what the plan holds: "tasks to be accomplished with a recommendation for completion before or after system authorization; resources required to accomplish the tasks; milestones established to meet the tasks; and the scheduled completion dates for the milestones and tasks". The system owner or common control provider has primary responsibility for the task. The FedRAMP POA&M template adds columns for the CVE and the KEV due date. The table says which field comes from which text. Fields of a plan of action and milestones entry and the text behind each, read 19 Nov 2026. Rows without a quoted basis are this site's practice, not a requirement. | Field | What goes in | Basis | |---|---|---| | Entry ID and CVE | One ID per finding and asset group; every CVE it covers | FedRAMP template column AD, "CVE", added 27 Jun 2022 | | Weakness and source of detection | The flaw in plain words, the scanner or feed that found it, the finding ID | FedRAMP legacy playbook: Column F, Weakness Detector Source | | Asset | Name, version, environment, exposure (public or internal) | BOD 26-04 asset tags: environment and exposure | | Owner | A role that answers for the entry and one owner per milestone | SP 800-37 Task A-6, primary responsibility: system owner | | Resources required | Change window length, people, money, vendor support | SP 800-37 Task A-6 | | Authority date | The KEV due date, when one exists | FedRAMP template column AC, named for BOD 22-01, which BOD 26-04 revoked on 10 Jun 2026 | | Milestones | Each task with a date, an owner and the record that proves it | SP 800-37 Task A-6: scheduled completion dates | | Scheduled completion date | The date the finding is expected to read as closed | SP 800-37 Task A-6 | | Milestone changes | Every change with the date and the reason | Keeps the original dates readable | | Vendor dependency | Yes or no, the product, the last vendor check-in date | FedRAMP legacy columns Q, R and S: check in "at least once a month" | | Status and status date | Open, in progress or closed, and when that was last true | SP 800-37: the plan "is subsequently used to monitor progress" | | Exception link | The exception ID, if the gap between mitigation and fix is accepted | This site's field, see the [Risk Acceptance Form](https://remediationvsmitigation.com/risk-acceptance-form/) | ## A filled entry for CVE-2026-19868 The finding is real. CISA added CVE-2026-19868 (SonicWall SonicOS SSLVPN, "Use of Single-factor Authentication Vulnerability", CWE-308) to the KEV catalog on 6 Nov 2026 with a due date of 20 Nov 2026 and the required action "Apply mitigations in accordance with vendor instructions". NVD describes it as a flaw that "may allow a user with valid credentials for an alternate authentication method to bypass SAML authentication controls". SonicWall advisory [SNWLID-2026-0014](https://psirt.global.sonicwall.com/vuln-detail/SNWLID-2026-0014) lists SonicOS Gen7 7.3.3-7015 and older and Gen8 8.2.2-8015 and older as affected, and names the fixed builds: Gen7 7.3.4-7013 and higher, Gen8 8.2.3-8015 and higher. (NVD's configuration data uses slightly different cut-offs, so the entry uses SonicWall's figures.) The flaw matters only where SAML is enabled for SSLVPN logins and at least one LDAP, RADIUS or TACACS+ server is active. The advisory was released on 5 Nov 2026 (version 1.0); version 1.1, dated 9 Nov 2026, updated the workaround: with SAML in use and no upgrade possible, disable or delete every active LDAP, RADIUS and TACACS+ server on the firewall. The cost is that anything on the firewall that signs in through those servers stops working until they are re-enabled. The firewall names, owners and every date after 19 Nov below are the example's own. **POA&M entry POAM-2026-0114 (example)** **Weakness:** SonicOS SSLVPN accepts a user's valid credentials for an alternate authentication method in place of SAML authentication, which can bypass SAML-enforced MFA. CVE-2026-19868, CWE-308. **Source of detection:** CISA KEV catalog (version 2026.11.18) matched to the asset inventory on 19 Nov 2026; vendor advisory SNWLID-2026-0014. **Original detection date:** 6 Nov 2026 (the day CISA added the CVE; BOD 26-04 starts the clock at KEV addition or the agency's own identification, whichever is first). **Asset:** fw-edge-a and fw-edge-b, SonicOS Gen7 (NSa 4700) build 7.3.3-7015, SAML SSLVPN in use, LDAP and RADIUS servers active, production, publicly exposed. **KEV due date:** 20 Nov 2026. Forensic triage flag in the feed: No. **Entry owner:** Network operations lead. **Reviewer:** Security operations lead. **Resources required:** one 2-hour change window, two engineers, the 7.3.4-7013 firmware image, vendor support contact. **Vendor dependency:** No. The fixed build exists: SonicOS 7.3.4-7013 or higher for Gen7. | # | Date | Milestone | Owner | Record that proves it | |---|---|---|---|---| | 1 | 19 Nov 2026 | Confirm the build and that SAML SSLVPN is enabled on both units; list the active LDAP and RADIUS servers and every login that depends on them (admin sign-ins included) and move those to local accounts first | Network operations lead | Configuration export and dependency list attached to the ticket | | 2 | Fri 20 Nov 2026 (KEV due date) | Apply SonicWall's workaround from SNWLID-2026-0014 version 1.1 on both units: disable or delete every active LDAP, RADIUS and TACACS+ server; a second person checks the server lists | Network operations lead; Security analyst | Change record and the checked server lists | | 3 | Fri 20 Nov 2026 | Compare SSLVPN login records since 6 Nov 2026 with the identity provider's sign-in log for logins that went through an alternate method | Security operations analyst | Saved log search and a one-line result | | 4 | Sat 5 Dec 2026, 02:00 to 04:00 | Upgrade both units to SonicOS 7.3.4-7013 or higher | Network operations lead | Change ticket and the version shown after the change | | 5 | Mon 7 Dec 2026 | Confirm 7.3.4-7013 or higher on both units, rescan, re-enable the LDAP and RADIUS servers, close | Security analyst | Scan ID, version output and close note | **Scheduled completion date:** 7 Dec 2026. **Milestone changes:** none (each change is logged with date and reason). **Status on 19 Nov 2026:** Open, in progress, milestone 1 under way. **Exception link:** none. If milestone 4 slips past 5 Dec 2026, open an exception record and enter its ID here. The date arithmetic is short. The KEV due date is 14 days after the 6 Nov addition. The upgrade in milestone 4 falls 29 days after the addition and 15 days after the due date, so for a federal asset the 5 Dec upgrade by itself misses the KEV date. Milestone 2 is what the entry relies on for 20 Nov: the required action is to apply mitigations in accordance with vendor instructions, and SonicWall's workaround is the vendor's instruction for a firewall that cannot be upgraded yet. An agency confirms that reading with the directive's contact before it depends on it. Milestone 5 closes the entry 31 days after the addition. The entry stores the KEV date and does not move it. Which timeline in [BOD 26-04](https://www.cisa.gov/news-events/directives/bod-26-04-prioritizing-security-updates-based-risk) Table 1 applies to an asset depends on exposure, KEV status, automatability and technical impact, so confirm that row before accepting a 5 Dec upgrade date. Milestone 2 is checked by a second person because [a mitigation has to be verified](https://remediationvsmitigation.com/guides/verifying-a-mitigation-works/), not assumed. ## Cases where one entry per finding is wrong - **Assets with different exposure.** BOD 26-04 applies its timeline "For each asset detected to contain an instance of a CVE ID", and a mitigation that removes an asset from the internet "will shift the required timeline". Two assets with one CVE and different exposure need separate milestone dates. - **Accepted risk.** With an acceptance signed by the authorizing official, SP 800-37 says no entry is necessary. The deficiency stays in the assessment report and the exception record carries it. - **Legacy FedRAMP combined risks.** The playbook (reference only since 24 Jun 2026) let assessors combine risks from unauthorized external services into one line of the risk exposure table, but providers "must create unique POA&M items" because "remediation plans and mitigating factors will likely differ for each service". - **Vendor dependencies.** The legacy playbook keeps them as open risks, asks for "evidence of vendor interactions" and requires High-risk vendor dependencies to be mitigated to Moderate within thirty days. See [FedRAMP deviation requests](https://remediationvsmitigation.com/guides/fedramp-deviation-requests/). - **Cloud provider findings.** FedRAMP's 2026 agency page limits agency entries to cases such as when "The agency needs to track a compensating control, contract action, monitoring activity, or other agency-owned response". For a service you cannot patch, see [SaaS you cannot patch](https://remediationvsmitigation.com/decision-table/saas-cannot-patch/). - **Operational technology.** NIST SP 800-82 Rev. 3 says to "take into account planned schedule maintenance or shutdowns of the OT system", so milestone dates follow outage windows, as in the [OT and medical row](https://remediationvsmitigation.com/decision-table/ot-medical-cannot-patch/). ## Sources - NIST SP 800-53 Rev. 5.2.0 OSCAL catalog, CA-5, CA-5(1), PM-4, RA-7: https://github.com/usnistgov/oscal-content/blob/main/nist.gov/SP800-53/rev5/json/NIST_SP-800-53_rev5_catalog.json (read 19 Nov 2026) - NIST SP 800-37 Rev. 2, Task A-6: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-37r2.pdf (read 19 Nov 2026) - CISA BOD 26-04, Required Actions and Appendix A: https://www.cisa.gov/news-events/directives/bod-26-04-prioritizing-security-updates-based-risk (read 19 Nov 2026) - CISA KEV catalog JSON feed, catalogVersion 2026.11.18, entry CVE-2026-19868: https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) - NVD record for CVE-2026-19868: https://services.nvd.nist.gov/rest/json/cves/2.0?cveId=CVE-2026-19868 (read 19 Nov 2026) - SonicWall PSIRT advisory SNWLID-2026-0014 (affected and fixed builds, workaround, revision history): https://psirt.global.sonicwall.com/vuln-detail/SNWLID-2026-0014 (read 19 Nov 2026) - FedRAMP 2026, managing agency POA&Ms: https://www.fedramp.gov/2026/agencies/use/ongoing/poams/ (read 19 Nov 2026) - FedRAMP legacy playbook, POA&M (reference only since 24 Jun 2026): https://www.fedramp.gov/legacy/playbook/csp/authorization/poam/ (read 19 Nov 2026) - FedRAMP, Update to the POA&M template, 27 Jun 2022: https://www.fedramp.gov/archive/2022-06-28-update-poam-template/ (read 19 Nov 2026) - NIST SP 800-82 Rev. 3, Task A-6: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-82r3.pdf (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Exception Register: 15 Columns, States and Blank CSV](https://remediationvsmitigation.com/exception-register/index.md) - [FedRAMP Deviation Requests and Their 2026 Replacements](https://remediationvsmitigation.com/guides/fedramp-deviation-requests/index.md) - [Risk Register vs Exception Register: How They Link](https://remediationvsmitigation.com/guides/risk-register-vs-exception-register/index.md) - [How to Verify a Mitigation Works: Checks for 5 Real CVEs](https://remediationvsmitigation.com/guides/verifying-a-mitigation-works/index.md) - [Recording an Exception an Auditor Accepts: Eight Items](https://remediationvsmitigation.com/guides/recording-an-exception-an-auditor-accepts/index.md) - [SaaS You Cannot Patch: Mitigate and Track the Vendor](https://remediationvsmitigation.com/decision-table/saas-cannot-patch/index.md) - [Risk Acceptance Form: Free Template With 15 Fields](https://remediationvsmitigation.com/risk-acceptance-form/index.md) --- # Recording an exception an auditor accepts Canonical: https://remediationvsmitigation.com/guides/recording-an-exception-an-auditor-accepts/ To document a risk acceptance, keep eight items together: finding, exposure, reason, verified controls, dated plan, residual risk, signature and expiry. FedRAMP's rule VER-RPT-AVI supplies the reporting behind the first three, and NIST SP 800-37 and SP 800-40 supply items 4, 5, 7 and 8. Item 6 is this site's grouping (read 19 Nov 2026). ## The eight items 1. **The finding with its identifiers.** FedRAMP VER-RPT-AVI wants the provider's "internally assigned tracking identifier" and the time and source of detection ([VER-RPT-AVI](https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-evaluation-and-reporting/)). Add the CVE; the FedRAMP POA&M template gained a CVE column on 27 Jun 2022 ([template post](https://www.fedramp.gov/archive/2022-06-28-update-poam-template/)). 2. **Exposure and likelihood.** Whether the flaw is internet-reachable, whether it is likely exploitable, and whether it is on the KEV catalog. The 2026 rule asks the first two as plain yes or no questions. 3. **The reason, in plain words.** The same rule asks for an "Explanation of why this is an accepted vulnerability". The wording table below shows weak and strong versions. 4. **Each control, with proof.** Description, owner, how it was checked and when. SP 800-37 expects plans to hold "an accurate description of implemented controls, including compensating controls" ([Task R-2](https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-37r2.pdf)), and SP 800-40 Rev. 4 asks that "no one uninstalls the patch, deactivates the additional security controls" ([Section 2.2](https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf)). 5. **A plan with dates.** Tasks, resources, milestones and "scheduled completion dates" ([Task A-6](https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-37r2.pdf)); the [plan of action and milestones guide](https://remediationvsmitigation.com/guides/poam-plan-of-action-and-milestones/) has a filled entry. 6. **Residual risk against tolerance.** One sentence on the level left, set against the tolerance the policy states. See [residual risk after mitigation](https://remediationvsmitigation.com/guides/residual-risk-after-mitigation/). 7. **The signature of the right role.** SP 800-37 says "The explicit acceptance of risk is the responsibility of the authorizing official and cannot be delegated to other officials within the organization" (Task R-4). For other organizations see [who may sign](https://remediationvsmitigation.com/risk-acceptance-form/who-may-sign/) and [the risk owner](https://remediationvsmitigation.com/guides/risk-owner/). 8. **An end date and a review rule.** The authorization decision conveys "the authorization termination date or time-driven authorization frequency" (Task R-4). SP 800-40 Rev. 4 adds that exceptions "should be reviewed regularly" (Section 3.5.5). What happens on the date is in [exception expiry and renewal](https://remediationvsmitigation.com/guides/exception-expiry-and-renewal/). The [filled example form](https://remediationvsmitigation.com/risk-acceptance-form/example/) shows one finding on the form itself; this page gives the reasoning behind each item and what an assessor asks of it. The [Risk Acceptance Form](https://remediationvsmitigation.com/risk-acceptance-form/) has 15 fields: 13 asked of every finding, which cover these eight items, plus a vendor field and a PCI field where they apply. ## What an assessor asks of each item Questions an assessor or reviewer can put to a risk acceptance record, with the published text each rests on, read 19 Nov 2026. PCI SSC states it cannot approve a compensating control itself (FAQ 1046). | Assessor's question | Evidence that answers it | Text behind the question | |---|---|---| | Who decided, and could they? | A named role, a signature, a date | SP 800-37 Task R-4: acceptance "cannot be delegated" | | Is the control working today? | Owner, verification method, verification date for each control | SP 800-40 Rev. 4 Section 2.2 step c: "ensure they are functioning as intended" | | Is the control enough for this requirement? | What the control protects and which other controls it depends on | PCI SSC FAQ 1046: a QSA decides whether a compensating control "is sufficient to meet the intent of a requirement during their review of all other controls in place" | | Is the time allowed justified? | A written risk analysis that sets the time frame | PCI SSC FAQ 1597: vulnerabilities below critical and high-risk are addressed "in accordance with the entity's risk as defined and documented in a targeted risk analysis" | | Does the record end, and is it revisited? | An expiry date and a review log | SP 800-37 Task R-4 termination date; SP 800-40 Rev. 4 Section 3.5.5 regular review | | Is it still open on the books? | A row in the register, not a closed ticket | FedRAMP legacy playbook: approved exceptions "must be captured on the POA&M's Open tab and periodically reassessed" | | Could a reader reproduce the attack from it? | Preconditions and controls only | FedRAMP VER-RPT-NID: disclose "sufficient information for informed risk-based decision-making" but nothing that "would likely lead to exploitation" | Keep request samples, payload text and step lists out of the record. Name the preconditions (reachable, authenticated, version) and the controls. The [exception register](https://remediationvsmitigation.com/exception-register/) shows the reassessment as row history. ## Reason field: weak and strong wording Patterns for the reason field; bracketed items are placeholders to fill from your own records. | Weak entry | Why it fails | Stronger entry (pattern) | |---|---|---| | "Too hard to patch." | States an opinion; names no constraint, date or alternative | "Vendor has not released a fix; vendor email dated [date] says planned for [quarter]; controls A and B hold until then." | | "Low risk." | Gives no basis and no tolerance statement | "Not internet-reachable (basis: [drawing, date]); not on the KEV catalog (version [x]); impact Moderate on our scale; within the stated tolerance for internal services." | | "Business critical." | Says why the asset matters, not why the finding is accepted | "Replacing the component before [date] stops [named business process]; replacement scheduled for [date] in the plan." | ## A record built on a real finding CVE-2026-75604 is a path traversal in the incremental cache of Next.js on Windows-hosted servers. The GitHub CNA's CVSS 3.1 vector is AV:N/AC:H/PR:N/UI:N/S:C/C:H/I:H/A:H, a base score of 9.0: no sign-in is needed, but attack complexity is high. Vercel released the fixes, Next.js 15.5.24 and 16.3.3, on 25 Aug 2026, and NVD published its record on 1 Sep 2026. The vendor states there is no known workaround and says to upgrade immediately if the server is on Windows. CISA's SSVC data in the NVD record shows exploitation as none, and the CVE is not in the KEV catalog (feed version 2026.11.18, read 19 Nov 2026). The estate, owners, controls and dates below are the example's own, and the record is signed on Thu 19 Nov 2026. Exception EXC-2026-044 (example organization, finding facts are real) 1. **Finding:** CVE-2026-75604, Vercel Next.js, path traversal in the incremental cache on Windows hosts. Not in the CISA KEV catalog (feed 2026.11.18). Two Windows Server hosts run the benefits portal on Next.js 15.5.21 (Pages Router). Vendor fix: 15.5.24 (GHSA-p293-qw3h-jr36). 2. **Exposure and likelihood:** The portal is reachable only from corporate subnets and the VPN (firewall rules and drawing dated 18 Nov 2026). Not internet-reachable. On the KEV catalog: no. Likely exploitable: only by someone already on the internal network, and the vendor rates attack complexity high. 3. **Reason:** The fix is a version bump that needs a rebuild, a redeploy and a regression test of the enrollment pages. The open-enrollment change freeze runs to Fri 27 Nov 2026, and the first window after it is Sat 28 Nov 2026. The vendor lists no workaround, so exposure is cut by network scope and detection. 4. **Controls:** (a) Inbound access to both hosts allowed only from the corporate subnets and the VPN range. Owner: Network operations lead. Verified 18 Nov 2026 by reading the rule list. (b) A web-log alert on any request path containing an encoded backslash (%5c), the example's own choice of traversal indicator. Owner: Security operations lead. Verified 18 Nov 2026 with a harmless test request. 5. **Plan:** Task 1, upgrade both hosts to Next.js 15.5.24 or later, rebuild and redeploy, from Sat 28 Nov 2026 (Web platform lead; needs 4 hours). Task 2, if not done by Mon 30 Nov 2026, decide between a second window and a renewed acceptance (Web platform lead). 6. **Residual risk:** Moderate on the organization's three-level scale while controls (a) and (b) hold; within the Moderate tolerance in policy section 4. 7. **Signature:** Director of Member Services (business owner of the benefits portal) and Head of Security (roles), signed Thu 19 Nov 2026. Who signs at Moderate is on [who may sign](https://remediationvsmitigation.com/risk-acceptance-form/who-may-sign/). The Head of Web Platforms owns the plan in item 5 and does not sign: the team that carries out the fix is not the one that accepts the risk it leaves. 8. **End and review:** Expires Fri 4 Dec 2026, 15 days after signing. The expiry sits four days past the 30 Nov decision point in item 5 so that a second window, Wed 2 Dec 2026, would still be covered; the plan date is a checkpoint, and after 4 Dec the acceptance lapses unless it is signed again. Review Thu 26 Nov 2026. Reopen at once if control (a) or (b) fails its check or the hosts gain a route from outside the corporate network. If CISA adds the CVE to the KEV catalog, the acceptance is not renewed: the signers end it and the upgrade moves to the earliest possible window. Two limits apply. First, a KEV listing during an acceptance means escalate, not renew. Where a directive applies, the KEV date binds: BOD 26-04 covers federal civilian agency assets, and the directive text describes no acceptance procedure that moves a KEV timeline, so an agency takes the question to the directive's contact, CyberDirectives@cisa.dhs.gov, and does not rely on an internal record. An organization outside the directive uses the listing as the trigger in item 8: the signers end the acceptance and the fix moves up. [Exception expiry and renewal](https://remediationvsmitigation.com/guides/exception-expiry-and-renewal/) shows that case on a KEV-listed CVE. Second, the reason depends on facts that can change: if a route to the hosts appears from outside the corporate network, the reasoning in item 3 no longer holds, which is why item 8 lists the reopen conditions. ## Sources - FedRAMP 2026, VER-RPT-AVI and VER-RPT-NID: https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-evaluation-and-reporting/ (read 19 Nov 2026) - FedRAMP, Update to the POA&M template, 27 Jun 2022: https://www.fedramp.gov/archive/2022-06-28-update-poam-template/ (read 19 Nov 2026) - NIST SP 800-37 Rev. 2, Tasks R-2, R-4 and A-6: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-37r2.pdf (read 19 Nov 2026) - NIST SP 800-40 Rev. 4, Sections 2.2 and 3.5.5: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf (read 19 Nov 2026) - PCI SSC FAQ 1046, approval of compensating controls: https://www.pcisecuritystandards.org/faqs/1046 (read 19 Nov 2026) - PCI SSC FAQ 1597, resolving and addressing vulnerabilities: https://www.pcisecuritystandards.org/faqs/1597 (read 19 Nov 2026) - FedRAMP legacy POA&M playbook (reference only since 24 Jun 2026): https://www.fedramp.gov/legacy/playbook/csp/authorization/poam/ (read 19 Nov 2026) - CISA KEV catalog JSON feed, catalogVersion 2026.11.18 (CVE-2026-75604 not listed): https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) - CISA BOD 26-04, scope and contact: https://www.cisa.gov/news-events/directives/bod-26-04-prioritizing-security-updates-based-risk (read 19 Nov 2026) - NVD record for CVE-2026-75604 (CVSS vector, SSVC exploitation entry): https://services.nvd.nist.gov/rest/json/cves/2.0?cveId=CVE-2026-75604 (read 19 Nov 2026) - GitHub advisory GHSA-p293-qw3h-jr36 (affected range 13.4.0 up to 15.5.24 and 16.3.3; no known workaround): https://github.com/vercel/next.js/security/advisories/GHSA-p293-qw3h-jr36 (read 19 Nov 2026) - Next.js 15.5.24 and 16.3.3 releases (both published 25 Aug 2026): https://api.github.com/repos/vercel/next.js/releases/tags/v15.5.24 (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Risk Acceptance Form: Free Template With 15 Fields](https://remediationvsmitigation.com/risk-acceptance-form/index.md) - [Risk Acceptance Form Example: Filled-In Worked Sample](https://remediationvsmitigation.com/risk-acceptance-form/example/index.md) - [Exception Register: 15 Columns, States and Blank CSV](https://remediationvsmitigation.com/exception-register/index.md) - [Who Can Accept Risk: Signer by Standard (2026)](https://remediationvsmitigation.com/risk-acceptance-form/who-may-sign/index.md) - [POA&M Example: A Filled Plan of Action and Milestones Entry](https://remediationvsmitigation.com/guides/poam-plan-of-action-and-milestones/index.md) - [Who Is the Risk Owner for a Security Finding?](https://remediationvsmitigation.com/guides/risk-owner/index.md) - [Residual Risk After Mitigation: Define, Estimate, Record](https://remediationvsmitigation.com/guides/residual-risk-after-mitigation/index.md) - [Risk Exception Expiry: What Happens and How to Renew](https://remediationvsmitigation.com/guides/exception-expiry-and-renewal/index.md) --- # What is remediation in cyber security Canonical: https://remediationvsmitigation.com/guides/remediation-in-cybersecurity/ Remediation in cyber security means eliminating the vulnerability: CISA BOD 26-04 (10 Jun 2026) says it is eliminated "through patching, decommissioning the system, or another action". FedRAMP adds that a remediated vulnerability is "no longer detected". In CISA's KEV feed of 19 Nov 2026, 893 of 1,786 entries carry the exact action "Apply updates per vendor instructions." ## The definition that deadlines use BOD 26-04 states that "the vulnerability shall be remediated (that is, the vulnerability is eliminated through patching, decommissioning the system, or another action)". Footnote 8 says the use is consistent with DoD Instruction 8531.01, NIST SP 800-53 Rev. 5 and NIST SP 800-216 ([BOD 26-04](https://www.cisa.gov/news-events/directives/bod-26-04-prioritizing-security-updates-based-risk)). NIST SP 800-216 words it as "The neutralization or elimination of a vulnerability or the likelihood of its exploitation". FedRAMP's 2026 definition adds the test: a remediated vulnerability "has been neutralized or eliminated and is no longer detected", and its rule text says "FedRAMP does not use the terms "mitigation" and "remediation" interchangeably" ([FedRAMP definitions](https://www.fedramp.gov/2026/definitions/), [VDR rules](https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-detection-and-response/)). Older glossary wording is looser. The [NIST glossary entry](https://csrc.nist.gov/glossary/term/remediation) lists CNSSI 4009-2022 definitions of remediation as "The act of mitigating a vulnerability or a threat" and as "Actions taken to correct known deficiencies and weaknesses once a vulnerability has been identified". A contract or policy that borrows the first wording lets a mitigation count as remediation, so state which definition applies. The comparison with mitigation is in the [difference guide](https://remediationvsmitigation.com/guides/difference-between-remediation-and-mitigation/). ## What counts, and what does not What the sources count as remediation, read 19 Nov 2026. | Action | Counts as remediation? | Why | |---|---|---| | Vendor patch or upgrade installed and in effect | Yes | Eliminates the flaw in the code (BOD 26-04) | | Software removed | Yes | The flaw is no longer present | | System decommissioned | Yes | Named in the BOD 26-04 definition | | Update installed but the scan still detects the flaw | Not yet | FedRAMP: a remediated vulnerability is "no longer detected" | | WAF or IPS rule | No | OWASP: a virtual patch "does not remove the underlying vulnerability" | | System taken off the internet | No | BOD 26-04 calls it a valid mitigation that changes the timeline | | Finding closed as a false positive | Not a response | Verify and record; see [false positive handling](https://remediationvsmitigation.com/guides/false-positive-handling/) | ## Six steps, who does them and what each leaves behind NIST SP 800-40 Rev. 4 Section 2.2 gives the life cycle of a software vulnerability response: know, plan, prepare, implement, verify, monitor. Applied to remediation: Steps from NIST SP 800-40 Rev. 4 Sections 2.2 and 2.3, owners and records as this site's practice, read 19 Nov 2026. | Step | Usual owner | Record it leaves | How it usually fails | |---|---|---|---| | 1. Know the asset is affected | Vulnerability management | Finding with product, version, asset and date | Stale inventory; SP 800-40 says inventories updated "monthly or quarterly" by scan is a model that "should no longer be used" | | 2. Choose remediation | System owner with security | Decision: remediate, mitigate or accept, with the reason | A mitigation chosen with no fix date; track it in a [plan of action and milestones](https://remediationvsmitigation.com/guides/poam-plan-of-action-and-milestones/) | | 3. Prepare | Change manager, system owner | Tested update, change ticket, window | SP 800-40 names the trade-off: testing "decreases the risk of operational disruption but increases the window of opportunity for attackers" | | 4. Implement | System administrator | Install log, new version | Installed but not in effect: SP 800-40 lists restarting software and rebooting the platform as steps that make a patch take effect | | 5. Verify | Security analyst | Rescan result showing the finding no longer detected, scan ID, date | Checking the installer's log instead of the scanner; SP 800-40 asks that a patch be shown "installed successfully and taken effect" | | 6. Keep watching, then close | Vulnerability management | Close note with the scan ID; monitoring rule | Rollback: SP 800-40 lists a patch uninstalled, an unpatched version restored from backup, and a device reset to a vulnerable factory default | ## A worked case: CVE-2026-20320 in Cisco BroadWorks Cisco's advisory cisco-sa-bworks-xxe-uwUd7CEt, first published 19 Aug 2026, rates the flaw High (CVSS base 7.5): an unauthenticated, remote attacker could read sensitive configuration information. It affects four products, BroadWorks Application Delivery Platform, Application Server, Profile Server and Xtended Services Platform, on releases earlier than RI.2026.07, and Cisco states that there are no workarounds. CISA added it to the KEV catalog on 10 Nov 2026, 83 days after the advisory, with a due date of 24 Nov 2026, 14 days later. On 12 Nov Cisco revised the advisory to say exploitation was known, and on 13 Nov corrected it to say only that the CVE was added to the KEV catalog. With no workaround, the KEV action leaves two paths: install the fixed release, or, in the feed's words, "discontinue use of the product if mitigations are unavailable". An operator running an affected Application Server on 19 Nov 2026 has this schedule. The dates after 19 Nov are the example's plan. Example schedule against the advisory of 19 Aug 2026 and the KEV due date of 24 Nov 2026. | Date | Step | Owner | Record | |---|---|---|---| | Thu 19 Nov 2026 | Find every affected BroadWorks server; note release and exposure | Vulnerability management | Finding list with versions | | Fri 20 Nov 2026 | Approve the upgrade to RI.2026.07 and book the window | System owner, change manager | Change ticket | | Sat 21 Nov 2026 | Install RI.2026.07, restart as the release notes direct | System administrator | Install log, release shown after the change | | Mon 23 Nov 2026 | Rescan; the finding reads as no longer detected; close | Security analyst | Scan ID, close note | | Tue 24 Nov 2026 | KEV due date, one day of margin | | Closed finding on file | If the window slips past 24 Nov, Cisco offers no workaround to hold the position, so the only interim step is taking the service out of reach. BOD 26-04 treats that as a mitigation that changes the timeline, not as remediation. ## How the KEV wording changed after BOD 26-04 The KEV catalog (version 2026.11.18, 1,786 entries) shows the shift when its required-action field is counted. Counts were made on 19 Nov 2026 by text match on that field in the [KEV feed](https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json): exact string match for the first row, case-insensitive match on "apply updates" and "mitigations" for the rest. KEV required-action wording by date added, catalog 2026.11.18, counted 19 Nov 2026. | Group | Entries | Required action wording | |---|---|---| | Added before 10 Jun 2026 | 1,617 | 893 carry exactly "Apply updates per vendor instructions."; 24 more contain "apply updates" in another form | | Added 10 Jun to 18 Nov 2026 | 169 | All 169 begin "Apply mitigations in accordance with vendor instructions, ensuring compliance with CISA’s BOD 26-04"; 108 are flagged for forensic triage | | Whole catalog | 1,786 | 917 contain "apply updates", all added before BOD 26-04; 791 contain "mitigations", of which 169 are the new wording | So the older entries are the ones that say "update", and the newer ones say "mitigations" even where the vendor ships a fix, as the Cisco case shows. Read the vendor advisory for whether a fixed release exists; do not infer it from the KEV wording. ## Cases where the usual answer is wrong - **Decommissioning counts.** Retiring the system is remediation under BOD 26-04, so a scheduled retirement can close a finding without any patch, provided the date is kept ([decommission row](https://remediationvsmitigation.com/decision-table/decommission-scheduled/)). - **PCI DSS separates two words.** PCI SSC FAQ 1597 defines "Resolved" as the entity "solves or fixes the vulnerability" and "Addressed" as the entity determining "whether to resolve the vulnerability or to mitigate the risk by addressing the vulnerability in another way (e.g., with a compensating control or by disabling a vulnerable service)". Critical and high-risk findings must be resolved; the rest need only be addressed. - **A fix that waits on a vendor.** Until the update exists, remediation is not available; the [decision table](https://remediationvsmitigation.com/decision-table/) and mitigation when no patch exists cover the interim step. ## Sources - CISA BOD 26-04, Appendix A and footnote 8: https://www.cisa.gov/news-events/directives/bod-26-04-prioritizing-security-updates-based-risk (read 19 Nov 2026) - NIST CSRC glossary, remediation: https://csrc.nist.gov/glossary/term/remediation (read 19 Nov 2026) - FedRAMP 2026 definitions, remediated vulnerability: https://www.fedramp.gov/2026/definitions/ (read 19 Nov 2026) - FedRAMP 2026, VDR rules: https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-detection-and-response/ (read 19 Nov 2026) - CISA KEV catalog JSON feed, catalogVersion 2026.11.18: https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) - OWASP Virtual Patching Cheat Sheet: https://cheatsheetseries.owasp.org/cheatsheets/Virtual_Patching_Cheat_Sheet.html (read 19 Nov 2026) - NIST SP 800-40 Rev. 4, Sections 2.2, 2.3 and 3: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf (read 19 Nov 2026) - Cisco Security Advisory cisco-sa-bworks-xxe-uwUd7CEt, version 1.2: https://sec.cloudapps.cisco.com/security/center/content/CiscoSecurityAdvisory/cisco-sa-bworks-xxe-uwUd7CEt (read 19 Nov 2026) - PCI SSC FAQ 1597, resolved and addressed: https://www.pcisecuritystandards.org/faqs/1597 (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Difference Between Remediation and Mitigation in IT Security](https://remediationvsmitigation.com/guides/difference-between-remediation-and-mitigation/index.md) - [POA&M Example: A Filled Plan of Action and Milestones Entry](https://remediationvsmitigation.com/guides/poam-plan-of-action-and-milestones/index.md) - [Decision Table: Remediate, Mitigate or Accept by Situation](https://remediationvsmitigation.com/decision-table/index.md) - [Risk Acceptance Form: Free Template With 15 Fields](https://remediationvsmitigation.com/risk-acceptance-form/index.md) - [Mitigation vs Remediation Wording in Vendor Advisories](https://remediationvsmitigation.com/guides/mitigation-vs-remediation-in-vendor-advisories/index.md) - [How to Verify a Mitigation Works: Checks for 5 Real CVEs](https://remediationvsmitigation.com/guides/verifying-a-mitigation-works/index.md) --- # Residual risk after mitigation Canonical: https://remediationvsmitigation.com/guides/residual-risk-after-mitigation/ Residual risk is the portion of risk left after security measures are applied, and a mitigation never takes it to zero. The worked case rates CVE-2024-10441 (CVSS base score 9.8, added to CISA's KEV catalog on 18 Nov 2026) High before any control and Moderate after two, on one scale, with the next reduction dated. ## Three wordings, one difference Definitions from the NIST CSRC glossary entry for residual risk, read 19 Nov 2026. | Source | Wording of "residual risk" | Condition built in | |---|---|---| | CNSSI 4009 (in NIST SP 800-30 Rev. 1) | "Portion of risk remaining after security measures have been applied." | Measures applied | | NIST SP 800-161 Rev. 1 | "Portion of risk remaining after controls/countermeasures have been applied." | Controls applied | | NIST IR 8286 | "Risk that remains after risk responses have been documented and performed." | Documented and performed | The third wording is the stricter one for a mitigation record. A control that is planned but not installed, or installed but not written down, is not yet a response, so the residual figure stays at the earlier level until it is. NIST SP 800-37 adds that "regardless of the risk response, there remains a degree of residual risk" and that organizations "determine acceptable degrees of residual risk based on organizational risk tolerance". The inherent-risk side of the pair is on the [inherent vs residual risk](https://remediationvsmitigation.com/guides/inherent-vs-residual-risk/) page. ## Estimate it the way the first estimate was made NIST IR 8286, Section 3.5.2: "The residual risk can be calculated using the same methods for calculating current risk". A record that rates the finding on one scale before a control and on another after cannot show a reduction. The same section tells the risk owner what to do with the result: "If the residual risk is beyond the acceptable level of risk, then the risk owner should evaluate whether the risk can be brought to an acceptable level (for example, by applying additional security controls)." ## A worked case: CVE-2024-10441 on a NAS The finding is real. Synology advisory Synology-SA-24:20, published 5 Nov 2024, rates CVE-2024-10441 Critical with a CVSS 3 base score of 9.8 (vector CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H) and says remote attackers can execute arbitrary code. The advisory lists the mitigation as "None" and the fixed release for DSM 7.2.2 as 7.2.2-72806-1. CISA added the CVE to the KEV catalog on 18 Nov 2026, 743 days after the advisory, with a due date of 21 Nov 2026 and the forensic triage flag set to Yes. The asset below, its owners and every date after 19 Nov are the example's own: one DiskStation running DSM 7.2.2 earlier than 7.2.2-72806-1, holding finance team shares, reachable from the internet through a port forward. The scale is also the example's. Both estimates use it, as IR 8286 requires. The example organization's rating scale. Policy in the example: Moderate may be accepted for up to 30 days; High may not be accepted. | Likelihood of exploitation | Impact Low | Impact Moderate | Impact High | |---|---|---|---| | Low | Low | Low | Moderate | | Medium | Low | Moderate | High | | High | Moderate | High | High | Example ladder for one finding, with the ratings in the scale above. Dates after 19 Nov 2026 are planned. | Date | Step and evidence | Likelihood | Impact | Rating | |---|---|---|---|---| | Thu 19 Nov 2026 | Baseline. Reachable from the internet, no login needed, on the KEV catalog; full control of the device would follow. Evidence: port-forward list, KEV entry. | High | High | High | | Thu 19 Nov 2026 | First control: the port forward is removed, so only the office network reaches the NAS. Evidence: port-forward list after the change. | Medium | High | High | | Fri 20 Nov 2026 | Second control: a firewall rule limits the NAS to two administrator workstations. Evidence: rule list, a refused connection from a non-administrator workstation in the log. | Low | High | Moderate | | Sat 21 Nov 2026 | KEV due date. Forensic triage of the NAS completed with no sign of compromise. Evidence: triage note. | Low | High | Moderate | | Sat 28 Nov 2026 | Next reduction: upgrade to 7.2.2-72806-1 or later. Evidence: version shown after the change. | | | | | Mon 30 Nov 2026 | Rescan reads the finding as no longer detected; entry closed. | | | Closed | Three things in the table need attention. The first control lowered the likelihood but left the rating at High, so a record that logs only the rating would show no progress, while a record that logs likelihood and impact separately shows it. The Moderate figure on 20 Nov is inside the 30-day tolerance and the upgrade is dated 8 days later, so the acceptance period is short and ends with a fix. The figure also describes the future only: the KEV entry flags forensic triage, which BOD 26-04 describes as completing "remediation or mitigation action within the timeline (three days)" and carrying out a triage "to assess whether the system is compromised". A control cannot lower the risk of an intrusion that already happened, which is why the 21 Nov row records the triage result. ## How FedRAMP records each step down FedRAMP's 2026 definitions call a vulnerability partially mitigated when "the risk of exploitation still exists and the vulnerability is still detected", and fully mitigated when likelihood or impact has been reduced "until either are negligible, but the vulnerability is still detected". Rule VER-RPT-VDT asks providers for the "Time and Potential Agency Impact N-rating of each completed and evaluated reduction" and the "Estimated time and target Potential Agency Impact N-rating of next reduction". The ladder above carries the same fields: each rung has a date, a level and a next target. The wording that matters for a record is in the [FedRAMP definitions](https://www.fedramp.gov/2026/definitions/). ## A control that only helps response leaves the rating unchanged NIST SP 800-82 Rev. 3 separates two kinds of compensating control for OT. Limiting exposure, restricting vulnerable services and virtual patching reduce the chance of exploitation. "Other compensating controls that do not decrease the residual risk but increase the ability to respond may be desirable (e.g., provide a timely response in case of an incident, devise a plan to ensure that the OT can identify exploitation of the flaw)." Adding the NAS to a monitoring rule would sit in the second group: it belongs on the record, but the rating in the ladder would not change. See [increased monitoring](https://remediationvsmitigation.com/compensating-controls/increased-monitoring/) in the register. ## Who estimates and who signs The security analyst proposes the likelihood and impact, the system owner confirms the facts they rest on (exposure, version, data held), and the risk owner decides whether the residual level is acceptable, per IR 8286. Under NIST SP 800-37 Task R-4, acceptance by the authorizing official "cannot be delegated". Other organizations name their own signer; see [who is the risk owner](https://remediationvsmitigation.com/guides/risk-owner/) and [who may sign](https://remediationvsmitigation.com/risk-acceptance-form/who-may-sign/). The record carries four things: the before and after levels on one scale with the control that moved each, the tolerance statement they are compared with, the date measured with the next reduction and its date, and the evidence for each control. The field for them is in the [Risk Acceptance Form](https://remediationvsmitigation.com/risk-acceptance-form/). The case where the control is already in front of the flaw is in [control already in place](https://remediationvsmitigation.com/decision-table/compensating-control-in-front/). ## Sources - NIST CSRC glossary, residual risk: https://csrc.nist.gov/glossary/term/residual_risk (read 19 Nov 2026) - NIST IR 8286, Section 3.5.2: https://nvlpubs.nist.gov/nistpubs/ir/2020/NIST.IR.8286.pdf (read 19 Nov 2026) - NIST SP 800-37 Rev. 2, Task R-3 discussion and Task R-4: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-37r2.pdf (read 19 Nov 2026) - NIST SP 800-82 Rev. 3, Section 6.2.11 (SI-2 OT discussion): https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-82r3.pdf (read 19 Nov 2026) - FedRAMP 2026, VER-RPT-VDT: https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-evaluation-and-reporting/ (read 19 Nov 2026) - FedRAMP 2026 definitions, partially and fully mitigated vulnerability: https://www.fedramp.gov/2026/definitions/ (read 19 Nov 2026) - Synology advisory Synology-SA-24:20: https://www.synology.com/en-global/security/advisory/Synology_SA_24_20 (read 19 Nov 2026) - CISA KEV catalog JSON feed, catalogVersion 2026.11.18, entry CVE-2024-10441: https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) - CISA BOD 26-04, Appendix A: https://www.cisa.gov/news-events/directives/bod-26-04-prioritizing-security-updates-based-risk (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Inherent vs Residual Risk on a Security Decision Record](https://remediationvsmitigation.com/guides/inherent-vs-residual-risk/index.md) - [Risk Acceptance Form: Free Template With 15 Fields](https://remediationvsmitigation.com/risk-acceptance-form/index.md) - [Increased Monitoring as a Compensating Control](https://remediationvsmitigation.com/compensating-controls/increased-monitoring/index.md) - [A Compensating Control Is in Front: Why a Fix Date Stays](https://remediationvsmitigation.com/decision-table/compensating-control-in-front/index.md) - [Who Is the Risk Owner for a Security Finding?](https://remediationvsmitigation.com/guides/risk-owner/index.md) - [Recording an Exception an Auditor Accepts: Eight Items](https://remediationvsmitigation.com/guides/recording-an-exception-an-auditor-accepts/index.md) --- # Risk acceptance criteria Canonical: https://remediationvsmitigation.com/guides/risk-acceptance-criteria/ Seven tests decide whether a finding may be accepted instead of fixed: tolerance, a signer with authority, verified controls, a dated plan, written residual risk, no bar from a directive or contract, and an expiry date. On 14 Oct 2026 a PAN-OS flaw fails four of the seven for a federal agency and three for anyone else; a Linux kernel flaw passes all seven. ## The seven tests No NIST publication gives one list of acceptance criteria. These seven are assembled from the clauses that each govern one condition. A finding that fails any one of them is fixed, mitigated or avoided instead. Seven tests for a risk acceptance, each with its source; texts read 19 Nov 2026. | Test | Pass condition | Source | |---|---|---| | 1. Tolerance or compelling need | The risk is within stated tolerance, or a named mission or business need justifies more | [SP 800-39, Task 3-1](https://nvlpubs.nist.gov/nistpubs/Legacy/SP/nistspecialpublication800-39.pdf) | | 2. Authority | The signer holds the authority the policy gives; for US federal systems, the authorizing official, undelegated | [SP 800-37 Rev. 2, Task R-4](https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-37r2.pdf) | | 3. Verified controls | Each control is described, owned and checked, and the record shows it "continues to be in place" | [SP 800-40 Rev. 4, Section 2.2](https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf) | | 4. Dated plan | Tasks, resources, milestones and "scheduled completion dates" | [SP 800-37 Rev. 2, Task A-6](https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-37r2.pdf) | | 5. Residual risk written | The risk left after the response is stated and compared with tolerance | [SP 800-37 Rev. 2, Task R-3](https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-37r2.pdf) | | 6. Nothing bars it | No directive, standard or contract term forbids acceptance | [BOD 26-04 guidance](https://www.cisa.gov/news-events/directives/bod-26-04-implementation-guidance-prioritizing-security-updates-based-risk), [PCI SSC FAQ 1046](https://www.pcisecuritystandards.org/faqs/1046) | | 7. Expiry | A termination date set by the approver | [SP 800-37 Rev. 2, Task R-4](https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-37r2.pdf) | Test 1 depends on a limit that NIST declines to set: "There is no correct level of organizational risk tolerance" ([SP 800-39](https://nvlpubs.nist.gov/nistpubs/Legacy/SP/nistspecialpublication800-39.pdf)). The organization has to write it down before the first acceptance is signed; [risk acceptance vs risk tolerance](https://remediationvsmitigation.com/guides/risk-acceptance-vs-risk-tolerance/) covers how. Test 7 borrows from authorization: in SP 800-37 the termination date "is established by the authorizing official and indicates when the authorization expires." ## Two findings run through the tests CVE-2026-0322 is a code injection flaw in the PAN-OS management interface. Palo Alto Networks published fixed builds on 14 Oct 2026, states that no workaround exists, rates exploit maturity as attacked and reports very limited exploitation ([advisory](https://security.paloaltonetworks.com/CVE-2026-0322)). CISA added it to the Known Exploited Vulnerabilities catalog the same day with a due date of 17 Oct 2026 ([KEV feed](https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json)). CVE-2026-80521 is a use-after-free in the Linux kernel's AF_UNIX garbage collector that a local, low-privilege user can reach. The 6.1 stable fix shipped in Linux 6.1.189 on 3 Oct 2026 ([changelog](https://cdn.kernel.org/pub/linux/kernel/v6.x/ChangeLog-6.1.189)), and the CVE was not in the catalog in version 2026.11.18 ([NVD record](https://services.nvd.nist.gov/rest/json/cves/2.0?cveId=CVE-2026-80521)). The host used for the second CVE is invented: a Linux 6.1 log-collector server, lc-02, with three named administrator accounts and no other local users. The same CVE is filled in on all 15 fields, for a different setting (six gateway appliances whose vendor has shipped no fixed kernel), in the [worked Risk Acceptance Form](https://remediationvsmitigation.com/risk-acceptance-form/example/). Both findings run through all seven tests. The CVE facts are from the cited advisories and the catalog read on 19 Nov 2026; the log collector, the policy and the signer on the right are invented. | Test | CVE-2026-0322, PAN-OS firewall | CVE-2026-80521, Linux 6.1 log collector | |---|---|---| | 1. Tolerance | Fails. Root code execution, active exploitation and a fix on day one. No stated tolerance covers that | Passes. Local access and a won race are needed; only three administrators can log in; the policy tolerates Moderate for 30 days on internal servers | | 2. Authority | Passes. The policy names the signer | Passes. The security head signs under the same policy | | 3. Verified controls | Fails. Palo Alto says restricting management access to one jump box lowers its CVSS 4.0 rating only from 8.6 to 8.5, so the control does not bring the risk down to tolerance | Passes. SSH is limited to the three accounts; the account list and the key list were read on 19 Nov 2026 | | 4. Dated plan | Passes. A plan can be dated: upgrade to the fixed build for the train, such as 11.2.16 or 10.2.18-h12 | Passes. Install 6.1.189 or later in the window of Sat 12 Dec 2026 | | 5. Residual risk | Fails. With the jump box in place the vendor still rates the flaw 8.5 (High), above any Moderate tolerance | Passes. Low: an attacker would first need one of the three accounts | | 6. Nothing bars it | Federal civilian agency: fails. CISA "does not issue waivers or exceptions for actions required in cyber directives" and the due date was 17 Oct 2026 ([guidance](https://www.cisa.gov/news-events/directives/bod-26-04-implementation-guidance-prioritizing-security-updates-based-risk)). Any other organization: passes, unless a contract or PCI DSS requirement applies | Passes; no directive, contract term or PCI DSS requirement applies | | 7. Expiry | Passes in form; any expiry date can be written, which is why this test cannot rescue a finding that failed earlier ones | Fri 18 Dec 2026, 29 days after signing on 19 Nov 2026, six days after the window | | Outcome | Patch. A federal agency fails tests 1, 3, 5 and 6; any other organization fails tests 1, 3 and 5. Either way acceptance is not available. A short mitigation is allowed only while the build is staged, and it is recorded as mitigation | Accept until 18 Dec 2026, with a reopen trigger if the CVE reaches the catalog | The right-hand column passes because every cell can be checked by someone other than the signer. The record for it holds: ID EXC-2026-044 (invented); CVE-2026-80521 on lc-02; reason (fix 6.1.189 exists, install waits for the 12 Dec window); controls (SSH limited to three accounts, verified 19 Nov 2026; key list read the same day); plan (install 6.1.189 or later on 12 Dec 2026); residual risk Low; signer the security head, signed 19 Nov 2026; expiry 18 Dec 2026; reopen if the CVE reaches the KEV catalog or a fourth account is added. If the account list could not be produced, test 3 would fail and the finding would go back for mitigation. The left-hand column fails whether or not a directive applies: a finding that fails tests 1, 3 and 5 is not a candidate for the [Risk Acceptance Form](https://remediationvsmitigation.com/risk-acceptance-form/), whatever the cost of patching. ## Where acceptance is not available at all Some findings fail test 6 before anyone weighs risk. A federal civilian agency facing a catalog action gets no waiver (the [Decision Table](https://remediationvsmitigation.com/decision-table/) row for a missed KEV due date sets out the steps). BOD 26-04 also counts removal as a fix: the vulnerability "shall be remediated (that is, the vulnerability is eliminated through patching, decommissioning the system, or another action)" ([BOD 26-04](https://www.cisa.gov/news-events/directives/bod-26-04-prioritizing-security-updates-based-risk)). The choice in that setting is between ways of fixing, not between fixing and accepting. In PCI DSS the entity does not approve its own substitute. PCI SSC states it is "not able to approve specific configurations or compensating controls," and points to a Qualified Security Assessor to evaluate whether a compensating control meets the intent of the requirement ([FAQ 1046](https://www.pcisecuritystandards.org/faqs/1046)). The PCI row of the [Decision Table](https://remediationvsmitigation.com/decision-table/) shows the route. Residual risk after the response (test 5) is covered in [residual risk after mitigation](https://remediationvsmitigation.com/guides/residual-risk-after-mitigation/). The owner who proposes the acceptance is a different role from the signer in test 2; see [who is the risk owner](https://remediationvsmitigation.com/guides/risk-owner/). ## Sources - NIST SP 800-39, Task 3-1 and the risk tolerance discussion: https://nvlpubs.nist.gov/nistpubs/Legacy/SP/nistspecialpublication800-39.pdf (read 19 Nov 2026) - NIST SP 800-37 Rev. 2, Tasks A-6, R-3 and R-4: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-37r2.pdf (read 19 Nov 2026) - NIST SP 800-40 Rev. 4, Section 2.2: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf (read 19 Nov 2026) - CISA BOD 26-04, definitions beneath Table 1: https://www.cisa.gov/news-events/directives/bod-26-04-prioritizing-security-updates-based-risk (read 19 Nov 2026) - CISA BOD 26-04 implementation guidance, waiver question: https://www.cisa.gov/news-events/directives/bod-26-04-implementation-guidance-prioritizing-security-updates-based-risk (read 19 Nov 2026) - PCI SSC FAQ 1046: https://www.pcisecuritystandards.org/faqs/1046 (read 19 Nov 2026) - Palo Alto Networks advisory, CVE-2026-0322: https://security.paloaltonetworks.com/CVE-2026-0322 (read 19 Nov 2026) - CISA Known Exploited Vulnerabilities feed, version 2026.11.18: https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) - NVD record, CVE-2026-80521: https://services.nvd.nist.gov/rest/json/cves/2.0?cveId=CVE-2026-80521 (read 19 Nov 2026) - kernel.org, ChangeLog-6.1.189: https://cdn.kernel.org/pub/linux/kernel/v6.x/ChangeLog-6.1.189 (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Risk Acceptance Form: Free Template With 15 Fields](https://remediationvsmitigation.com/risk-acceptance-form/index.md) - [Risk Acceptance vs Risk Tolerance: Who Sets the Limit](https://remediationvsmitigation.com/guides/risk-acceptance-vs-risk-tolerance/index.md) - [Exception Register: 15 Columns, States and Blank CSV](https://remediationvsmitigation.com/exception-register/index.md) - [Residual Risk After Mitigation: Define, Estimate, Record](https://remediationvsmitigation.com/guides/residual-risk-after-mitigation/index.md) - [Who Is the Risk Owner for a Security Finding?](https://remediationvsmitigation.com/guides/risk-owner/index.md) --- # Risk acceptance vs risk avoidance Canonical: https://remediationvsmitigation.com/guides/risk-acceptance-vs-risk-avoidance/ Risk acceptance keeps the asset because the risk is within tolerance; risk avoidance removes it because the risk is not (NIST SP 800-39). A BOD 26-02 case mitigates and accepts a real FortiOS flaw on 19 Nov 2026 on a firewall that cannot migrate to a fixed release, then closes it as avoided at replacement on 15 May 2027. ## The line NIST draws NIST SP 800-39 gives each response its own trigger ([Task 3-1](https://nvlpubs.nist.gov/nistpubs/Legacy/SP/nistspecialpublication800-39.pdf)). "Risk acceptance is the appropriate risk response when the identified risk is within the organizational risk tolerance." "Risk avoidance may be the appropriate risk response when the identified risk exceeds the organizational risk tolerance." Avoidance "involves taking specific actions to eliminate the activities or technologies that are the basis for the risk". NIST's example ends with eliminating a networked connection and using an "air gap". For software, SP 800-40 Rev. 4 gives the two in concrete terms ([Section 2.1](https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf)). Accept: rely on "existing security controls to prevent vulnerability exploitation" or find the impact low enough that no action is needed. Avoid: "uninstalling the vulnerable software, decommissioning assets with the vulnerabilities, or disabling computing capabilities in assets that can function without them." Acceptance and avoidance compared; SP 800-39 Task 3-1 and SP 800-40 Rev. 4 Sections 2.1 and 2.2, read 19 Nov 2026. | Question | Risk acceptance | Risk avoidance | |---|---|---| | What changes on the asset | Nothing, or only controls around it | The software is removed, the asset retired or the capability turned off | | Record to keep | Signed [form](https://remediationvsmitigation.com/risk-acceptance-form/) with an expiry; a row in the [exception register](https://remediationvsmitigation.com/exception-register/) | Decommission or removal record; a rescan that no longer finds the asset | | What NIST says to check afterward | That the response "continues to be in place" | "verify that vulnerable assets were decommissioned or replaced", and that no one "restarts the decommissioned asset" | | Ends when | The expiry date, a trigger, or a fix | The asset or feature no longer exists | ## Which response fits which situation Decision table for acceptance and avoidance; sources read 19 Nov 2026. A row gives the response the source supports, not a rule for every organization. | Situation | Response | Why, and from where | |---|---|---| | Risk is inside stated tolerance, no directive applies | Accept, with expiry | SP 800-39 Task 3-1: acceptance is appropriate within tolerance | | Risk is above tolerance and the feature or software is not needed | Avoid: uninstall or disable | SP 800-40 Rev. 4 lists disabling "computing capabilities in assets that can function without them" under avoid | | Risk is above tolerance, the asset is needed, the fix is weeks away | Mitigate now, accept the rest for a dated period, fix on a milestone | SP 800-39 describes accepting "for a limited time", then avoiding or mitigating in sequence | | A federal civilian agency holds an end-of-support edge device | Avoid by the directive's dates | BOD 26-02 requires decommissioning; see the case below | | A federal civilian agency cannot meet a catalog due date | Remove the asset or apply the catalog action; no acceptance | CISA "does not issue waivers or exceptions for actions required in cyber directives" ([guidance](https://www.cisa.gov/news-events/directives/bod-26-04-implementation-guidance-prioritizing-security-updates-based-risk)) | | A critical mission system has no newer technology or is too isolated to replace | Accept narrowly, with controls and a replacement plan | SA-22 guidance names these as exceptions to replacing unsupported components ([SP 800-53 Rev. 5](https://github.com/usnistgov/oscal-content/blob/main/nist.gov/SP800-53/rev5/json/NIST_SP-800-53_rev5_catalog.json)) | SP 800-39 treats the responses as a sequence, not a menu: in an emergency an organization "might accept the risk associated with unfiltered connection to an external communications provider for a limited time; then avoid risk by cutting the connection; mitigate risk in the near-term" and then mitigate in the long term. Accept first, avoid on a date is therefore NIST's own pattern. ## Worked case: a firewall that cannot migrate, replaced under BOD 26-02 BOD 26-02 times its required actions from issuance. At twelve months agencies decommission devices on the CISA EOS Edge Device List and inventory edge devices that will reach end of support within the succeeding twelve months. At eighteen months they "Decommission all identified EOS edge devices from agency networks", and at twenty-four months they decommission devices "on or before the date these devices reach EOS". Each decommission is reported "to CISA using the CISA-provided reporting template" ([BOD 26-02](https://www.cisa.gov/news-events/directives/bod-26-02-mitigating-risk-end-support-edge-devices), Required Actions). The timeline below prints the dates. The finding is real. Fortinet advisory FG-IR-26-273, published 10 Nov 2026 and updated 16 Nov 2026, covers CVE-2026-126259, a missing authentication flaw in FortiOS that lets an unauthenticated attacker on the high availability (HA) segment run commands as super admin. Fortinet rates it Critical, CVSS 3 score 9.0, and lists "Known Exploited: No". Fixes are FortiOS 8.0.1, 7.6.8, 7.4.13 and 7.2.14. For FortiOS 7.0, all versions, the solution reads "Migrate to a fixed release", so the fix on that branch is a move to another branch, such as 7.2.14, not a patch. The device is invented: a FortiGate pair at a federal civilian agency, running FortiOS 7.0 on an HA link, with a vendor end-of-support date of 30 Jun 2027. By the example's assumption, the pair's hardware model is not supported by the 7.2 line, so migrating to 7.2.14 is closed to it and replacement is its only way to a fixed release. On 19 Nov 2026 it is still vendor-supported, so the directive does not yet require anything of it, and the CVE is not a CISA catalog entry in the 2026.11.18 feed. Fortinet's text says exploitation needs access to the HA segment, so the agency can narrow the exposure without an upgrade. Timeline of the worked case. Directive dates are from BOD 26-02 and the advisory facts from Fortinet's page for FG-IR-26-273, both read on 19 Nov 2026; the firewall, its vendor end-of-support date, the acceptance, its backstop expiry and the signer are invented. | Date | Step | Response | Real or invented | |---|---|---|---| | 5 Feb 2026 | BOD 26-02 issued | None | Real | | 19 Nov 2026 | The agency's scan matches F-2210 to CVE-2026-126259 on the firewall. The authorizing official accepts that finding, not the age of the device, until the device is replaced. Controls: the HA link sits on a dedicated segment reachable by no other host, and management access is limited to one jump host. Backstop expiry Mon 31 May 2027 (193 days). Replacement is ordered | Mitigate and accept | Real CVE; invented device and decision | | 5 Feb 2027 | Twelve-month deadline. The firewall appears in the inventory of edge devices that will become end of support within the succeeding twelve months. Its end-of-support date is after 5 Feb 2027, so it is not yet due for decommissioning | Accept, review | Real date; invented firewall | | Sat 15 May 2027 | Cutover to the new device, 177 days after the acceptance. Old device powered off and removed from the inventory; a rescan finds nothing at its address; the decommission is reported to CISA on the CISA-provided reporting template (BOD 26-02, Required Actions 3 and 4). The acceptance of F-2210 is closed the same day with the status "avoided", with the decommission record and the rescan attached; the device no longer exists, so nothing remains to accept | Avoid, close | Invented | | 30 Jun 2027 | Vendor end-of-support date, 46 days after cutover | None | Invented device date | | 5 Aug 2027 | Eighteen-month limit for every identified end-of-support edge device, 36 days after the vendor date | None | Real | Two points carry over to any case. First, the acceptance ends at cutover, and its backstop expiry is earlier than the vendor's end-of-support date, so the decision never outlives the device or the support it relies on. If cutover slipped past 31 May 2027, the acceptance would have to be renewed on a new decision, not left open. The firewall is inventoried at twelve months as going end of support, so the twenty-four-month action applies to it: decommission "on or before the date these devices reach EOS", which is 30 Jun 2027 here, earlier than the 5 Aug 2027 eighteen-month limit. Second, a device that cannot be replaced by the directive's date is not an acceptance question for that agency: the agency would contact CISA, and the [BOD 26-04 guidance](https://www.cisa.gov/news-events/directives/bod-26-04-implementation-guidance-prioritizing-security-updates-based-risk) says CISA issues no waivers for directive actions. Outside federal agencies the same sequence works with the organization's own dates, as in [accepting risk on end-of-life software](https://remediationvsmitigation.com/guides/accepting-risk-on-end-of-life-software/) and the [end-of-life row of the Decision Table](https://remediationvsmitigation.com/decision-table/end-of-life-software/). ## The same action under three names Decommissioning is avoidance in SP 800-40 Rev. 4 and remediation in CISA BOD 26-04, which says a vulnerability "shall be remediated (that is, the vulnerability is eliminated through patching, decommissioning the system, or another action)" ([BOD 26-04](https://www.cisa.gov/news-events/directives/bod-26-04-prioritizing-security-updates-based-risk)). A register row for a decommission should therefore record the action, not only the word. A second overlap sits inside NIST's own list. Disabling "a vulnerable feature" is filed under mitigate, while disabling capabilities in assets that can function without them is filed under avoid. Pick by what remains: if the vulnerable code stays installed but unreachable, record mitigation; if it is gone, record avoidance. The [disable a feature or service control](https://remediationvsmitigation.com/compensating-controls/disable-feature-or-service/) covers the first case, and [risk transfer vs acceptance](https://remediationvsmitigation.com/guides/risk-acceptance-vs-risk-transfer/) the remaining response. ## Sources - NIST SP 800-39, Task 3-1: https://nvlpubs.nist.gov/nistpubs/Legacy/SP/nistspecialpublication800-39.pdf (read 19 Nov 2026) - NIST SP 800-40 Rev. 4, Sections 2.1 and 2.2: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf (read 19 Nov 2026) - CISA BOD 26-04, definitions beneath Table 1: https://www.cisa.gov/news-events/directives/bod-26-04-prioritizing-security-updates-based-risk (read 19 Nov 2026) - CISA BOD 26-04 implementation guidance, waiver question: https://www.cisa.gov/news-events/directives/bod-26-04-implementation-guidance-prioritizing-security-updates-based-risk (read 19 Nov 2026) - CISA BOD 26-02, Required Actions: https://www.cisa.gov/news-events/directives/bod-26-02-mitigating-risk-end-support-edge-devices (read 19 Nov 2026) - NIST SP 800-53 Rev. 5 catalog (OSCAL), SA-22 guidance: https://github.com/usnistgov/oscal-content/blob/main/nist.gov/SP800-53/rev5/json/NIST_SP-800-53_rev5_catalog.json (read 19 Nov 2026) - Fortinet PSIRT advisory FG-IR-26-273 (CVE-2026-126259), published 10 Nov 2026, updated 16 Nov 2026: https://fortiguard.fortinet.com/psirt/FG-IR-26-273 (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Risk Acceptance Criteria: 7 Tests From NIST and CISA](https://remediationvsmitigation.com/guides/risk-acceptance-criteria/index.md) - [End-of-Life Software: Replace, Isolate or Accept With Dates](https://remediationvsmitigation.com/decision-table/end-of-life-software/index.md) - [Risk Acceptance Form: Free Template With 15 Fields](https://remediationvsmitigation.com/risk-acceptance-form/index.md) - [Exception Register: 15 Columns, States and Blank CSV](https://remediationvsmitigation.com/exception-register/index.md) - [Risk Transfer vs Risk Acceptance for Security Findings](https://remediationvsmitigation.com/guides/risk-acceptance-vs-risk-transfer/index.md) - [Risk Acceptance vs Risk Tolerance: Who Sets the Limit](https://remediationvsmitigation.com/guides/risk-acceptance-vs-risk-tolerance/index.md) - [Accepting Risk on End-of-Life Software: Limits and Record](https://remediationvsmitigation.com/guides/accepting-risk-on-end-of-life-software/index.md) --- # Risk acceptance vs risk exception Canonical: https://remediationvsmitigation.com/guides/risk-acceptance-vs-risk-exception/ A risk acceptance records who agreed to live with a risk. A risk exception records that an asset may depart from a rule, plan or setting. Two hosts missing the 2 Nov 2026 deadline for CVE-2026-80521 show it: one needs an exception only, the other an exception with an acceptance behind it. ## What the primary sources call each one NIST SP 800-39 defines acceptance as a response to a risk: it "is the appropriate risk response when the identified risk is within the organizational risk tolerance" ([Task 3-1](https://nvlpubs.nist.gov/nistpubs/Legacy/SP/nistspecialpublication800-39.pdf)). In US federal systems the signer is fixed: "The explicit acceptance of risk is the responsibility of the authorizing official and cannot be delegated" ([SP 800-37 Rev. 2, Task R-4](https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-37r2.pdf)). An exception is a departure from something else. SP 800-40 Rev. 4 uses the word for maintenance plans: "Organizations should closely track and monitor all exceptions to maintenance plans," and "having some exceptions is inevitable" ([Section 3.5.5](https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf)). SP 800-53 Rev. 5 uses "deviations": CM-6 c requires organizations to "Identify, document, and approve any deviations from established configuration settings" ([CM-6](https://github.com/usnistgov/oscal-content/blob/main/nist.gov/SP800-53/rev5/json/NIST_SP-800-53_rev5_catalog.json)). Neither text says who approves, or whether residual risk must be accepted. Uses of the two words in the primary sources; all read 19 Nov 2026. The 800-53 clauses are from the OSCAL catalog of Rev. 5 ([source](https://github.com/usnistgov/oscal-content/blob/main/nist.gov/SP800-53/rev5/json/NIST_SP-800-53_rev5_catalog.json)). | Term | What it governs | Who decides in the text | Source | |---|---|---|---| | Risk acceptance | A residual risk that stays | The authorizing official for US federal systems; the policy elsewhere | SP 800-39 Task 3-1; SP 800-37 Rev. 2 Task R-4 | | Exception to a maintenance plan | An asset that cannot follow its patch plan | Not stated; track and review regularly | SP 800-40 Rev. 4, Section 3.5.5 | | Deviation from a configuration setting | A setting that differs from the baseline | "approve" is required; the role is not named | SP 800-53 Rev. 5, CM-6 c | | Permit by exception | A deny-all rule with a list of allowed programs or traffic | Not a risk decision; a control | SP 800-53 Rev. 5, CM-7(5) and SC-7(5) | | Waiver or exception to a federal directive | Not available | Nobody: CISA "does not issue waivers or exceptions" | [BOD 26-04 guidance](https://www.cisa.gov/news-events/directives/bod-26-04-implementation-guidance-prioritizing-security-updates-based-risk) | SP 800-40 Rev. 4 also pushes against exceptions as a habit. "Instead of denoting certain assets or types of assets as 'exceptions,' there should be maintenance groups for them," and assets with similar long-term exceptions "might need to be moved to a separate maintenance group with its own maintenance plan." A medical device or legacy controller that is renewed every quarter as an exception is telling the team to build a plan for that class. ## Which record does a late patch need Decision table for choosing the record. The sources name no approver for exceptions, so the policy must; the first row assumes it does. | Is the asset departing from a rule or plan? | Does a risk remain that someone must own? | Record needed | |---|---|---| | Yes | No: the control or the fix closes the gap inside the exception period | Exception only, approved by whoever owns the plan | | Yes | Yes: the gap is open during the exception | Exception, plus a signed acceptance with its own expiry | | No | Yes: a risk is kept and no rule is bent | Acceptance only | | No | No | Neither; fix it in the normal cycle | The two records have different owners and lifetimes. An exception ends when the plan can be followed or the period lapses. An acceptance ends at its own date or when a reopen trigger fires. One exception can rest on an acceptance, and the [exception register](https://remediationvsmitigation.com/exception-register/) links the two rows; the [risk register versus exception register guide](https://remediationvsmitigation.com/guides/risk-register-vs-exception-register/) covers the layout. ## Two hosts, one late patch CVE-2026-80521 is a use-after-free in the Linux kernel's AF_UNIX garbage collector. An attacker must already run code on the host as an unprivileged user and win a narrow race. The 6.1 stable fix shipped in Linux 6.1.189 on 3 Oct 2026 ([changelog](https://cdn.kernel.org/pub/linux/kernel/v6.x/ChangeLog-6.1.189)), and the CVE was not in the CISA catalog in version 2026.11.18 ([NVD record](https://services.nvd.nist.gov/rest/json/cves/2.0?cveId=CVE-2026-80521), [KEV feed](https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json)). The maintenance plan below is invented: kernel fixes within 30 days of release, so the due date is Mon 2 Nov 2026. Host B needs an exception only; Host A needs an exception with an acceptance behind it. The CVE facts and fix date are real; the hosts, the plan, the people and the approvals are invented. | | Host B, backup appliance | Host A, shared build server | |---|---|---| | Who can run code locally | Three administrators; no unprivileged users | About 40 developers with shell accounts | | Why the plan is missed | Reboot needs a window; the next is Sat 14 Nov 2026 | Release freeze until Tue 15 Dec 2026 | | Late by | 12 days | 43 days | | Risk that remains | None beyond what the three administrators already hold | An unprivileged user could reach the flaw while the patch is waiting | | Record | Exception to the maintenance plan, approved by the plan owner on 30 Oct 2026 | Exception requested 28 Oct 2026; acceptance signed by the security head on 29 Oct 2026, expiry 15 Dec 2026 | | Controls named | None needed | Developer logins limited to a jump host; unusual process creation by the build user alerted; both checked 29 Oct 2026 | | Reopen when | Window moves again | The CVE reaches the catalog; either control fails | | Closed | 14 Nov 2026, after the kernel version check | Planned 15 Dec 2026 | Writing "exception" on Host A's record would have left the signer's name off it. A reviewer reading it would learn which rule was bent but not who accepted the exposure of 40 accounts for 43 days. Writing "acceptance" on Host B's would have sent a signer a decision with no risk in it. The word on the record tells a reader which question was answered. How the end of each period is handled is in [risk exception expiry](https://remediationvsmitigation.com/guides/exception-expiry-and-renewal/), and the tests behind Host A's signature are in [risk acceptance criteria](https://remediationvsmitigation.com/guides/risk-acceptance-criteria/). ## Sources - NIST SP 800-39, Task 3-1: https://nvlpubs.nist.gov/nistpubs/Legacy/SP/nistspecialpublication800-39.pdf (read 19 Nov 2026) - NIST SP 800-37 Rev. 2, Task R-4: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-37r2.pdf (read 19 Nov 2026) - NIST SP 800-40 Rev. 4, Sections 3.4 and 3.5.5: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf (read 19 Nov 2026) - NIST SP 800-53 Rev. 5 catalog (OSCAL), CM-6, CM-7(5), SC-7(5): https://github.com/usnistgov/oscal-content/blob/main/nist.gov/SP800-53/rev5/json/NIST_SP-800-53_rev5_catalog.json (read 19 Nov 2026) - CISA BOD 26-04 implementation guidance, waiver question: https://www.cisa.gov/news-events/directives/bod-26-04-implementation-guidance-prioritizing-security-updates-based-risk (read 19 Nov 2026) - NVD record, CVE-2026-80521: https://services.nvd.nist.gov/rest/json/cves/2.0?cveId=CVE-2026-80521 (read 19 Nov 2026) - kernel.org, ChangeLog-6.1.189: https://cdn.kernel.org/pub/linux/kernel/v6.x/ChangeLog-6.1.189 (read 19 Nov 2026) - CISA Known Exploited Vulnerabilities feed, version 2026.11.18: https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Exception Register: 15 Columns, States and Blank CSV](https://remediationvsmitigation.com/exception-register/index.md) - [Risk Acceptance Form: Free Template With 15 Fields](https://remediationvsmitigation.com/risk-acceptance-form/index.md) - [Risk Acceptance Criteria: 7 Tests From NIST and CISA](https://remediationvsmitigation.com/guides/risk-acceptance-criteria/index.md) - [Risk Register vs Exception Register: How They Link](https://remediationvsmitigation.com/guides/risk-register-vs-exception-register/index.md) - [Risk Exception Expiry: What Happens and How to Renew](https://remediationvsmitigation.com/guides/exception-expiry-and-renewal/index.md) - [Application Allow-Listing as a Compensating Control](https://remediationvsmitigation.com/compensating-controls/application-allow-listing/index.md) --- # Risk acceptance vs risk tolerance Canonical: https://remediationvsmitigation.com/guides/risk-acceptance-vs-risk-tolerance/ Risk tolerance is the level of risk an organization will bear, written once for many decisions. Risk acceptance keeps one named finding because it falls inside that level. NIST SP 800-39 calls acceptance "the appropriate risk response when the identified risk is within the organizational risk tolerance" (read 19 Nov 2026). ## What NIST says tolerance is The NIST glossary lists several definitions of risk tolerance, each taken from a different publication. Acceptance is defined by its response text in SP 800-39, quoted in the next section. Definitions from the NIST CSRC glossary entry for risk tolerance, read 19 Nov 2026. | Source | Wording of "risk tolerance" | What it adds | |---|---|---| | NIST SP 800-53 Rev. 5 (from SP 800-39) | "The level of risk or the degree of uncertainty that is acceptable to an organization." | Covers uncertainty as well as level | | NIST SP 800-137 | "The level of risk an entity is willing to assume in order to achieve a potential desired result." | Tied to a goal | | NIST IR 8286 (adapted from ISO Guide 73) | "The organization's or stakeholder's readiness to bear the remaining risk after responding to or considering the risk in order to achieve its objectives." | Measured on the risk that remains after a response | | NIST SP 800-221 (from OMB Circular A-123) | "The acceptable level of variance in performance relative to the achievement of objectives." | Defined as variance in performance | | NIST SP 800-221 (from ISO Guide 73) | "The readiness of an organization or stakeholders to bear the remaining risk after responding to or considering the risk to achieve its objectives (while recognizing that such tolerance can be influenced by legal or regulatory requirements)" | Law and regulation can narrow it | ## Tolerance is set above the finding, acceptance at it SP 800-39 puts tolerance in the organization-wide tier: "Risk tolerance is the level of risk or degree of uncertainty that is acceptable to organizations and is a key element of the organizational risk frame." It adds that "There is no correct level of organizational risk tolerance", and that the degree is "generally indicative of organizational culture", "potentially different for different types of losses/compromises" and shaped by the individual views of senior executives. SP 800-53 PM-9 says the risk management strategy "includes an expression of the security and privacy risk tolerance for the organization". RA-7 requires that findings from assessments, monitoring and audits be responded to "in accordance with organizational risk tolerance". So the person asked to accept one finding starts from a document someone else wrote. If it is silent on vulnerabilities, the [risk acceptance criteria](https://remediationvsmitigation.com/guides/risk-acceptance-criteria/) have nothing to be tested against. Acceptance is the response at the finding. SP 800-39: "Risk acceptance is the appropriate risk response when the identified risk is within the organizational risk tolerance. Organizations can accept risk deemed to be low, moderate, or high depending on particular situations or conditions." NIST IR 8286, Table 3, adds the working consequence: for Accept, "No additional risk response action is needed except for monitoring." The same table pairs Transfer with risks "that fall outside of tolerance levels"; see [risk transfer versus risk acceptance](https://remediationvsmitigation.com/guides/risk-acceptance-vs-risk-transfer/). The quantity compared with tolerance is residual risk. SP 800-39 Task 3-3 has organizations "determine acceptable degrees of residual risk based on organizational risk tolerance and the specific risk tolerances of particular decision makers". How to state the remaining level is covered in [residual risk after mitigation](https://remediationvsmitigation.com/guides/residual-risk-after-mitigation/). ## Sorting a statement into the right word Appetite is the third word. NIST IR 8286 quotes OMB Circular A-123: appetite "is the broad-based amount of risk an organization is willing to accept in pursuit of its mission/vision", set by the most senior leadership, while tolerance is "generally established at the program, objective, or component level". For a vulnerability record, tolerance is the one that matters, because it can be written for a class of systems or findings. Sorting of example statements by this site, using the SP 800-39 and IR 8286 wording above. Read 19 Nov 2026. | Statement | What it is | Why | Where it lives | |---|---|---|---| | Leadership will take more risk to enter a new market | Appetite | Broad amount of risk in pursuit of the mission | Strategy, board papers | | A day of intranet downtime is bearable, a day of payment downtime is not | Tolerance | Differs by type of loss (SP 800-39) | Risk management strategy (PM-9) | | No internet-facing system may carry a known exploited vulnerability past its CISA due date | Tolerance, written as a limit on findings | Sets what RA-7 responses are compared with | Vulnerability policy | | Accept anything below CVSS 8.0 | A proxy for tolerance, not tolerance | Severity is one input; exploitation can change while the rating stays put (see the case below) | Often found in scanner settings | | Host reporting-01 keeps finding F-4417 until 17 Feb 2027, signed by the director | Acceptance | One finding, one signer, one expiry | [Risk acceptance form](https://remediationvsmitigation.com/risk-acceptance-form/) and [exception register](https://remediationvsmitigation.com/exception-register/) | | Patching on host X may run 10 days past the monthly cycle | Exception to a rule | A departure from a plan, not a statement about risk level | Exception register; see [acceptance versus exception](https://remediationvsmitigation.com/guides/risk-acceptance-vs-risk-exception/) | ## Worked case: a local-access kernel flaw that left tolerance on day 149 The sample organization below is invented. The CVE facts are real, from the NVD and Red Hat records and the CISA catalog read 19 Nov 2026. Its tolerance statement for internal Linux servers: SAMPLE tolerance statement TS-4 (invented organization) The infrastructure director may accept a finding on an internal Linux server for up to 180 days when all four hold: 1. CVSS base score below 8.0. 2. Not on the CISA KEV catalog. 3. The attacker needs local access (CVSS attack vector Local). 4. A mitigation the vendor documents is in place and has been checked. If any condition stops holding, the acceptance ends that day and the finding is treated as unaccepted. CVE-2026-46300 ("Fragnesia") is a Linux kernel flaw in socket buffer coalescing: a local low-privilege user can alter cached copies of read-only files and reach root. NVD gives it a CVSS 3.1 base of 7.8 (AV:L/AC:L/PR:L). Red Hat released the fixed RHEL 9 kernel, kernel-5.14.0-687.10.1.el9_8 (RHSA-2026:19568), on 20 May 2026, and names blocking the esp4 and esp6 IPsec modules as a mitigation where IPsec is not in use. 1. Mon 1 Jun 2026: all four conditions hold (7.8 is below 8.0; not in the catalog; local access; the esp4 and esp6 modules are blocked and a module listing confirms it, on a server that does not use IPsec). The reason, invented, is that the database vendor has not yet certified the fixed kernel for this host. The director signs an acceptance for 180 days, expiring 28 Nov 2026. Field 7 of the [risk acceptance form](https://remediationvsmitigation.com/risk-acceptance-form/) cites TS-4 and field 11 records residual risk as Low. 2. Wed 28 Oct 2026, day 149 of 180: CISA adds the CVE to the KEV catalog, with a due date of 31 Oct 2026, 3 days later. Condition 2 fails. By the statement's own words the acceptance ended on 28 Oct, not on 28 Nov. 3. 28 to 30 Oct 2026: the finding goes to the [decision table](https://remediationvsmitigation.com/decision-table/) as unaccepted. The database vendor's certification is escalated, the fixed kernel is installed and the host rebooted on Fri 30 Oct, a day before the due date. The record keeps the 1 Jun acceptance as closed on 28 Oct, with the reason. A tolerance written only as "CVSS below 8.0" would have left the acceptance in force for 31 more days, until 28 Nov 2026, because the base score stayed 7.8 throughout. The statement also shows the limit of its own wording: NVD has carried CISA's SSVC mark of exploitation "active" for this CVE since 14 May 2026, 18 days before the acceptance, and condition 2 looks only at the catalog. A statement that added "no active exploitation in CISA's SSVC data" would have refused the acceptance on 1 Jun. Which signals the statement names decides when an acceptance ends. ## Sources - NIST CSRC glossary, risk tolerance: https://csrc.nist.gov/glossary/term/risk_tolerance (read 19 Nov 2026) - NIST SP 800-39, Section 2.3.3 and Task 3-1 (risk responses): https://nvlpubs.nist.gov/nistpubs/Legacy/SP/nistspecialpublication800-39.pdf (read 19 Nov 2026) - NIST SP 800-53 Rev. 5 catalog (OSCAL), controls PM-9 and RA-7: https://github.com/usnistgov/oscal-content/blob/main/nist.gov/SP800-53/rev5/json/NIST_SP-800-53_rev5_catalog.json (read 19 Nov 2026) - NIST IR 8286, Tables 2 and 3: https://nvlpubs.nist.gov/nistpubs/ir/2020/NIST.IR.8286.pdf (read 19 Nov 2026) - NIST SP 800-37 Rev. 2, Appendix D roles and Task R-4: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-37r2.pdf (read 19 Nov 2026) - NVD record for CVE-2026-46300, with CISA's KEV date, due date and SSVC data: https://services.nvd.nist.gov/rest/json/cves/2.0?cveId=CVE-2026-46300 (read 19 Nov 2026) - Red Hat CVE page for CVE-2026-46300, errata and mitigation: https://access.redhat.com/security/cve/CVE-2026-46300 (read 19 Nov 2026) - CISA KEV catalog JSON feed, catalogVersion 2026.11.18, entry CVE-2026-46300: https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Risk Acceptance Criteria: 7 Tests From NIST and CISA](https://remediationvsmitigation.com/guides/risk-acceptance-criteria/index.md) - [Risk Transfer vs Risk Acceptance for Security Findings](https://remediationvsmitigation.com/guides/risk-acceptance-vs-risk-transfer/index.md) - [Risk Acceptance Form: Free Template With 15 Fields](https://remediationvsmitigation.com/risk-acceptance-form/index.md) - [Residual Risk After Mitigation: Define, Estimate, Record](https://remediationvsmitigation.com/guides/residual-risk-after-mitigation/index.md) --- # Risk transfer vs risk acceptance Canonical: https://remediationvsmitigation.com/guides/risk-acceptance-vs-risk-transfer/ Risk transfer shifts liability for a risk to another party; risk acceptance keeps it. NIST SP 800-39 says transfer "reduces neither the likelihood of harmful events occurring nor the consequences", so a transferred vulnerability still needs a fix, a mitigation or a dated exception. Sources read 19 Nov 2026. ## What transfer is in three NIST texts SP 800-39 defines the pair this way: "Risk transfer shifts the entire risk responsibility or liability from one organization to another organization (e.g., using insurance to transfer risk from particular organizations to insurance companies). Risk sharing shifts a portion of risk responsibility or liability to other organizations (usually organizations that are more qualified to address the risk)." The other two NIST texts treat transfer as partial. Wording of transfer in three NIST publications, read 19 Nov 2026. | Text | Transfer means | Example given | |---|---|---| | NIST SP 800-39, Task 3-1 | Shifting liability and responsibility when organizations "desire and have the means to shift risk liability and responsibility to other organizations" | Insurance | | NIST IR 8286, Table 3 | "For cybersecurity risks that fall outside of tolerance levels, reduce them to an acceptable level by sharing a portion of the consequences with another party" | Cybersecurity insurance | | NIST SP 800-40 Rev. 4, Section 2.1 | Reduce the risk by sharing consequences with another party | Insurance, or replacing installed software with SaaS "where the SaaS vendor/managed service provider takes care of patching" | ## What transfer leaves in place SP 800-39: "It is important to note that risk transfer reduces neither the likelihood of harmful events occurring nor the consequences in terms of harm to organizational operations and assets, individuals, other organizations, or the Nation." NIST IR 8286 adds that "there are often consequences that cannot be transferred, like loss of customer trust". A vulnerability stays exploitable after a policy is bought; part of the bill moves. SP 800-40 Rev. 4 states the default: "By default, an organization accepts the risk posed by using its software." Transfer is a step in addition to a response to the flaw, never the response itself. The finding still takes one of the outcomes on the [decision table](https://remediationvsmitigation.com/decision-table/). ## Which response each arrangement is Arrangements sorted by this site against the response types in NIST IR 8286 Table 3 and SP 800-40 Rev. 4 Section 2.1. SP 800-40 lists replacing installed software with SaaS under Transfer because it shares consequences with another party. This table files it as Share because SP 800-39 reserves transfer for shifting the entire liability, and the customer keeps its data, configuration and the decision to move. NIST's mitigation covers both fixing the flaw and reducing its likelihood or impact; this site records the first as remediation and the second as mitigation. Read 19 Nov 2026. | Arrangement | Response type | What moves | What stays with you | Record | |---|---|---|---|---| | Cyber insurance policy covering a finding you keep | Accept, plus partial transfer | Part of the financial consequence | Likelihood, the outage, lost customer trust, and the finding itself | Acceptance record with expiry; policy number and exclusions noted | | Move the product to SaaS where the vendor patches | Share (SP 800-40 Rev. 4 files this under Transfer) | The patching task | Configuration, data, and the decision to move; see [SaaS you cannot patch](https://remediationvsmitigation.com/decision-table/saas-cannot-patch/) | Exception or vendor-dependency row until the move is done | | Managed-service contract that names patch times | Share | Responsibility for the action | Liability for missed dates unless the contract says otherwise | Contract clause and owner for checking the provider | | Federal agency buys a policy | Generally not available | Nothing | Everything (SP 800-39: self-initiated transfers by public sector organizations are "generally not possible") | Authorizing official's acceptance in the [risk acceptance form](https://remediationvsmitigation.com/risk-acceptance-form/) | | Decommission the system | Avoid | The whole risk, by removing the asset | Nothing, once it is off the network | Decommission record | | Apply the fix | Remediate (NIST SP 800-39 files it under risk mitigation) | The flaw | Residual risk, usually none | Rescan evidence | | Keep the finding, no policy | Accept | Nothing | All of it | [Exception register](https://remediationvsmitigation.com/exception-register/) row with expiry | ## Worked case: a policy does not move the 19 Sep due date The sample organization is invented, including its policy clause; the CVE, advisory and KEV dates are real (NVD and the Cisco advisory, read 19 Nov 2026). It runs two Cisco ISE nodes on release 3.3 and carries a cyber insurance policy. CVE-2026-76460 is an authentication bypass in an ISE API, CVSS 10.0. Cisco published advisory cisco-sa-ISE-ABP-VNSW7Tn5 on 16 Sep 2026 with fixed patches (for 3.3, Patch 12) and said it was aware of active exploitation. CISA added the CVE to KEV the same day with a due date of 19 Sep 2026. Cisco lists no workaround and names access control lists as its mitigation. SAMPLE policy clause (invented wording; read your own policy) The insurer is not liable for loss arising from a vulnerability for which the software vendor published a fix more than 30 days before the loss, where the insured did not apply the fix or an equivalent control. 1. 16 Sep 2026: the security lead asks whether the policy covers the exposure until the maintenance window. The answer is partial transfer at most: it moves some cost, leaves the exploitable API in place, and does not change the 19 Sep date the organization has adopted from CISA. 2. 17 Sep to Wed 23 Sep: access to the ISE nodes is limited with access control lists, a dated exception covers the gap between 19 and 23 Sep, and ISE 3.3 Patch 12 goes on both nodes on 23 Sep. The steps, owners and evidence for that sequence are in [mitigate vs manage risk](https://remediationvsmitigation.com/guides/mitigate-vs-manage-risk/). The register records the response as Mitigate and notes the policy in a separate field as partial transfer of financial consequence. 3. 16 Oct 2026: the sample clause's 30-day mark (16 Sep plus 30 days). A team that had treated the policy as the response would have been uninsured for this flaw from that day. The patch on 23 Sep, day 7, is well inside it. The point of the case: the transfer instrument can itself require the fix. Read exclusions before writing "transferred" in a record. ## The public sector case SP 800-39 states that "the concept of risk transfer is less applicable in the public sector (e.g., federal, state, local governments) than the private sector, as liability of organizations is generally established by legislation or policy". For a federal system a policy purchase cannot be the stated response, and the authorizing official's acceptance is still needed. Risk sharing does remain available: SP 800-39 says it "may be a sharing of liability or a sharing of responsibility for other, adequate risk responses such as mitigation". ## Transfer line for the exception record Transfer or sharing (complete only if used) - What was shifted: liability / responsibility / financial consequence (choose, per SP 800-39) - To whom, and the instrument: policy or contract number, effective dates - What was not shifted: likelihood, outage, loss of customer trust (IR 8286) - Exclusions read: yes / no, and any that depend on patching - Response that still applies to the flaw: fix date, mitigation or dated exception (link the exception row) ## Sources - NIST SP 800-39, Section 2.3.3 and Task 3-1 (risk responses): https://nvlpubs.nist.gov/nistpubs/Legacy/SP/nistspecialpublication800-39.pdf (read 19 Nov 2026) - NIST IR 8286, Tables 2 and 3: https://nvlpubs.nist.gov/nistpubs/ir/2020/NIST.IR.8286.pdf (read 19 Nov 2026) - NIST SP 800-40 Rev. 4, Section 2.1: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf (read 19 Nov 2026) - NVD record for CVE-2026-76460, with the CISA KEV date and due date it carries: https://services.nvd.nist.gov/rest/json/cves/2.0?cveId=CVE-2026-76460 (read 19 Nov 2026) - Cisco advisory cisco-sa-ISE-ABP-VNSW7Tn5: https://sec.cloudapps.cisco.com/security/center/content/CiscoSecurityAdvisory/cisco-sa-ISE-ABP-VNSW7Tn5 (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Risk Acceptance vs Risk Tolerance: Who Sets the Limit](https://remediationvsmitigation.com/guides/risk-acceptance-vs-risk-tolerance/index.md) - [SaaS You Cannot Patch: Mitigate and Track the Vendor](https://remediationvsmitigation.com/decision-table/saas-cannot-patch/index.md) - [Risk Acceptance Form: Free Template With 15 Fields](https://remediationvsmitigation.com/risk-acceptance-form/index.md) - [Mitigating Risk vs Managing Risk, With a Worked Case](https://remediationvsmitigation.com/guides/mitigate-vs-manage-risk/index.md) --- # Who is the risk owner Canonical: https://remediationvsmitigation.com/guides/risk-owner/ A risk owner is the designated party responsible and accountable for keeping a risk within enterprise requirements, as NIST IR 8286 defines it. The risk owner is not automatically the system owner or the signer: in a federal system only the authorizing official can accept risk. NIST wording read 19 Nov 2026. ## The definition and where it is used NIST IR 8286 lists the risk owner as a field of the cybersecurity risk register: "The designated party responsible and accountable for ensuring that the risk is maintained in accordance with enterprise requirements. The Risk Owner may work with a designated Risk Manager who is responsible for managing and monitoring the selected risk response". NIST IR 8286B, Section 2.4.3, says the field records "the personnel and/or organizational element responsible for ensuring that the described risk response is implemented", chosen "at the discretion of the enterprise but should be consistently used". IR 8286C is stricter about the person: "The party designated as the risk owner must be continually knowledgeable about relevant risk conditions and must also have the accountability and authority to manage the risk." It asks that responsibility "be periodically reviewed (e.g., monthly) to ensure that the risk owner is the appropriate designee". ## Which role answers which question Role wording as read on 19 Nov 2026. | Role | Question it answers | What the source says | |---|---|---| | Risk owner | Who owns the loss if the finding is exploited? | "in whom the organization has invested the authority and accountability for making risk-based decisions and who owns the loss associated with a realized risk scenario" (NIST IR 8286B) | | Risk manager | Who recommends the treatment? | "may or may not be the system owner"; helps recommend treatment (NIST IR 8286B) | | System owner | Who schedules the patch and runs the change? | "an organizational official responsible for the procurement, development, integration, modification, operation, maintenance, and disposal of a system" (NIST SP 800-37 Rev. 2, Appendix D); the provider that operates the system when it is hosted | | Authorizing official | Who signs a dated acceptance in a federal system? | "the only organizational official who can accept the security and privacy risk to organizational operations, organizational assets, and individuals" (NIST SP 800-37 Rev. 2, Appendix D and Task R-4) | | Authorizing official designated representative | Who runs day-to-day risk activities for the authorizing official? | The one activity that "cannot be delegated" to the representative is "the authorization decision and signing of the associated authorization decision document (i.e., the acceptance of risk)" (NIST SP 800-37 Rev. 2, Appendix D) | | Signer outside the federal setting | Who signs when SP 800-37 does not apply? | The role named in your own policy; see [who may sign](https://remediationvsmitigation.com/risk-acceptance-form/who-may-sign/) | | More senior stakeholder named in advance | Who decides when a time limit or threshold is crossed (escalation), or when the matter is sensitive or critical (elevation)? | Escalation by time frame or risk condition, elevation of the response decision (NIST IR 8286B) | IR 8286B separates the last two terms. "Risk Escalation occurs when a particular threshold is reached, either based on a time frame or some other risk condition", for example "a risk that has remained through more than two fiscal periods without adequate treatment". "Risk Elevation is the process of transferring the decisions on risk response to a more senior stakeholder when the factors involved (e.g., a regulatory compliance risk) are particularly sensitive or critical." Both need criteria written in advance, so an expired [exception](https://remediationvsmitigation.com/exception-register/) has a defined next owner. ## Worked case: SharePoint, 11 Aug to 14 Oct 2026 The sample organization is invented; the CVE, Microsoft and CISA dates are real (NVD, the Microsoft Security Response Center update guide and the CISA KEV feed, version 2026.11.18, read 19 Nov 2026). Its on-premises SharePoint Server 2019 farm of four servers holds leave and payroll forms. Roles in its policy: risk owner, the Director of HR Operations; system owner, the Head of Collaboration Platforms; signer for any date past a CISA due date, the CISO. 1. 11 Aug 2026: Microsoft publishes CVE-2026-65660 with the August security updates, as revision 1.0 titled a spoofing issue with a CVSS 3.1 base score of 6.5. On 14 Aug the owner and system owner triage it as Medium on that rating, because 6.5 falls in the Medium band of CVSS 3.1 (4.0 to 6.9), and set the patch for the 30 Sep cycle. The risk register entry names the owner and a backup, per IR 8286B. 2. 27 Aug: Microsoft's revision 1.1 updates the impact, title and FAQs; the vulnerability is now titled SharePoint Server Remote Code Execution, with CVSS 3.1 base score 8.8 (up from 6.5). Nobody is notified; the entry still reads Medium. 3. 14 Sep: the monthly owner review (the IR 8286C example interval) picks up the revision, 18 days after it. The owner re-rates the entry High and moves the patch to 5 Oct. 4. 25 Sep: Microsoft revision 1.2 records evidence of observed attacks and CISA adds the CVE to KEV with a due date of 28 Sep. The sample policy's escalation rule (any KEV listing goes to the CISO the same day) fires. The CISO, as signer, rules that the 5 Oct date cannot stand. 5. 27 Sep: the system owner installs KB5002894 and KB5002896 (the SharePoint Server 2019 updates, fixed build 16.0.10417.20198) on all four servers, one day inside the 28 Sep due date, and reports the build. The owner confirms; the CISO closes the entry. 6. 14 Oct: the next monthly review confirms the owner is still the right designee, as IR 8286C asks, and the closed entry keeps the dates above. Three roles acted on one finding. The risk owner acted four times: 14 Aug, 14 Sep, 27 Sep and 14 Oct; the system owner on 14 Aug and 27 Sep; the CISO on 25 Sep and 27 Sep. The gap that mattered was 27 Aug to 14 Sep, a monthly review catching a vendor revision; the escalation rule did the rest. A weekly feed check on vendor revisions would close that gap, and the record would show who is assigned to it. ## How owner assignment fails IR 8286C asks for an owner with "the accountability and authority to manage the risk" and for periodic review of the designation. Three failures break that, and each leaves a trace in the finding record. The causes are this site's reading; the requirement is NIST's. Failure modes of risk owner assignment, with the NIST wording they depart from named in the text. Read 19 Nov 2026. | Failure | What happens | What the record shows | Repair | |---|---|---|---| | Orphaned owner after a reorganization | The named person or role no longer exists, so escalation criteria point at nobody and the finding ages | A "last confirmed" date older than the review interval (IR 8286C gives monthly as an example), or an owner field naming a retired role | Name a role and a backup, not only a person; re-confirm the owner whenever the organization chart changes | | Owner is the requester | The person who asks for the exception also owns the risk and signs it, so no second check exists | The requester and approver fields hold the same role | Keep the approver a separate role, as in the [sample signing rule](https://remediationvsmitigation.com/risk-acceptance-form/who-may-sign/) | | Shared platform with no single business unit | Several units use one farm or service, none has the authority to decide for the others, and the owner defaults to IT operations, which runs the system but does not bear the loss | Owner and system owner hold the same role; the affected business units appear nowhere | Name an owner at the level that holds authority over all users of the platform, list the units affected, and use elevation (IR 8286B) when they disagree | ## When the usual answer is wrong The usual split is the risk owner decides, the system owner patches and the signer accepts. Two setups change it. - **Vendor-hosted software.** The vendor is the system owner and schedules the fix, so the customer's owner cannot schedule the patch. IR 8286B gives the example: "a system owner may be a cloud service provider that operates a hosted application, while the risk owner may be a corporate business unit that processes important data within that application". The customer's business unit still owns the loss and decides whether to wait, add a control or leave the service. The record carries a vendor-dependency row until the vendor ships; see [SaaS you cannot patch](https://remediationvsmitigation.com/decision-table/saas-cannot-patch/). - **Shared services.** A shared service has one provider-side owner and several consuming business units. Each consumer owns its own loss, so the register holds one row for the platform finding and one owner per consuming unit, with the provider as system owner on all of them. This is this site's practice; IR 8286B supplies the terms. ## What the finding record holds - One named risk owner, with a backup if the enterprise uses one (IR 8286B). - The system owner or operating provider, as a separate field. A finding waiting on a fix also carries a [plan of action and milestones](https://remediationvsmitigation.com/guides/poam-plan-of-action-and-milestones/) entry. - A separate approver role for acceptance; for a federal system the authorizing official (SP 800-37 Rev. 2). - The date the owner was last confirmed, against the monthly example in IR 8286C. - The escalation and elevation criteria that apply, copied from policy. ## Sources - NIST IR 8286, Tables 2 and 3: https://nvlpubs.nist.gov/nistpubs/ir/2020/NIST.IR.8286.pdf (read 19 Nov 2026) - NIST IR 8286B, Sections 2.4.3 and 2.4.4: https://nvlpubs.nist.gov/nistpubs/ir/2022/NIST.IR.8286B.pdf (read 19 Nov 2026) - NIST IR 8286C, risk register guidance: https://nvlpubs.nist.gov/nistpubs/ir/2022/NIST.IR.8286C.pdf (read 19 Nov 2026) - NIST SP 800-37 Rev. 2, Appendix D roles and Task R-4: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-37r2.pdf (read 19 Nov 2026) - NVD record for CVE-2026-65660, with the CISA KEV date and due date it carries: https://services.nvd.nist.gov/rest/json/cves/2.0?cveId=CVE-2026-65660 (read 19 Nov 2026) - Microsoft Security Response Center, August 2026 CVRF, CVE-2026-65660 revision history: https://api.msrc.microsoft.com/cvrf/v3.0/cvrf/2026-Aug (read 19 Nov 2026) - CISA KEV catalog entry for CVE-2026-65660: added 25 Sep 2026, due 28 Sep 2026: https://www.cisa.gov/known-exploited-vulnerabilities-catalog?field_cve=CVE-2026-65660 (read 19 Nov 2026) - Microsoft Security Response Center, CVE-2026-65660 advisory record (revision 1.0 spoofing, CVSS 6.5): https://api.msrc.microsoft.com/sug/v2.0/en-US/vulnerability/CVE-2026-65660 (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Who Can Accept Risk: Signer by Standard (2026)](https://remediationvsmitigation.com/risk-acceptance-form/who-may-sign/index.md) - [Exception Register: 15 Columns, States and Blank CSV](https://remediationvsmitigation.com/exception-register/index.md) - [Risk Register vs Exception Register: How They Link](https://remediationvsmitigation.com/guides/risk-register-vs-exception-register/index.md) --- # Risk register vs exception register Canonical: https://remediationvsmitigation.com/guides/risk-register-vs-exception-register/ A risk register records risks and the responses chosen; an exception register records approved departures from a rule, plan or setting, each with an end date. NIST IR 8286 lists 12 risk register elements; the exception register has 15 columns. Two fields join them: finding_ref and vulnerability_state. ## Two records, two questions NIST's glossary gives two definitions of a risk register. One, from OMB Circular A-11, is "A repository of risk information including the data understood about risks over time." The other, from NISTIR 8170, is "A central record of current risks, and related information, for a given scope or organization." It adds that current risks are both accepted risks and risks with a planned mitigation path. So a risk register already holds accepted risks. It has no lifecycle for each departure from a rule, which is the job of the [exception register](https://remediationvsmitigation.com/exception-register/). ## The 12 elements against the 15 columns Elements from NIST IR 8286, Table 2. All 12 appear: priority, category, likelihood, impact, exposure and response cost are in the row marked none, and the other six are paired with columns. The pairing is a reading of the two. Read 19 Nov 2026. | NIST IR 8286 element | Exception register column | Reading | |---|---|---| | ID (Risk Identifier) | exception_id | Each register carries its own ID: the risk register holds the risk ID (R-0207 in the case below), the exception register holds exception_id (EX-0031). The exception row names the risk ID in finding_ref, and the risk row names the exception ID in its response description | | Risk Description | finding_ref, asset | The register holds the pointer; the form holds the scenario | | Risk Response Type | response | IR 8286 has accept, transfer, mitigate, avoid; the exception register has accepted, interim mitigation, vendor dependency | | Risk Response Description | control_ids, plan_milestone_date | Controls link to the compensating control entries | | Risk Owner | approver_role | Not the same: the approver signs, the owner manages; see [who is the risk owner](https://remediationvsmitigation.com/guides/risk-owner/) | | Status | vulnerability_state | IR 8286 leaves terms to the organization; the register fixes six FedRAMP values for the finding | | Priority, Risk Category, Current Assessment (likelihood, impact, exposure), Risk Response Cost | none | Held on the [risk acceptance form](https://remediationvsmitigation.com/risk-acceptance-form/) or the risk register | | none | exception_state, decision_date, expiry_date, next_review_date, vendor_last_checkin, closed_date, closure_evidence | The state of the exception itself, its dates and the closing proof are what the exception register adds | The exception register has 15 columns. Eight are paired in the table (exception_id, finding_ref, asset, response, control_ids, plan_milestone_date, approver_role and vulnerability_state). Seven are its own: exception_state, decision_date, expiry_date, next_review_date, vendor_last_checkin, closed_date and closure_evidence. Two fields join the registers. First, finding_ref holds the risk register's ID (R-0207) beside the CVE ID, so an auditor can move from a departure to the risk it covers; the risk register's response description holds the exception_id (EX-0031) for the way back. Second, vulnerability_state takes the same six values as the risk register's Status, so the two rows cannot disagree about the finding. The approver and the risk owner are not a join: approver_role records who signed, and the risk owner field records who manages the risk, so the signature and the accountability can differ without confusing the record. ## Who keeps each register, and how often Maintenance and review of the two registers. The risk register column rests on NIST IR 8286 and IR 8286C; the exception register column is this site's routine. Read 19 Nov 2026. | | Risk register | Exception register | |---|---|---| | Kept by | The risk owner for each entry; ERM leadership "should ensure that accountability" (NIST IR 8286) | The register owner, who moves rows between states ([exception register](https://remediationvsmitigation.com/exception-register/)) | | Reviewed | On the enterprise's reporting cycle: IR 8286C says that where risk is reported quarterly, register processes "need to cycle at least quarterly", and gives monthly as an example interval for confirming the owner | Monthly, by next_review_date; a row enters "renewal due" 30 days before expiry (this site's routine) | | A change to one entry | Re-rating, new response, new owner | New decision, renewal as a new row, or closure with evidence | ## Where an entry goes Placement by this site from the cited NIST texts. Read 19 Nov 2026. | Situation | Risk register | Exception register | POA&M | |---|---|---|---| | Accepted risk, no fix planned | Yes | Yes, with expiry | No: SP 800-37 Rev. 2 says entries "are not necessary when deficiencies are accepted by the authorizing official as residual risk" | | Mitigation waiting for a fix | Yes | Yes | Yes: SP 800-53 RA-7 generates a plan entry when a mitigation "cannot be completed immediately"; see the [POA&M guide](https://remediationvsmitigation.com/guides/poam-plan-of-action-and-milestones/) | | Exception to a patching schedule, risk response unchanged | No | Yes (SP 800-40 Rev. 4 asks that exceptions be tracked closely) | No | | False positive | No: closed with evidence | Only if a row was opened | No | ## Statuses NIST IR 8286C suggests a status that could be "a simple indicator (e.g., open, closed, pending, waived, transferred)" or a more detailed explanation. IR 8286B says the terms are "at the discretion of the organization, but the options available should be specified in the risk management policy and/or procedures to enable consistent review". The exception register takes the stricter route for vulnerability_state and uses six values from FedRAMP's 2026 definitions. The risk register row in the case below uses the same six values, so both registers describe one finding in one vocabulary. exception_state is separate (active, renewal due, lapsed, escalated, closed) and has no risk register counterpart: FedRAMP Consolidated Rules for 2026, definitions, read 19 Nov 2026. | Status | FedRAMP definition (abridged) | |---|---| | accepted | "A vulnerability that the provider does not intend to fully mitigate or remediate", or that has not been or will not be within the maximum overdue period | | overdue | "intends to fully mitigate or remediate but has not or will not do so within the time frames recommended or required by FedRAMP" | | partially mitigated | "likelihood or Potential Agency Impact N-rating has been reduced from the original evaluation but the risk of exploitation still exists and the vulnerability is still detected" | | fully mitigated | reduced "until either are negligible, but the vulnerability is still detected" | | remediated | "neutralized or eliminated and is no longer detected" | | false positive | "not actually present in an exploitable state in the information resource" | ## Worked case: one ScreenConnect finding in both registers The sample organization is invented; the CVE, vendor and CISA dates are real (NVD, the [ConnectWise bulletin](https://www.connectwise.com/company/trust/security-bulletins/2026-09-08-screenconnect-bulletin) and [advisory page](https://www.connectwise.com/company/trust/advisories), and the CISA KEV feed, read 19 Nov 2026). It runs ScreenConnect for its help desk, with 40 technician clients. CVE-2026-84869 (file transfer flaw, CVSS 9.9) was fixed in version 26.6.5, released 8 Sep 2026. CISA added it to KEV on 11 Sep with a due date of 14 Sep. ConnectWise's interim step, published 3 Sep, was to deselect the TransferFiles permission in every role. Sample records using the exception register columns and FedRAMP's six status values. Vendor and CISA dates are real (ConnectWise bulletin, NVD and KEV feed 2026.11.18); the organization is invented. | Date (2026) | Event | Risk register R-0207 | Exception register EX-0031 | |---|---|---|---| | 4 Sep | Interim step applied: TransferFiles deselected in all roles | Opened. Response: mitigate. Owner: Head of IT Service Desk. Status: partially mitigated | No row: no rule has been departed from | | 8 Sep | 26.6.5 released | Plan: upgrade server and clients by 14 Sep | No row | | 11 Sep | CISA adds the CVE to KEV, due 14 Sep | Exposure re-rated; escalated to the CISO | No row | | 14 Sep | Server upgraded. 11 of 40 clients at a branch office cannot be reached before the due date | Status: partially mitigated. Response description: exception EX-0031 | Opened. finding_ref: CVE-2026-84869, R-0207; response: interim mitigation; exception_state: active; vulnerability_state: partially mitigated; approver_role: CISO; decision_date 14 Sep; expiry_date 25 Sep; control_ids: the TransferFiles role change; plan_milestone_date 21 Sep | | 15 Sep | Due date has passed | Status: overdue | vulnerability_state: overdue (exception_state stays active) | | 21 Sep | 10 of 11 clients upgraded; 1 laptop is back in the office on 24 Sep | Unchanged | next_review_date 24 Sep | | 24 Sep | Last client upgraded; rescan shows no detection | Closed. Status: remediated | closed_date 24 Sep; closure_evidence: rescan reference; exception_state: closed; vulnerability_state: remediated | Both rows share the finding, the dates and the final status, but they answer different questions. The risk register shows that the risk was owned and re-rated; the exception register shows that the 14 Sep rule was departed from for 10 days under a signature and that the departure ended with evidence. The word overdue in the 15 Sep rows is this site's convention: FedRAMP defines overdue against FedRAMP's own time frames, and here it is applied to the missed KEV due date of 14 Sep. ## Sources - NIST CSRC glossary, risk register: https://csrc.nist.gov/glossary/term/risk_register (read 19 Nov 2026) - NIST IR 8286, Tables 2 and 3: https://nvlpubs.nist.gov/nistpubs/ir/2020/NIST.IR.8286.pdf (read 19 Nov 2026) - NIST IR 8286B, Sections 2.4.3 and 2.4.4: https://nvlpubs.nist.gov/nistpubs/ir/2022/NIST.IR.8286B.pdf (read 19 Nov 2026) - NIST IR 8286C, risk register guidance: https://nvlpubs.nist.gov/nistpubs/ir/2022/NIST.IR.8286C.pdf (read 19 Nov 2026) - NIST SP 800-37 Rev. 2, Appendix D roles and Task R-4: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-37r2.pdf (read 19 Nov 2026) - NIST SP 800-53 Rev. 5 catalog (OSCAL), controls PM-9 and RA-7: https://github.com/usnistgov/oscal-content/blob/main/nist.gov/SP800-53/rev5/json/NIST_SP-800-53_rev5_catalog.json (read 19 Nov 2026) - FedRAMP Consolidated Rules for 2026, definitions of vulnerability terms: https://www.fedramp.gov/2026/definitions/ (read 19 Nov 2026) - NVD record for CVE-2026-84869, with the CISA KEV date and due date it carries: https://services.nvd.nist.gov/rest/json/cves/2.0?cveId=CVE-2026-84869 (read 19 Nov 2026) - ConnectWise security bulletin, ScreenConnect 26.6.5, 8 Sep 2026 (CVE-2026-84869, TransferFiles mitigation): https://www.connectwise.com/company/trust/security-bulletins/2026-09-08-screenconnect-bulletin (read 19 Nov 2026) - ConnectWise trust center advisories, interim Guest File Transfer advisory of 3 Sep 2026: https://www.connectwise.com/company/trust/advisories (read 19 Nov 2026) - CISA KEV feed, catalog version 2026.11.18: CVE-2026-84869 added 11 Sep 2026, due 14 Sep 2026: https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Exception Register: 15 Columns, States and Blank CSV](https://remediationvsmitigation.com/exception-register/index.md) - [Risk Acceptance vs Risk Exception in Cyber Security](https://remediationvsmitigation.com/guides/risk-acceptance-vs-risk-exception/index.md) - [POA&M Example: A Filled Plan of Action and Milestones Entry](https://remediationvsmitigation.com/guides/poam-plan-of-action-and-milestones/index.md) - [Who Is the Risk Owner for a Security Finding?](https://remediationvsmitigation.com/guides/risk-owner/index.md) --- # How to verify a mitigation works Canonical: https://remediationvsmitigation.com/guides/verifying-a-mitigation-works/ Verifying a mitigation means testing the control against the path the flaw needs, then re-checking it on a fixed interval. NIST SP 800-40 Rev. 4 makes these two steps. FedRAMP's 2026 rules set drift checks from every 7 days (Class D) to every 3 months (Class A), read 19 Nov 2026. ## Two steps, and a record that shows both NIST SP 800-40 Rev. 4, Section 2.2, lists step c, "Verify the risk response", and step d, "Continuously monitor the risk response". For controls, step c reads "For deploying additional security controls, ensure they are functioning as intended." Step d reads: "Make sure that the risk response continues to be in place: no one uninstalls the patch, deactivates the additional security controls, lets the cybersecurity insurance lapse, or restarts the decommissioned asset." A mitigation record with an install date and no check date has completed neither step. The first check answers whether the control closes the path the flaw needs. It is not a check that the control exists. A rule that is installed in log-only mode exists and blocks nothing; a setting that was changed on one node of three exists too. So each check below reads the state that matters for that flaw, and each says who does it and how it usually fails. ## The check for five real mitigations These come from the CVE records in the site's data, read 19 Nov 2026. The mitigation is the vendor's or, where marked, a general control. The check reads configuration or logs on a system you own; none of it needs an exploit. Verification step for five mitigations. Sources: Fortinet FG-IR-26-175, Palo Alto CVE-2026-0322, Node.js October 2026 security releases with ELLIO and Aikido analysis, SonicWall SNWLID-2026-0014, Atlassian CVE-2026-21589 advisory. Read 19 Nov 2026. | CVE and mitigation | Verification step | Owner | How it usually fails | |---|---|---|---| | CVE-2026-104286, FortiMail: turn off the IBE service (Fortinet FG-IR-26-175) | In the GUI open Encryption > IBE and read IBE Service. In the CLI run show system encryption ibe, and confirm the status line reads disable. Then confirm from outside the trusted network that the web interface is not published, if that was the chosen option. | Messaging administrator | Status was changed on one appliance of a pair. A later configuration restore or upgrade brings back set status enable. Users who need IBE mail complain and someone re-enables it. | | CVE-2026-0322, PAN-OS: limit the management interface to one jump box (Palo Alto) | Open Device > Setup > Interfaces > Management Interface Settings and read Permitted IP Addresses: one address, the jump box. Then open Network > Network Profiles > Interface Mgmt and list every profile that allows HTTPS or SSH, and which data-plane interfaces use it. | Firewall administrator | The management address is closed but an interface management profile on a data-plane interface still allows HTTPS. Panorama or a second firewall was not included. Palo Alto itself rates the jump-box case 8.5 instead of 8.6, so this control lowers risk only slightly. | | CVE-2026-111497, Node.js and Brotli: reject inbound requests with Content-Encoding br at the proxy (researcher stopgap; the Node.js advisory lists no workaround) | From a test client you control, send a request with Content-Encoding br and an empty or harmless body to a staging endpoint, and confirm the proxy answers with an error before the application sees it. Check the proxy log for the rejected request. Confirm the rule is on every proxy node. | Platform or web operations | The rule covers inbound requests only. Brotli responses your own application fetches from other servers still reach the decoder. A node behind a different load balancer has no rule. | | CVE-2026-19868, SonicOS SSLVPN: disable or delete every LDAP, RADIUS and TACACS+ server (SonicWall workaround of 9 Nov 2026) | Open Device > Users > Settings > Authentication, then Configure RADIUS, Configure LDAP and Configure TACACS+. In each Servers list, confirm none is listed and enabled. Then sign in once with each login type that should still work, such as the local admin. | Network security administrator | Someone re-adds a server for administrator logins and reopens the condition. Or the check stops at one of the three lists. | | CVE-2026-21589, Atlassian Data Center: block parent-directory sequences with a WAF rule or the vendor's rewrite rule | For the rewrite option, check conf/server.xml for the RewriteValve line and WEB-INF/rewrite.config for the rule on every node (Bitbucket: the first rule in app/WEB-INF/urlrewrite.xml). Atlassian says to test that the rule blocks the encoded and double-encoded forms, so run that test on a staging copy and keep the test request and its forbidden or 404 response in the record. | Application administrator | One node of a cluster was not restarted after the edit, or a node was rebuilt from an older image. The rule matches only the plain form and not the encoded ones. | ## A check that passes on a vulnerable build For CVE-2026-111497 the obvious version check is wrong. ELLIO reports that process.versions.brotli shows 1.2.0 on both patched and vulnerable Node.js builds, because the fix was cherry-picked without changing the version string. The check that works is node --version on the host and inside each image, compared with 22.23.4, 24.22.1 or 26.12.1. A Dockerfile pinned to node:24.22.0 stays vulnerable until the pin moves. If Node.js links the system Brotli library, the check is the distribution package version instead. The lesson holds beyond this CVE: verify with the identifier the vendor says decides the outcome, and confirm it can tell a fixed system from an unfixed one. ## Worked example: FortiMail IBE switched off on 1 Oct 2026, drift found on 8 Oct Fortinet published FG-IR-26-175 on Thu 1 Oct 2026 and said the flaw was exploited. CISA added CVE-2026-104286 to the KEV catalog the same day with a due date of 4 Oct. Fortinet's solution update of 5 Oct 2026 lists FortiMail 7.4.9 as the fixed 7.4 build. The record below is an illustration: the vendor dates are real, the organization's dates and the cause of the drift are assumed. The organization is not a federal agency, so the 4 Oct date is a target for it, not a directive. It runs FortiMail 7.4.8 and picks the weekly interval from the FedRAMP Class D ladder. Illustrative record for one appliance. Vendor dates from Fortinet FG-IR-26-175 (published 1 Oct, solution update 5 Oct, cloud clarification 7 Oct 2026) and CISA KEV, read 19 Nov 2026. | Date | Event | Record entry | |---|---|---| | Thu 1 Oct 2026 | IBE Service set to off on the appliance. Step c: status read as disable, and the web interface confirmed closed to the internet. | Change ticket, configuration export, name of the person who read the status, next check 8 Oct | | Mon 5 Oct 2026 | Fortinet's solution update lists 7.4.9 as the fixed 7.4 build. The upgrade is booked for the Sat 10 Oct window and the mitigation stays until then. | Advisory revision read, upgrade booking, mitigation kept with removal tied to the upgrade | | Tue 6 Oct 2026 (assumed) | A configuration is restored from a backup taken before 1 Oct. Nobody reads the status. | None: the setting is silently enable again | | Thu 8 Oct 2026 | Interval check finds set status enable. The finding is opened as drift, IBE is turned off again, the restore procedure is amended, and the appliance logs are checked against the indicators Fortinet lists, because the flaw was exploitable while the setting was on. | Check date, result, drift finding, logs reviewed, next check 10 Oct | | Sat 10 Oct 2026 | Appliance upgraded to 7.4.9. Version read as 7.4.9 on the appliance. The IBE setting is no longer needed as a control, so it is returned to the state the business wants, and the mitigation is marked removed. | Upgrade date, version read, mitigation removal date 10 Oct, closure | The 8 Oct row is the case the record exists for. Without the weekly check, the restore would have stood until the 10 Oct upgrade or the next audit, and the register would still have said "mitigated". The residual figure on the [residual risk page](https://remediationvsmitigation.com/guides/residual-risk-after-mitigation/) is only as old as the last check, so the next check date belongs in the entry. ## How often to re-check FedRAMP's rule on drift is the only published ladder of intervals. It defines drift as "Changes to information resources that cause deviations from the intended and assessed state; common forms of drift include changes to configurations, deployed software, privileges, running processes, and availability." The table is for detection on resources likely to drift. Non-machine-based resources must be verified and validated at least every 3 months (VDR-TFR-NMV). FedRAMP Consolidated Rules for 2026, Vulnerability Detection and Response, read 19 Nov 2026. | FedRAMP class | Drift detection (VDR-TFR-PDD, SHOULD) | Machine-based status check (VDR-TFR-MVF) | |---|---|---| | Class A | At least once every 3 months | Not listed | | Class B | At least once every month | SHOULD, at least monthly | | Class C | At least once every 14 days | MUST, at least monthly | | Class D | At least once every 7 days | MUST, at least monthly | These are provider timeframes. An organization with no rule of its own can adopt the ladder and write the interval on the [exception register](https://remediationvsmitigation.com/exception-register/). A KEV-listed flaw on an internet-facing system belongs at the weekly end. ## Which exposure result settles the question When the mitigation is meant to change exposure, the method matters. FedRAMP says prevention that intercepts, inspects or filters triggering payloads means that, "once this prevention is in place the vulnerability should no longer be considered an internet-reachable vulnerability". CISA's BOD 26-04 is stricter about proof: agencies "may use multiple overlapping systems to determine whether an asset or set of assets are publicly exposed", and "If one such method finds that a system is publicly exposed, the value for the decision point should be set to Yes." One method showing the asset closed does not outvote another showing it open. The control itself is covered under [removing internet exposure](https://remediationvsmitigation.com/compensating-controls/remove-internet-exposure/). ## Check without weakening the system - FedRAMP rule VDR-CSO-MSP says providers "SHOULD NOT weaken the security of information resources to facilitate vulnerability scanning, detection, or assessment activities." Do not switch a rule to log-only or open a port to make it easier to test. - NIST SP 800-115 says "Because of the risk of operational impact, it is recommended that an established incident response plan be in place during testing." - For OT, SP 800-82 Rev. 3 recommends "A complete verification" to ensure the controls "are not affecting or degrading the performance and safety of the OT system". ## Sources - NIST SP 800-40 Rev. 4, Section 2.2: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf (read 19 Nov 2026) - FedRAMP 2026, Vulnerability Detection and Response: VDR-TFR-PDD, MVF, NMV, CSO-MSP: https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-detection-and-response/ (read 19 Nov 2026) - FedRAMP 2026 definitions: Drift: https://www.fedramp.gov/2026/definitions/ (read 19 Nov 2026) - FedRAMP 2026, Vulnerability Evaluation and Reporting, internet-reachable note: https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-evaluation-and-reporting/ (read 19 Nov 2026) - CISA BOD 26-04, Appendix A: https://www.cisa.gov/news-events/directives/bod-26-04-prioritizing-security-updates-based-risk (read 19 Nov 2026) - NIST SP 800-115, Section 7: https://nvlpubs.nist.gov/nistpubs/Legacy/SP/nistspecialpublication800-115.pdf (read 19 Nov 2026) - NIST SP 800-82 Rev. 3, Task I-1: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-82r3.pdf (read 19 Nov 2026) - Fortinet PSIRT FG-IR-26-175 (FortiMail): https://fortiguard.fortinet.com/psirt/FG-IR-26-175 (read 19 Nov 2026) - Palo Alto Networks advisory CVE-2026-0322: https://security.paloaltonetworks.com/CVE-2026-0322 (read 19 Nov 2026) - Node.js October 2026 security releases: https://nodejs.org/en/blog/vulnerability/october-2026-security-releases (read 19 Nov 2026) - ELLIO analysis of CVE-2026-111497: https://ellio.tech/en/blog/node2shell-cve-2026-111497-rce-on-stock-nodejs-builds/ (read 19 Nov 2026) - SonicWall PSIRT SNWLID-2026-0014: https://psirt.global.sonicwall.com/vuln-detail/SNWLID-2026-0014 (read 19 Nov 2026) - Atlassian advisory CVE-2026-21589: https://confluence.atlassian.com/security/cve-2026-21589-arbitrary-file-access-vulnerability-impacts-multiple-products-1870495748.html (read 19 Nov 2026) - CISA KEV entry, CVE-2026-104286: https://www.cisa.gov/known-exploited-vulnerabilities-catalog?field_cve=CVE-2026-104286 (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Residual Risk After Mitigation: Define, Estimate, Record](https://remediationvsmitigation.com/guides/residual-risk-after-mitigation/index.md) - [Exception Register: 15 Columns, States and Blank CSV](https://remediationvsmitigation.com/exception-register/index.md) - [How Virtual Patching Works: Six Phases and a Real Rule](https://remediationvsmitigation.com/guides/how-virtual-patching-works/index.md) --- # When virtual patching is enough Canonical: https://remediationvsmitigation.com/guides/virtual-patching-and-when-it-is-enough/ A virtual patch is enough only while a fix is out of reach, every route to the flaw passes the filter, and the deadline you answer to accepts mitigation. FedRAMP asks for KEV remediation by the due date even when fully mitigated; PCI SSC separates "resolved" from "addressed" (read 19 Nov 2026). ## Seven questions that decide it OWASP defines virtual patching as "A security policy enforcement layer that detects and blocks matching exploitation attempts against a known vulnerability without changing the vulnerable code." It adds that it "reduces exposure only for traffic the enforcement layer inspects and attack variants its rules recognize" ([OWASP](https://cheatsheetseries.owasp.org/cheatsheets/Virtual_Patching_Cheat_Sheet.html)). Whether that is enough is a question about the flaw and the deadline, not about the rule. How a rule is built and rolled out is on [how virtual patching works](https://remediationvsmitigation.com/guides/how-virtual-patching-works/). Conditions for relying on a virtual patch. Sources named in the cells and listed below, read 19 Nov 2026. | Question | Enough when | Not enough when | |---|---|---| | 1. Is a fix available now? | No vendor fix exists, or the fix needs a maintenance window that is dated. OWASP: code fixes and virtual patching "are NOT mutually exclusive", so run the rule while the fix is scheduled. | The fix is installable today and the rule is being used to avoid the work. | | 2. Do all routes to the flaw pass the filter? | The vulnerable component is reached only through the inspected path. | Internal callers, a second hostname, an API gateway or a direct port reach the component without passing the rule. | | 3. Can a rule describe the flaw? | The flaw is an injection or path pattern. OWASP says the approach is "ideally suited for injection-type flaws". | The flaw is a logic or authentication error with no request pattern to match, so the rule would block legitimate traffic or nothing. | | 4. Has the rule been tested in block mode? | A dated test shows the rule blocks the plain and the encoded forms. OWASP: "A passing test for one payload does not establish complete protection." | The rule is in log-only mode, or the test used one sample. | | 5. Does the deadline accept mitigation? | The rule governs an internal target date or an exposure value (FedRAMP: once deflected, the flaw "should no longer be considered an internet-reachable vulnerability"). | One of the three deadlines in the next section applies. | | 6. Was the asset exposed before the rule? | The flaw became known and exploited after the rule was in place, or triage shows no compromise. | The flaw was exploited before the rule. The rule stops new attempts and removes nothing already planted. | | 7. Is there an end date? | The rule has a removal date set to the fix date. NIST SP 800-40 Rev. 4, Section 3.5.3: "Schedules should be set and enforced for both patch deployment and mitigation removal." | The rule is open-ended. Move the finding to a [dated exception](https://remediationvsmitigation.com/risk-acceptance-form/) instead. | ## Where a deadline rules it out - **FedRAMP, KEV due dates.** "Providers SHOULD remediate Known Exploited Vulnerabilities according to the due dates in the CISA Known Exploited Vulnerabilities Catalog (even if the vulnerability has been fully mitigated)" (VDR-TFR-KEV). The rules were optional from 4 Jul 2026 and are required for Rev5 initial and ongoing certification from 7 Dec 2026, with a grace period to 7 Mar 2027, so in Oct 2026 the KEV clause bound only providers that had adopted early ([FedRAMP](https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-detection-and-response/)). - **Federal civilian agencies, BOD 26-04.** CISA's implementation guidance says: "Aside from removing affected assets from the network, the only known technical mitigation to vulnerabilities in the KEV Catalog is to apply the required actions listed in the catalog." ([CISA](https://www.cisa.gov/news-events/directives/bod-26-04-implementation-guidance-prioritizing-security-updates-based-risk)). - **PCI DSS, critical and high-risk flaws.** PCI SSC FAQ 1597 says Requirements 11.3.1 and 11.3.1.1 together specify that critical and high-risk vulnerabilities are "resolved" and lower-risk ones "addressed"; "resolved" is defined as "the entity solves or fixes the vulnerability". Lower-risk flaws are "addressed", meaning the entity decides whether to resolve them or to mitigate "with a compensating control or by disabling a vulnerable service". A WAF rule can address a lower-risk finding; for a critical one it does not by itself meet "resolved", and whether an assessor accepts it as a compensating control for the requirement is the assessor's call ([audit guide](https://remediationvsmitigation.com/guides/compensating-controls-in-audit/); [PCI SSC](https://www.pcisecuritystandards.org/faqs/1597)). ## Worked example: FortiMail CVE-2026-104286, 1 Oct to 19 Nov 2026 Fortinet published advisory FG-IR-26-175 on Thu 1 Oct 2026 for CVE-2026-104286, a path traversal in FortiMail that lets an unauthenticated attacker write files through crafted HTTP or HTTPS requests, reported as exploited in the wild (CVSS 3 score 9.8). The advisory's fixed builds are 8.0.2, 7.6.7 and 7.4.9; on 1 Oct it called all three "upcoming", and the 5 Oct solution update first listed them as available. For the 7.2 line it says "Upgrade to upcoming 7.2.10 or above", and for 7.0 "Migrate to a fixed release". An appliance on 7.2 can therefore move to the 7.4 or 7.6 branch: the 7.4.9 release notes give the upgrade path 6.0.5 > 6.2.4 > 6.4.5 > 7.0.6 > 7.2.2 > 7.4.9 and list the same supported appliance models as the 7.2.9 notes (200F, 400F, 900F, 2000E, 2000F, 3000E, 3200E, 3000F). Fortinet's options for an unfixed appliance are to turn the IBE service off, to block internet access to the webmail interface or limit it to a trusted private network, or, if a WAF sits in front, to block POST requests to /ibe that contain '../'. That third option is the virtual patch. 1. **Thu 1 Oct:** Fortinet publishes FG-IR-26-175; CISA adds CVE-2026-104286 to KEV with a due date of Sun 4 Oct, 3 days later. 2. **Mon 5 Oct:** Fortinet revises the advisory (revision 2, solution update), which first lists 8.0.2, 7.6.7 and 7.4.9 as fixed builds, and remediates FortiMail Cloud by updating all firmware to 7.6.7 or 8.0.2, so cloud customers take no action. 3. **Wed 7 Oct:** revision 3 clarifies the FortiMail Cloud fix. This is the last revision. 4. **Thu 19 Nov:** the advisory, still at revision 3, reads "Upgrade to upcoming 7.2.10 or above" and gives no release date. Fortinet's document library lists FortiMail 7.2 documents up to 7.2.9 and none for 7.2.10, so no fixed 7.2 build is published 49 days after the KEV add date. The fixed 7.4.9 and 7.6.7 builds are. Take an appliance on 7.2.9 with a WAF in front. The example's own dates: the WAF rule goes live on Fri 2 Oct 2026 after a block-mode test that morning. Working the seven questions: Seven-question test applied to FortiMail 7.2.9. Facts from Fortinet FG-IR-26-175 and its CVRF revision history, the FortiMail 7.4.9 and 7.2.9 release notes, and CISA KEV (catalog 2026.11.18), read 19 Nov 2026. | Question | Answer for the 7.2.9 appliance | |---|---| | 1. Fix available? | Yes, in another branch. 7.2.10 is still only announced in the advisory on 19 Nov 2026, with no date, but 7.4.9 and 7.6.7 are listed as fixed and the 7.4.9 release notes give an upgrade path from 7.2.2 to 7.4.9 on the same models. The cost is a branch upgrade: a configuration backup, a reboot and a mail-flow interruption, plus the release-note caveats of the new branch. The rule is a bridge for the time that change needs, not for a missing fix. | | 2. All routes pass the filter? | Only if no one reaches the appliance's web interface except through the WAF. Check the appliance's allowed access settings, not only the WAF policy. | | 3. Can a rule describe it? | Yes, it is a path traversal on one path, which suits a pattern rule. Fortinet gives the rule's shape. | | 4. Tested in block mode? | Required before it counts. Record the test date. | | 5. Deadline accepts mitigation? | Not under a KEV clause. The 4 Oct due date came 3 days after the add date. FedRAMP's VDR-TFR-KEV is a SHOULD, bound on 1 Oct only providers that had adopted early, and it counts remediation, not mitigation, so a provider on 7.2.9 could not follow it by the 4 Oct date: 7.4.9 and 7.6.7 were first listed as fixed on Mon 5 Oct, a day later. The provider reports the open finding and its reductions, and mitigation does not satisfy the clause; the steps are on [FedRAMP deviation requests](https://remediationvsmitigation.com/guides/fedramp-deviation-requests/). An organization with no such clause can treat the rule as enough for the window to its scheduled upgrade. | | 6. Exposed before the rule? | Possibly. The advisory already reported exploitation on Thu 1 Oct, so the 1 to 2 Oct window before the rule went live is searched in the logs, and the configuration is backed up before changes, as CISA's forensic triage note asks. Fortinet's indicators are listed below the table. | | 7. End date? | Set to the date of the scheduled 7.4.9 or 7.6.7 upgrade, or to 7.2.10 if it ships first. The 7.0 appliances have no fix inside their branch, so their end date is the migration date, and the same release notes list a path from 7.0.6 through 7.2.2 to 7.4.9. Until then the rule is the control. That is an exception, not a patch state. | Fortinet's published indicators of compromise for FG-IR-26-175, as the advisory prints them (Fortinet writes addresses with a bracketed dot): - **IP addresses:** 79[.]141.169.187 and 45[.]129.0.192. - **System event log:** a cron entry beginning `type=event subtype=system pri=debug user=system ui=cron msg="(root) CMD (/bin/sh -c 'O=/migadmin ...` (Fortinet's text is cut off there). - **Admin session log:** `type=kevent subtype=admin pri=information user=admin ui=(null) action=logout status=success reason=unknown msg="User admin logged out from (null)."` - **Config log:** `type=kevent subtype=config ... msg="Added 'archive234' to 'archive account' : rotation-size[50]rotation-time[1] rotation-hour[14]destination[remote]remote-ip[79.141.169.187]remote-username[archive234]remote-password[***]remote-directory[/uploads] (user: admin, from: cli)"` - **Encryption log:** `Caught BufferException(2), BufferImpl.cpp:973, 'Invalid Base64 Encoding at pos 0. Character=0x2a'` and `Internal user *@domain.tld failed to log in.` The simplest of Fortinet's three options, switching IBE off, removes the path with no rule at all and costs the IBE secure mail feature. The WAF rule keeps the feature running and carries the testing burden in question 4. Which to choose is a cost question; the evidence needed for either is on [verifying a mitigation](https://remediationvsmitigation.com/guides/verifying-a-mitigation-works/). Products that ship WAF and IPS rules are on tools. ## What to write in the record when you rely on it Rule ID and version, block or log-only mode, the routes covered, the dated test, the approver and the removal date. The control entry for a [WAF rule](https://remediationvsmitigation.com/compensating-controls/waf-rule/) lists the same evidence, and the row for [a control already in front](https://remediationvsmitigation.com/decision-table/compensating-control-in-front/) shows when the finding stays open. NIST SP 800-40 Rev. 4 says organizations "should plan to replace emergency mitigations with permanent fixes". ## Sources - OWASP Virtual Patching Cheat Sheet: https://cheatsheetseries.owasp.org/cheatsheets/Virtual_Patching_Cheat_Sheet.html (read 19 Nov 2026) - FedRAMP 2026, VDR-TFR-KEV: https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-detection-and-response/ (read 19 Nov 2026) - FedRAMP 2026, VER rules, internet-reachable note: https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-evaluation-and-reporting/ (read 19 Nov 2026) - CISA, BOD 26-04 implementation guidance, FAQ: https://www.cisa.gov/news-events/directives/bod-26-04-implementation-guidance-prioritizing-security-updates-based-risk (read 19 Nov 2026) - PCI SSC FAQ 1597, resolved and addressed: https://www.pcisecuritystandards.org/faqs/1597 (read 19 Nov 2026) - NIST SP 800-40 Rev. 4, Section 3.5.3: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf (read 19 Nov 2026) - Fortinet PSIRT FG-IR-26-175: https://fortiguard.fortinet.com/psirt/FG-IR-26-175 (read 19 Nov 2026) - Fortinet PSIRT FG-IR-26-175, CVRF revision history: https://fortiguard.fortinet.com/psirt/cvrf/FG-IR-26-175 (read 19 Nov 2026) - CISA KEV entry, CVE-2026-104286: https://www.cisa.gov/known-exploited-vulnerabilities-catalog?field_cve=CVE-2026-104286 (read 19 Nov 2026) - Fortinet Document Library, FortiMail 7.2 release notes: https://docs.fortinet.com/product/fortimail/7.2 (read 19 Nov 2026) - CISA KEV feed, catalog version 2026.11.18: https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) - Fortinet, FortiMail 7.4.9 release notes, Firmware Upgrade and Downgrade (upgrade path): https://docs.fortinet.com/document/fortimail/7.4.9/release-notes/569892/firmware-upgrade-and-downgrade (read 19 Nov 2026) - Fortinet, FortiMail 7.2.9 release notes, supported models: https://docs.fortinet.com/document/fortimail/7.2.9/release-notes (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [How Virtual Patching Works: Six Phases and a Real Rule](https://remediationvsmitigation.com/guides/how-virtual-patching-works/index.md) - [WAF Rule as a Virtual Patch: Atlassian Case and Limits](https://remediationvsmitigation.com/compensating-controls/waf-rule/index.md) - [A Compensating Control Is in Front: Why a Fix Date Stays](https://remediationvsmitigation.com/decision-table/compensating-control-in-front/index.md) - [How to Verify a Mitigation Works: Checks for 5 Real CVEs](https://remediationvsmitigation.com/guides/verifying-a-mitigation-works/index.md) - [FedRAMP Deviation Requests and Their 2026 Replacements](https://remediationvsmitigation.com/guides/fedramp-deviation-requests/index.md) --- # Workaround vs mitigation Canonical: https://remediationvsmitigation.com/guides/workaround-vs-mitigation/ Advisories do not use the two words alike. Fortinet and LiteLLM call a feature switch a workaround, Red Hat files the same kind of step under mitigation, and Cisco prints "no workarounds" beside a listed mitigation. On 19 Nov 2026 CISA's KEV feed used mitigations in 795 of 1,786 required actions and workaround in one. ## The definitions vendors rarely quote OASIS CSAF 2.0, the format many vendors publish advisories in, defines a workaround as remediation information about "a configuration or specific deployment scenario that can be used to avoid exposure to the vulnerability". It defines a mitigation as information that "helps to reduce the risk of the vulnerability but that does not resolve the vulnerability on the affected product" ([CSAF 2.0, remediation category](https://docs.oasis-open.org/csaf/csaf/v2.0/os/csaf-v2.0-os.html)). The general meaning is in [the difference between remediation and mitigation](https://remediationvsmitigation.com/guides/difference-between-remediation-and-mitigation/). Microsoft once defined the words the other way round. Bulletin MS10-046 (2010) says a mitigation is a setting or common configuration "existing in a default state, that could reduce the severity of exploitation of a vulnerability", and a workaround "does not correct the underlying vulnerability but would help block known attack vectors before you apply the update" ([MS10-046](https://learn.microsoft.com/en-us/security-updates/securitybulletins/2010/ms10-046)). In Microsoft's August 2026 CVRF file the entry for CVE-2026-56179 (Windows Network Address Translation spoofing) says: "The mitigation is disabled by default. To enable it, apply the following registry keys and reboot the machine." A 2026 Microsoft mitigation is something the administrator switches on, not a default that already helps ([MSRC CVRF 2026-Aug](https://api.msrc.microsoft.com/cvrf/v3.0/cvrf/2026-Aug)). ## How twelve 2026 advisories word it Eleven of the twelve rows are CVEs in this site's data, read again on 19 Nov 2026 at the vendor's own page or data feed; the Microsoft row is from the MSRC file. Vendors are listed alphabetically; the right column describes the step, in this site's words. Quoted from the vendor pages and data feeds linked under Sources; read 19 Nov 2026. | Advisory | Wording, quoted | What the step does | |---|---|---| | BerriAI LiteLLM, GHSA-7488-6r32-c95q (CVE-2026-59822) | "Workarounds If upgrading is not immediately possible, disable MCP routes or block access to /mcp/ and related MCP endpoints at your reverse proxy or API gateway." | Closes the vulnerable path by turning off a set of routes. | | Broadcom, VMSA-2026-0006.2 (includes CVE-2026-41709) | "Workarounds: None" | Offers nothing besides the patch. | | Cisco, cisco-sa-asaftd-tunnel-dos-s6mgkb75 (CVE-2026-76418) | "There are no workarounds that address this vulnerability. However, there is a mitigation, which is to disable secondary authentication on any user-configured tunnel group list or on the default tunnel group, DefaultWEBVPNGroup." | A configuration change that Cisco files under mitigation while stating that no workaround exists. | | Citrix, CTX696939 (CVE-2026-19490) | "Workarounds/ Mitigating Factors None" | One heading covers both words; no step is offered. | | F5, K000162605 (CVE-2026-94127), CVE record field "workarounds" | "An iRule is available upon request. Open a ticket with F5 support to request this." | A traffic rule that sits in front of the flaw, supplied on request. | | Fortinet, FG-IR-26-175 (CVE-2026-104286) | "customers are urged to apply the workaround below"; "Workaround: Disable the IBE feature support via the GUI" | Turns off the feature that contains the flaw. | | Microsoft, CVRF 2026-Aug (CVE-2026-56179) | "The mitigation is disabled by default. To enable it, apply the following registry keys and reboot the machine." | A hardening setting that stays off until an administrator sets it. | | Palo Alto Networks (CVE-2026-0322) | "Workarounds and Mitigations No known workarounds exist for this issue. Customers with a Threat Prevention subscription are provided with limited coverage against this vulnerability by enabling Threat ID 510046 and 510047" | One heading for both words; a detection signature is described as limited coverage, not as a workaround. | | PHP Group, GHSA-w8hw-w9fx-j79w (CVE-2026-109618) | "Updating to a PHP release containing the patch to this vulnerability is the only complete mitigation." "...setting and enforcing memory_limit=-1 administratively is the primary configuration mitigation." "This is not a complete mitigation." | Counts the patch as a mitigation and the configuration change as a partial one. | | Red Hat, CVE-2026-53266, field "mitigation" | "Disable ARP hardware address rewriting in ebtables SNAT rules, or remove ebtables SNAT rules that operate on ARP traffic on bridge interfaces." | Turns off the kernel feature path the flaw sits in. | | Red Hat, CVE-2026-80521, field "mitigation" | "Mitigation for this issue is either not available or the currently available options do not meet the Red Hat Product Security criteria comprising ease of use and deployment, applicability to widespread installation base, or stability." | The field holds a statement that nothing meets Red Hat's bar. | | SonicWall, SNWLID-2026-0014 (CVE-2026-19868), field "workaround" | "If SAML authentication is in use and upgrading the firmware is not possible, disable or delete any active LDAP, RADIUS, and TACACS+ authentication server configured on the firewall." | Removes the alternate sign-in paths the flaw uses. | ## The label does not say what the step does Fortinet, SonicWall and LiteLLM each call a step that turns off a feature or a path a workaround. Red Hat calls a step of the same kind a mitigation, and Cisco calls its step a mitigation in the same sentence that denies any workaround exists. PHP goes further and names the patch itself a mitigation. A reader who sorts findings by the vendor's label will sort the same kind of step into two piles. Sort by effect instead. A step that removes the vulnerable path (a disabled feature, a blocked route) changes whether the flaw can be reached. A step that narrows who can reach it (an access list, a jump host) changes exposure. A signature or alert changes detection or likelihood. Each can be a compensating control; none is the fix. The wording for a configuration change is on [configuration change](https://remediationvsmitigation.com/compensating-controls/configuration-change/). When the vendor lists nothing, say so in the record and mark any step you add as your own. Broadcom's "Workarounds: None" and Citrix's "None" leave the patch as the only vendor instruction, so access limits placed in front of those products are your decision and carry your name. Mitigation when no patch exists covers the case where even the patch is absent. ## CISA's KEV entries say mitigations In the KEV feed (catalogVersion 2026.11.18, 1,786 entries) the required action of 795 entries contains "mitigat" and one contains "workaround". The one is CVE-2026-0300, a Palo Alto Networks entry that repeats the vendor's text: "Until the vendor releases an official fix, the following workaround should be implemented". Entries with a vendor workaround still use CISA's word. CVE-2026-94127 (F5, due 25 Sep 2026) and CVE-2026-104286 (Fortinet, due 4 Oct 2026) both read: "Apply mitigations in accordance with vendor instructions" and end "discontinue use of the product if mitigations are unavailable". For a KEV item, the vendor's workaround is the mitigation CISA means. What follows if none exists is on [KEV required actions: mitigate or discontinue](https://remediationvsmitigation.com/guides/kev-required-actions-mitigate-or-discontinue/). ## What to write in the exception record Quote the vendor's label and then say what the step does in your own terms, so a later reader can compare vendors without learning each one's vocabulary. Vendor label and wording: "" (advisory ID, revision date) Effect, in our words: removes the vulnerable path / narrows who can reach it / reduces impact / detects only Function lost while in place: Set by: Verified by: Ends when: Status of the finding: open until the vendor fix is installed The "set by" line matters most when the vendor lists no workaround. The rest of the record follows the [Risk Acceptance Form](https://remediationvsmitigation.com/risk-acceptance-form/). ## Sources - OASIS, CSAF Version 2.0, remediation category values: https://docs.oasis-open.org/csaf/csaf/v2.0/os/csaf-v2.0-os.html (read 19 Nov 2026) - Microsoft, security bulletin MS10-046 (2010), mitigation and workaround definitions: https://learn.microsoft.com/en-us/security-updates/securitybulletins/2010/ms10-046 (read 19 Nov 2026) - Microsoft Security Response Center, CVRF document 2026-Aug, CVE-2026-56179: https://api.msrc.microsoft.com/cvrf/v3.0/cvrf/2026-Aug (read 19 Nov 2026) - BerriAI, LiteLLM advisory GHSA-7488-6r32-c95q: https://github.com/BerriAI/litellm/security/advisories/GHSA-7488-6r32-c95q (read 19 Nov 2026) - Broadcom, VMSA-2026-0006.2: https://support.broadcom.com/web/ecx/support-content-notification/-/external/content/SecurityAdvisories/0/38017 (read 19 Nov 2026) - Cisco, advisory cisco-sa-asaftd-tunnel-dos-s6mgkb75: https://sec.cloudapps.cisco.com/security/center/content/CiscoSecurityAdvisory/cisco-sa-asaftd-tunnel-dos-s6mgkb75 (read 19 Nov 2026) - Citrix, security bulletin CTX696939: https://support.citrix.com/external/article/CTX696939/netscaler-adc-and-netscaler-gateway-secu.html (read 19 Nov 2026) - CVE Services, F5 record for CVE-2026-94127 (workarounds field): https://cveawg.mitre.org/api/cve/CVE-2026-94127 (read 19 Nov 2026) - Fortinet PSIRT, FG-IR-26-175: https://fortiguard.fortinet.com/psirt/FG-IR-26-175 (read 19 Nov 2026) - Palo Alto Networks, CVE-2026-0322 advisory: https://security.paloaltonetworks.com/CVE-2026-0322 (read 19 Nov 2026) - PHP Group, php-src advisory GHSA-w8hw-w9fx-j79w: https://github.com/php/php-src/security/advisories/GHSA-w8hw-w9fx-j79w (read 19 Nov 2026) - Red Hat Security Data API, CVE-2026-53266: https://access.redhat.com/hydra/rest/securitydata/cve/CVE-2026-53266.json (read 19 Nov 2026) - Red Hat Security Data API, CVE-2026-80521: https://access.redhat.com/hydra/rest/securitydata/cve/CVE-2026-80521.json (read 19 Nov 2026) - SonicWall PSIRT data feed, SNWLID-2026-0014: https://psirtapi.global.sonicwall.com/api/v1/vulndetail/?advisory_id=SNWLID-2026-0014 (read 19 Nov 2026) - CISA, KEV catalog JSON feed, catalogVersion 2026.11.18, requiredAction field: https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Mitigation vs Remediation Wording in Vendor Advisories](https://remediationvsmitigation.com/guides/mitigation-vs-remediation-in-vendor-advisories/index.md) - [Difference Between Remediation and Mitigation in IT Security](https://remediationvsmitigation.com/guides/difference-between-remediation-and-mitigation/index.md) - [Configuration Change as a Mitigation: Cisco ASA Case](https://remediationvsmitigation.com/compensating-controls/configuration-change/index.md) - [KEV Required Action: Mitigate or Discontinue Use](https://remediationvsmitigation.com/guides/kev-required-actions-mitigate-or-discontinue/index.md) --- # Privacy Canonical: https://remediationvsmitigation.com/privacy/ The site sets no cookies, runs no analytics and loads no third-party scripts. The only record of a visit is a CloudFront access log line in a private bucket, deleted after 400 days. The decision tool and the form send nothing. First edition 19 Nov 2026. ## What is stored when you visit Pages are static files on Amazon S3 served by Amazon CloudFront. Each request produces one log line in CloudFront standard logging. AWS documents 33 fields per line in its [standard logging reference](https://docs.aws.amazon.com/AmazonCloudFront/latest/DeveloperGuide/standard-logs-reference.html), read 19 Nov 2026. Those fields include the date and time in UTC, the edge location, bytes sent, the viewer's IP address, the HTTP method, the host, the path, the query string, the status code, the Referer header, the User-Agent header and the cookie header. The site sets no cookies, so the cookie field holds no cookie from this site. The log has no field for a name or an email address. The IP address, user agent and referring page are the fields that can point toward a person or a device. What the site keeps and for how long. Read 19 Nov 2026. | Item | Where it is kept | How long | |---|---|---| | CloudFront access log lines | A private S3 bucket | Deleted after 400 days | | Cookies from this site | None are set | Not applicable | | Analytics or advertising data | None is collected | Not applicable | | Text typed into the decision tool or the form | Only in the browser tab; nothing is sent | Gone when the tab closes | | Email you send to editor@remediationvsmitigation.com | The mailbox | Until the thread is deleted | ## What the pages load Scripts and styles come from the same site. No page calls a third-party script, font service, tag manager or embedded widget, so no other company receives your address or your visit as a result of reading a page. Links to sources, such as [CISA](https://www.cisa.gov/) or [NIST](https://csrc.nist.gov/), leave the site only when you follow them, and those sites have their own policies. ## Using the tool and the form The [decision tool](https://remediationvsmitigation.com/) and the [Risk Acceptance Form](https://remediationvsmitigation.com/risk-acceptance-form/) run as code in your browser. Answers, names and dates you enter are not transmitted, and the page script makes no network request of its own. The Word and PDF versions of the form are ready-made files stored on the site, not documents built from what you type. Downloading one is an ordinary request and produces one log line like any page. Once saved, a filled-in copy is yours to keep under your own controls, because an exception record can name systems and approvers. ## Questions about log data To ask about a log line, mail [editor@remediationvsmitigation.com](mailto:editor@remediationvsmitigation.com) with the date and time range of your visit. Because logs hold an IP address and not a name, a match can only be made from details you supply. The [terms](https://remediationvsmitigation.com/terms/) cover reuse of the content, and the [about page](https://remediationvsmitigation.com/about/) describes the hosting. ## Sources - AWS, CloudFront Developer Guide, Standard logging reference (33 log file fields): https://docs.aws.amazon.com/AmazonCloudFront/latest/DeveloperGuide/standard-logs-reference.html (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [About Remediation vs Mitigation: Scope and Hosting](https://remediationvsmitigation.com/about/index.md) - [Terms of Use and CC BY 4.0 License](https://remediationvsmitigation.com/terms/index.md) - [Contact: editor@remediationvsmitigation.com](https://remediationvsmitigation.com/contact/index.md) - [Risk Acceptance Form: Free Template With 15 Fields](https://remediationvsmitigation.com/risk-acceptance-form/index.md) --- # Risk acceptance form Canonical: https://remediationvsmitigation.com/risk-acceptance-form/ A risk acceptance form is a signed record that a named approver accepts a stated residual risk until a stated expiry date. This one has 15 fields: 13 for every finding, plus a vendor field and a PCI field when they apply. Each field traces to NIST, FedRAMP, CISA or PCI SSC text read on 19 Nov 2026. ## The form in full Copy the text below or download it as [Word](https://remediationvsmitigation.com/downloads/risk-acceptance-form.docx) or [PDF](https://remediationvsmitigation.com/downloads/risk-acceptance-form.pdf). Fill it in, have the approver sign and date it, and write an expiry date before anyone files it. A filled-in copy is on the [worked example](https://remediationvsmitigation.com/risk-acceptance-form/example/). # Risk Acceptance Form Form version 1.0, first published 19 Nov 2026. License: CC BY 4.0. Cite as: Remediation vs Mitigation, Risk Acceptance Form, 19 Nov 2026, remediationvsmitigation.com/risk-acceptance-form/. Write no exploit steps, payloads or probing instructions in this record. ## A. The finding Field 1. Exception ID (your own number): ________ CVE ID(s) and scanner finding ID(s) covered: ________ Field 2. First detected (date, time, tool or report): ________ Evaluation completed (date, time): ________ ## B. Where it sits and how likely it is to be used Field 3. Asset(s) or system ID(s): ________ Environment: [ ] prod [ ] dev Exposure: [ ] public [ ] internal Asset type: [ ] server [ ] application [ ] network device Field 4. Internet-reachable? [ ] yes [ ] no Basis (test or architecture drawing): ________ Field 5. Likely exploitable? [ ] yes [ ] no Reason: ________ On the CISA KEV catalog? [ ] yes [ ] no Catalog version checked: ________ Field 6. Estimated impact if exploited, on your own scale: ________ ## C. The decision Field 7. Why this is accepted instead of fixed or mitigated: ________ Field 8. Mitigating factors and compensating controls in place (control, owner, how verified, date verified, entry link): ________ Field 9. Plan: corrective tasks, owner of each task, resources, milestones, scheduled completion dates: ________ Field 10. Reopen this record when: ________ Review every ____ days. Next review: ________ Field 11. Residual risk after the controls (one sentence) and the risk tolerance it is compared with: ________ Field 12. Only if the fix depends on a vendor: product, vendor status evidence, last vendor check-in date, next check-in date: ________ ## D. Decision and expiry Field 13. Approver name, title, signature, date: ________ Statement: "I accept the residual risk in Field 11 until the date in Field 14." Field 14. Expiry date. The exception lapses on this date unless the approver renews it in writing: ________ ## E. Only if PCI DSS applies Field 15. Compensating control worksheet: attach the worksheet your QSA uses, and name the PCI DSS requirement it stands in for: ________ ## What goes in each field and why it is there A field is on the form only where a source read on 19 Nov 2026 asks for it. Fields a team might add for its own reasons are left off. Fields 12 and 15 are conditional; the other 13 are asked of every finding. The 15 fields of the form, what goes in each, and the quoted wording behind it; sources are listed at the foot of the page; read 19 Nov 2026. | No. | Field | What goes in it | The wording that asks for it | |---|---|---|---| | 1 | Exception ID, CVE IDs, finding IDs | One internal number for the exception, then every CVE and scanner finding it covers. The CVE list is what lets a later KEV listing find this record. | FedRAMP VER-RPT-AVI: "Provider's internally assigned tracking identifier". FedRAMP's 27 Jun 2022 POA&M template update added column AD, "CVE", for the CVEs on each line item. | | 2 | Detection and evaluation | Date, time and tool or report of the first detection, then date and time the evaluation finished. Two timestamps, because a clock such as FedRAMP's 192 days runs from evaluation. | VER-RPT-AVI: "Time and source of the detection" and "Time of completed evaluation". | | 3 | Asset, environment, exposure | Hostnames or system IDs, prod or dev, public or internal, and the asset type (server, application or network device). | CISA BOD 26-04 asset tags: "Environment (prod/dev)", "Exposure (public/internal)", "Asset type (server, application, or network device)". | | 4 | Internet-reachable | Yes or no, and the basis: a test or an architecture drawing. Reachable means a payload from the internet can arrive at the vulnerable part, including through a load balancer, so internal-only is not enough of an answer. | VER-RPT-AVI: "Is it an internet-reachable vulnerability or not?" VER-EVA-EIR notes that FedRAMP "focuses on internet-reachable (rather than internet-accessible)" and uses SQL injection behind a load balancer as its example. | | 5 | Likely exploitable; KEV status | Yes or no with the reason, KEV yes or no, and the catalog version you checked. If you cannot show that exploitation cannot be automated, write that automation is assumed. | VER-RPT-AVI: "Is it a likely exploitable vulnerability or not?" VER-EVA-AIA: "Providers MUST assume the exploitation of vulnerabilities can be automated UNLESS they have evidence proving otherwise." BOD 26-04 lists "KEV Status" as a variable. | | 6 | Estimated impact if exploited | A rating on your own scale, with the reason in a clause. FedRAMP's scale has six levels, N0 ("extremely unlikely to have any adverse effects") to N5 ("a debilitating customer effect on more than one agency"). | VER-RPT-AVI: "Currently estimated Potential Agency Impact N-rating"; the levels are defined in VER-EVA-EPA. | | 7 | Why accepted | The constraint in plain words: no vendor fix yet, a patch that breaks a system, a retirement date. Add anything that helps the approver judge the risk. | VER-RPT-AVI: "Explanation of why this is an accepted vulnerability", and supplementary information that helps assess or mitigate the risk. NIST SP 800-39: "Risk acceptance is the appropriate risk response when the identified risk is within the organizational risk tolerance." | | 8 | Mitigating factors and compensating controls | For each control: what it is, who owns it, how it was checked, the date checked, and a link to its control entry. Link rather than describe, so a change to the control updates one record. | NIST SP 800-37 Rev. 2, Task R-2: plans contain "an accurate description of implemented controls, including compensating controls". FedRAMP legacy POA&M playbook: "CSPs must describe mitigating factors or compensating controls in place that reduce likelihood and/or impact of exploitation." | | 9 | Plan | The tasks that end the exception, an owner for each, the resources needed, milestones and dates. At least one task must be the fix or the decision to replace. | SP 800-37 Rev. 2, Task A-6: the plan includes "the scheduled completion dates for the milestones and tasks". SP 800-53 Rev. 5, CA-5 a: "document the planned remediation actions of the organization". | | 10 | Reopen triggers and review cadence | The events that reopen the record (scan result, audit, monitoring finding, change in exposure, new KEV listing, vendor fix released), a number of days between reviews, and the next review date. | CA-5 b: update the plan "based on the findings from control assessments, independent audits or reviews, and continuous monitoring activities". SP 800-40 Rev. 4, Section 3.5.5: "All exceptions to maintenance plans should be reviewed regularly". | | 11 | Residual risk against tolerance | One sentence on what risk remains once the controls hold, and the stated tolerance it is compared with. See [residual risk after mitigation](https://remediationvsmitigation.com/guides/residual-risk-after-mitigation/). | SP 800-37 Rev. 2, Task R-3: "regardless of the risk response, there remains a degree of residual risk". SP 800-53 Rev. 5, RA-7: respond to findings "in accordance with organizational risk tolerance". | | 12 | Vendor dependency | Only when the fix waits on a vendor: product, evidence of the vendor's status (advisory, vendor email), last check-in date and next check-in date. | FedRAMP legacy POA&M playbook: "CSPs are required to check in with the vendor at least once a month"; "Do provide evidence of vendor interactions regarding the status of patches/fixes (e.g., vendor notifications, email exchanges, etc.)." | | 13 | Approver | Name, title, signature, date and the acceptance statement. The person who asked for the exception does not sign it; see [who is the risk owner](https://remediationvsmitigation.com/guides/risk-owner/). | SP 800-37 Rev. 2, Task R-4: "The explicit acceptance of risk is the responsibility of the authorizing official and cannot be delegated to other officials within the organization." | | 14 | Expiry date | The date the exception lapses unless the approver renews it in writing. See [risk exception expiry](https://remediationvsmitigation.com/guides/exception-expiry-and-renewal/). | SP 800-37 Rev. 2: "The authorization termination date is established by the authorizing official and indicates when the authorization expires." | | 15 | PCI compensating control worksheet | Only when PCI DSS applies: attach the worksheet your Qualified Security Assessor uses and name the requirement it stands in for. | PCI SSC FAQ 1046: assessors "are responsible to determine whether a compensating control is sufficient to meet the intent of a requirement". FAQ 1597 defines "addressed" as deciding whether to resolve a vulnerability or mitigate the risk "in another way (e.g., with a compensating control or by disabling a vulnerable service)". | ## Three fields where the sources differ **Expiry (field 14).** NIST leaves the date to the authorizing official and allows no date at all: "Organizations may eliminate the authorization termination date if the system is operating under an ongoing authorization" (SP 800-37 Rev. 2, Task R-4). FedRAMP's 2026 rule sets no expiry; it sets a reporting trigger, 192 days from evaluation, after which an unfixed finding must be categorized as accepted (VER-TFR-MAV). The form asks for a date because an exception with no end is the case no source allows by default; what each source allows is on [maximum durations](https://remediationvsmitigation.com/risk-acceptance-form/maximum-durations/). **Review cadence (field 10).** CA-5 b leaves the update frequency to the organization, and SP 800-40 Rev. 4 says only "reviewed regularly". The form asks for a number of days and a date, which is stricter than either. **Vendor check-in (field 12).** The monthly interval comes from the FedRAMP legacy POA&M playbook. FedRAMP's two 2026 rule pages on vulnerabilities (evaluation and reporting, detection and response) set no vendor check-in interval. Keep the monthly check-in as a convention, not as a current FedRAMP requirement. ## What stays out of the record FedRAMP's disclosure rule draws the line a record should follow: providers "MUST NOT irresponsibly disclose specific sensitive information about vulnerabilities that would likely lead to exploitation, but MUST disclose sufficient information for informed risk-based decision-making" (VER-RPT-NID). Fields 4, 5 and 7 should say whether the flaw is reachable and likely to be used, and why, without payloads, request samples or step lists. ## What signing does not change Signing the form does not move a federal civilian agency's due date. CISA's BOD 26-04 implementation guidance answers the waiver question directly: "CISA does not issue waivers or exceptions for actions required in cyber directives." The [Decision Table](https://remediationvsmitigation.com/decision-table/) row for a missed KEV due date shows what to do instead. For PCI DSS, the form is not the compensating control worksheet. PCI SSC says it "is not able to approve specific configurations or compensating controls" (FAQ 1046); a Qualified Security Assessor judges them. See the PCI row of the [Decision Table](https://remediationvsmitigation.com/decision-table/). Related guide: [inherent versus residual risk](https://remediationvsmitigation.com/guides/inherent-vs-residual-risk/) for fields 4 to 6 against 11. ## Sources - FedRAMP Consolidated Rules for 2026, Vulnerability Evaluation and Reporting (VER-RPT-AVI, VER-RPT-NID, VER-EVA-EIR, VER-EVA-AIA, VER-EVA-EPA, VER-TFR-MAV): https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-evaluation-and-reporting/ (read 19 Nov 2026) - FedRAMP Consolidated Rules for 2026, Vulnerability Detection and Response: https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-detection-and-response/ (read 19 Nov 2026) - FedRAMP, Update to the POA&M template, 27 Jun 2022: https://www.fedramp.gov/archive/2022-06-28-update-poam-template/ (read 19 Nov 2026) - FedRAMP legacy documentation, Plans of Action and Milestones: https://www.fedramp.gov/legacy/playbook/csp/authorization/poam/ (read 19 Nov 2026) - NIST SP 800-37 Rev. 2, Tasks A-6, R-2, R-3, R-4: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-37r2.pdf (read 19 Nov 2026) - NIST SP 800-39, Task 3-1, risk acceptance: https://nvlpubs.nist.gov/nistpubs/Legacy/SP/nistspecialpublication800-39.pdf (read 19 Nov 2026) - NIST SP 800-40 Rev. 4, Section 3.5.5: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf (read 19 Nov 2026) - NIST SP 800-53 Rev. 5, CA-5 and RA-7 (OSCAL catalog): https://csrc.nist.gov/projects/cprt/catalog#/cprt/framework/version/SP_800_53_5_2_0/home/CA-5 (read 19 Nov 2026) - CISA, BOD 26-04, asset tags and variables: https://www.cisa.gov/news-events/directives/bod-26-04-prioritizing-security-updates-based-risk (read 19 Nov 2026) - CISA, BOD 26-04 implementation guidance, waiver question: https://www.cisa.gov/news-events/directives/bod-26-04-implementation-guidance-prioritizing-security-updates-based-risk (read 19 Nov 2026) - PCI SSC, FAQ 1046, approval of compensating controls: https://www.pcisecuritystandards.org/faqs/1046 (read 19 Nov 2026) - PCI SSC, FAQ 1597, resolving or addressing vulnerabilities: https://www.pcisecuritystandards.org/faqs/1597 (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Risk Acceptance Form Example: Filled-In Worked Sample](https://remediationvsmitigation.com/risk-acceptance-form/example/index.md) - [Who Can Accept Risk: Signer by Standard (2026)](https://remediationvsmitigation.com/risk-acceptance-form/who-may-sign/index.md) - [Maximum Duration of a Risk Acceptance: 9 Sources](https://remediationvsmitigation.com/risk-acceptance-form/maximum-durations/index.md) - [Exception Register: 15 Columns, States and Blank CSV](https://remediationvsmitigation.com/exception-register/index.md) - [Residual Risk After Mitigation: Define, Estimate, Record](https://remediationvsmitigation.com/guides/residual-risk-after-mitigation/index.md) - [Who Is the Risk Owner for a Security Finding?](https://remediationvsmitigation.com/guides/risk-owner/index.md) - [Risk Exception Expiry: What Happens and How to Renew](https://remediationvsmitigation.com/guides/exception-expiry-and-renewal/index.md) - [Inherent vs Residual Risk on a Security Decision Record](https://remediationvsmitigation.com/guides/inherent-vs-residual-risk/index.md) --- # Risk acceptance form example Canonical: https://remediationvsmitigation.com/risk-acceptance-form/example/ A worked sample fills all 15 fields of the Risk Acceptance Form for a real flaw, CVE-2026-80521 in the Linux kernel, on six invented appliances whose vendor has shipped no fixed kernel. The approver signs on 19 Nov 2026 and the exception expires 17 Feb 2027, 90 days later. ## The finding behind the sample CVE-2026-80521 is a race in the Linux kernel's AF_UNIX garbage collector that can corrupt kernel memory. It needs local code execution (CVSS 3.1 vector AV:L/AC:L/PR:L, base 7.8) and is not in the CISA KEV catalog, version 2026.11.18 ([NVD](https://services.nvd.nist.gov/rest/json/cves/2.0?cveId=CVE-2026-80521), [CISA KEV feed](https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json)). The fix went into Linux 7.2, and stable releases carry it from 7.1.10, 6.18.53, 6.12.111, 6.6.158 and 6.1.189. The last 7.0 release on kernel.org, 7.0.14, lacks it ([ChangeLog-7.0.14](https://cdn.kernel.org/pub/linux/kernel/v7.x/ChangeLog-7.0.14); [ChangeLog-7.1.10](https://cdn.kernel.org/pub/linux/kernel/v7.x/ChangeLog-7.1.10) has it), and kernel.org's release list on 19 Nov 2026 names no 7.0 branch among the maintained kernels ([kernel.org releases](https://www.kernel.org/releases.json)). Red Hat states that a mitigation "is either not available or the currently available options do not meet" its criteria ([Red Hat](https://access.redhat.com/hydra/rest/securitydata/cve/CVE-2026-80521.json)). That makes it the situation the [Decision Table](https://remediationvsmitigation.com/decision-table/) calls a fix that waits on a vendor: an appliance vendor must rebase its firmware onto a maintained kernel, and until it does the choice is between a compensating control and a dated acceptance. The sample takes both: two controls and a 90-day acceptance. Everything about the organization is invented, including the vendor's message. ## The completed form # Risk Acceptance Form: WORKED SAMPLE (invented organization, real CVE) Sample record. The organization, systems, people, IDs and vendor messages are invented. CVE-2026-80521 and the kernel facts about it are real. ## A. The finding Field 1. Exception ID: SAMPLE-EXC-0001 CVE ID(s) and scanner finding ID(s) covered: CVE-2026-80521; inventory finding SAMPLE-F-4417 Field 2. First detected: 14 Oct 2026, 09:20 UTC, by the weekly firmware inventory match (sample): appliance firmware 4.2 reports kernel 7.0.x Evaluation completed: 21 Oct 2026, 16:30 UTC ## B. Where it sits and how likely it is to be used Field 3. Asset(s): six gateway appliances, gw-01 to gw-06 (sample names). Environment: prod. Exposure: internal. Field 4. Internet-reachable? No. Basis: the flaw needs code already running on the appliance; the appliances run no service that accepts user-supplied code; architecture drawing dated 10 Nov 2026 shows no route from the internet to their management or shell access. Field 5. Likely exploitable? Not on current facts. The CVSS 3.1 vector is AV:L/AC:L/PR:L (base 7.8): the attacker must already run code as a local user and win a narrow race. On the CISA KEV catalog? No. Catalog version checked: 2026.11.18. Field 6. Estimated impact if exploited, on the sample organization's three-level scale: Moderate (root on a gateway exposes internal routing configuration; no customer data is stored on it). ## C. The decision Field 7. Why accepted: the appliance vendor controls the firmware and ships kernel 7.0.x. The kernel project's fix is in 7.2 and in the 7.1.10, 6.18.53, 6.12.111, 6.6.158 and 6.1.189 stable releases, and no 7.0.x release contains it, so the fix can only arrive as a firmware rebase by the vendor. The vendor's email of 12 Nov 2026 (sample) says firmware 4.3 with a fixed kernel is planned for the first quarter of 2027. Replacing six gateways before then would interrupt site-to-site traffic. Policy SAMPLE-POL-3 allows Moderate residual risk on an internal service for up to 90 days. Field 8. Controls in place: - Local shell and code-execution rights on the appliances limited to three named administrator accounts; all other local accounts disabled. Owner: Platform lead. Verified 18 Nov 2026 by listing accounts on all six units. Register entry: access-restriction, SAMPLE-CC-12. - Alert on any new local account or new shell session on an appliance. Owner: Security operations lead. Verified 18 Nov 2026 by a test alert on gw-03. Register entry: increased-monitoring, SAMPLE-CC-31. - Neither control is a vendor instruction: Red Hat states a mitigation is "either not available or the currently available options do not meet" its criteria, and no vendor lists a workaround. Field 9. Plan: - Task 1: install firmware 4.3 on all six units when released. Owner: Platform lead. Needs: a 2-hour change window per unit. Milestone: within 14 days of release. - Task 2: if the vendor has not shipped by 15 Feb 2027, decide between replacing the appliances and a renewed acceptance. Owner: Application owner. Date: 15 Feb 2027. Field 10. Reopen when: the vendor ships a kernel containing the fix; CVE-2026-80521 appears on the KEV catalog; any appliance gains a route from outside the internal network or a feature that runs user-supplied code; either control fails its check. Review every 30 days. Next review: 18 Dec 2026. Field 11. Residual risk: Low to Moderate while both controls hold; inside the Moderate tolerance stated in policy SAMPLE-POL-3. Field 12. Vendor dependency: product: gateway firmware 4.2 (sample). Vendor status evidence: vendor email of 12 Nov 2026 giving the planned quarter. Last check-in: 12 Nov 2026. Next check-in: 14 Dec 2026 (one month after 12 Nov is Saturday 12 Dec 2026, so the next working day). ## D. Decision and expiry Field 13. Approver: Chief Information Security Officer. Name: (the person holding the role; the sample names a role so as not to invent a person). Signed 19 Nov 2026. "I accept the residual risk in Field 11 until the date in Field 14." Field 14. Expiry: 17 Feb 2027 (90 days from the decision). Lapses unless renewed in writing. ## E. Only if PCI DSS applies Field 15. Not applicable in this sample. Prepared by: Platform lead (sample role). Reviewed by: Security lead (sample role). ## The dates, worked out Date arithmetic for the sample by this site, in date order. The 192-day rule is FedRAMP VER-TFR-MAV and applies to FedRAMP providers; the sample organization is not one. | Date | How it is derived | Where it appears | |---|---|---| | 14 Oct 2026 | First detection (sample) | Field 2 | | 21 Oct 2026 | Evaluation finished (sample) | Field 2 | | 12 Nov 2026 | Vendor email and last check-in (sample) | Field 12 | | 19 Nov 2026 | Decision and signature (sample) | Field 13 | | 14 Dec 2026 | Next vendor check-in: one month after the last is Saturday 12 Dec 2026, so the next working day, Monday | Field 12 | | 18 Dec 2026 | First 30-day review, set one day inside the 30 days | Field 10 | | 15 Feb 2027 | Task 2 decision date, two days before expiry | Field 9 | | 17 Feb 2027 | Decision date plus 90 days, chosen by the approver | Field 14 | | 1 May 2027 | Evaluation date plus 192 days, the point at which FedRAMP's 2026 rule requires an unfixed finding to be categorized as accepted; shown for comparison, not on the form | Not on the form | ## Choices that make the sample record hold up The approver field names a role and says why: NIST SP 800-37 Rev. 2 states "The explicit acceptance of risk is the responsibility of the authorizing official and cannot be delegated to other officials within the organization" ([Task R-4](https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-37r2.pdf)). That sentence covers federal systems. For other organizations SP 800-37 names no title, so the policy must say who signs; the [who may sign page](https://remediationvsmitigation.com/risk-acceptance-form/who-may-sign/) sets out the options. A real record carries the person's name. Both controls carry an owner, a verification method and a date, which answers SP 800-40 Rev. 4: "Make sure that the risk response continues to be in place" ([Section 2.2, step 3d](https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf)). The vendor field is filled because the fix depends on one, and the legacy FedRAMP playbook asks for "evidence of vendor interactions regarding the status of patches/fixes" ([playbook](https://www.fedramp.gov/legacy/playbook/csp/authorization/poam/)), which is why field 12 cites the vendor's email. The expiry falls inside the vendor's planned quarter, so the lapse arrives when the vendor's status is known. ## Four ways a record like this fails - **No expiry date.** Field 14 left blank turns a dated decision into permanent acceptance; [maximum durations](https://remediationvsmitigation.com/risk-acceptance-form/maximum-durations/) shows what each source allows. - **Controls named but not checked.** "Access limited" with no owner and no verification date does not meet SP 800-40 Rev. 4 Section 2.2, step 3d. - **Signed by the requester.** NIST SP 800-37 and the FedRAMP rules do not address it, so the policy must say that the person who asked for the exception does not sign it. - **Residual risk missing.** SP 800-37 Task R-3 says "there remains a degree of residual risk" whatever the response; a record that omits it has not been compared with tolerance. ## What happens on 18 Dec 2026 The first review has three outcomes. The record stays as signed, the exception closes because the vendor shipped a kernel with the fix and the rescan no longer detects the flaw, or the case escalates because a trigger in field 10 fired. Each outcome becomes a row change in the [exception register](https://remediationvsmitigation.com/exception-register/). [Risk exception expiry](https://remediationvsmitigation.com/guides/exception-expiry-and-renewal/) covers what the signer does on the expiry date, and [residual risk after mitigation](https://remediationvsmitigation.com/guides/residual-risk-after-mitigation/) covers how the figure in field 11 is formed. ## Sources - NVD, CVE-2026-80521 (description, CVSS 3.1 vector and base score): https://services.nvd.nist.gov/rest/json/cves/2.0?cveId=CVE-2026-80521 (read 19 Nov 2026) - CISA KEV catalog JSON feed, catalogVersion 2026.11.18 (no entry for CVE-2026-80521): https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json (read 19 Nov 2026) - Red Hat Security Data API, CVE-2026-80521 (mitigation statement): https://access.redhat.com/hydra/rest/securitydata/cve/CVE-2026-80521.json (read 19 Nov 2026) - kernel.org, ChangeLog-7.0.14 (no fix commit) and ChangeLog-7.1.10 (fix commit present): https://cdn.kernel.org/pub/linux/kernel/v7.x/ChangeLog-7.0.14 (read 19 Nov 2026) - kernel.org, list of current releases: https://www.kernel.org/releases.json (read 19 Nov 2026) - NIST SP 800-37 Rev. 2, Tasks R-3 and R-4: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-37r2.pdf (read 19 Nov 2026) - NIST SP 800-40 Rev. 4, Section 2.2 step 3d: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf (read 19 Nov 2026) - FedRAMP legacy documentation, Plans of Action and Milestones, POA&M Do's and Don'ts: https://www.fedramp.gov/legacy/playbook/csp/authorization/poam/ (read 19 Nov 2026) - FedRAMP Consolidated Rules for 2026, VER-TFR-MAV: https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-evaluation-and-reporting/ (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Risk Acceptance Form: Free Template With 15 Fields](https://remediationvsmitigation.com/risk-acceptance-form/index.md) - [Maximum Duration of a Risk Acceptance: 9 Sources](https://remediationvsmitigation.com/risk-acceptance-form/maximum-durations/index.md) - [Who Can Accept Risk: Signer by Standard (2026)](https://remediationvsmitigation.com/risk-acceptance-form/who-may-sign/index.md) - [Exception Register: 15 Columns, States and Blank CSV](https://remediationvsmitigation.com/exception-register/index.md) - [Recording an Exception an Auditor Accepts: Eight Items](https://remediationvsmitigation.com/guides/recording-an-exception-an-auditor-accepts/index.md) - [Risk Exception Expiry: What Happens and How to Renew](https://remediationvsmitigation.com/guides/exception-expiry-and-renewal/index.md) - [Residual Risk After Mitigation: Define, Estimate, Record](https://remediationvsmitigation.com/guides/residual-risk-after-mitigation/index.md) --- # How long a risk acceptance can last Canonical: https://remediationvsmitigation.com/risk-acceptance-form/maximum-durations/ FedRAMP's 192 days is a reporting trigger, not a permission: a finding still open 192 days after evaluation must be reported as accepted. BOD 26-02 has a 24-month step for end-of-support edge devices, PCI DSS Requirement 6.3.3 asks critical patches within one month, NIST sets no maximum and CISA allows no waivers. Read 19 Nov 2026. ## Nine sources and what each limits A source that sets no number is listed because its silence matters to the expiry date. The maximum duration each source sets, with its own wording; read 19 Nov 2026. | Source | Limit | Wording | |---|---|---| | [FedRAMP 2026, VER-TFR-MAV](https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-evaluation-and-reporting/) | A finding open 192 days after evaluation is reported as an accepted vulnerability. Rev5 providers must follow the rules from 7 Dec 2026 (initial and ongoing certification); the grace period ends 7 Mar 2027. | "Providers MUST categorize any vulnerability that is not or will not be fully mitigated or remediated within 192 days of evaluation as an accepted vulnerability." | | [FedRAMP legacy playbook](https://www.fedramp.gov/legacy/playbook/csp/authorization/poam/) | Critical and High 30 days, Moderate 90, Low 180, from discovery | "FedRAMP requires Critical and High risks to be remediated within 30 days of discovery, Moderate risks within 90 days of discovery, and Low risks within 180 days of discovery." | | [CISA BOD 26-04, Table 1](https://www.cisa.gov/news-events/directives/bod-26-04-prioritizing-security-updates-based-risk) | 3, 14 or 60 calendar days, or "Fix on system upgrade", by exposure, KEV listing, automatability and technical impact | "Days are calendar days." Table 1 cells: "3 days & forensic triage", "3 days", "14 days", "60 days", "Fix on system upgrade". | | [CISA BOD 26-04 implementation guidance](https://www.cisa.gov/news-events/directives/bod-26-04-implementation-guidance-prioritizing-security-updates-based-risk) | No waivers and no exceptions | "CISA does not issue waivers or exceptions for actions required in cyber directives." | | [CISA BOD 26-02](https://www.cisa.gov/news-events/directives/bod-26-02-mitigating-risk-end-support-edge-devices) | End-of-support edge devices: steps at 12, 18 and 24 months from issuance on 5 Feb 2026, so 5 Feb 2027, 5 Aug 2027 and 5 Feb 2028; does not apply to operational technology | "Within eighteen (18) months of issuance, all FCEB agencies shall: Decommission all identified EOS edge devices from agency networks" | | [PCI SSC FAQ 1597](https://www.pcisecuritystandards.org/faqs/1597) and [FAQ 1046](https://www.pcisecuritystandards.org/faqs/1046) | PCI DSS Requirement 6.3.3: critical vulnerabilities one month from release; other patches on the entity's own risk-based time frames. Neither FAQ sets a duration for a compensating control; an assessor judges whether it meets the intent of the requirement. | "PCI DSS Requirement 6.3.3 specifies time frames for installing security patches/updates: those for critical vulnerabilities must be resolved within one month of release." | | [NIST SP 800-37 Rev. 2](https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-37r2.pdf) | No fixed maximum; the authorizing official sets the termination date and may drop it under ongoing authorization | "Organizations may eliminate the authorization termination date if the system is operating under an ongoing authorization" | | [NIST SP 800-53 Rev. 5, SI-2](https://csrc.nist.gov/projects/cprt/catalog#/cprt/framework/version/SP_800_53_5_2_0/home/SI-2) | No fixed number; the update window is organization-defined | "Organization-defined time periods for updating security-relevant software and firmware may vary based on a variety of risk factors" | | [NIST SP 800-40 Rev. 4](https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf) | No fixed number; exceptions are tracked and reviewed | "Organizations should closely track and monitor all exceptions to maintenance plans." | ## What 192 days does and does not mean FedRAMP defines an accepted vulnerability as one "that the provider does not intend to fully mitigate or remediate, OR that has not or will not be fully mitigated or remediated within the maximum overdue period" ([FedRAMP definitions](https://www.fedramp.gov/2026/definitions/)). An overdue vulnerability is a separate term: one the provider does intend to fix but has not fixed in the recommended time. Accepted findings stay in the monthly report, because providers must report vulnerability activity "at least monthly" (VER-TFR-MHR). BOD 26-02's last step, at 24 months from 5 Feb 2026, is a different kind of limit: it applies to end-of-support edge devices as a whole, not to a single vulnerability. The clock runs from evaluation, not from discovery as in the legacy playbook. For an evaluation finished on 21 Oct 2026, day 192 falls on 1 May 2027 (date arithmetic by this site); the [worked example](https://remediationvsmitigation.com/risk-acceptance-form/example/) uses that date. Before day 192 the rules set shorter "SHOULD" timeframes to partially mitigate, fully mitigate or remediate to a lower impact rating, which depend on the certification class, the impact rating (PAIN), whether the flaw is internet-reachable (IRV) and whether it is likely exploitable (LEV). Timeframes from evaluation in FedRAMP VDR-TFR-PVR (Classes A and B share one table); read 19 Nov 2026. The tables list PAIN-2 to PAIN-5. | Class | Impact rating | Likely exploitable and internet-reachable | Likely exploitable, not internet-reachable | Not likely exploitable | |---|---|---|---|---| | A and B | PAIN-5 | 4 days | 8 days | 32 days | | A and B | PAIN-4 | 8 days | 32 days | 64 days | | A and B | PAIN-3 | 32 days | 64 days | 192 days | | A and B | PAIN-2 | 96 days | 160 days | 192 days | | C | PAIN-5 | 2 days | 4 days | 16 days | | C | PAIN-4 | 4 days | 8 days | 64 days | | C | PAIN-3 | 16 days | 32 days | 128 days | | C | PAIN-2 | 48 days | 128 days | 192 days | | D | PAIN-5 | 12 hours | 1 day | 8 days | | D | PAIN-4 | 2 days | 8 days | 32 days | | D | PAIN-3 | 8 days | 16 days | 64 days | | D | PAIN-2 | 24 days | 96 days | 192 days | ## A mitigation and a KEV date FedRAMP's KEV rule says providers should remediate known exploited vulnerabilities by the CISA due date "even if the vulnerability has been fully mitigated" (VDR-TFR-KEV). A control in front of a KEV item can lower the risk but does not count as the fix against that date. See the [Decision Table row for KEV items with a patch available](https://remediationvsmitigation.com/decision-table/kev-patch-available-exposed/) and the [difference between remediation and mitigation](https://remediationvsmitigation.com/guides/difference-between-remediation-and-mitigation/). BOD 26-04 treats one kind of mitigation differently: "one valid mitigation is to remove the system from the internet; that action changes the value of 'Publicly Exposed' from Yes to No and will shift the required timeline". The shift depends on the other columns of Table 1, as the rows below show. The rows are read from CISA's Table 1; row numbers are CISA's. Agency timeline for remediation in BOD 26-04 Table 1 before and after internet exposure is removed; read 19 Nov 2026. | Facts of the finding | Publicly exposed | No longer exposed | |---|---|---| | On the KEV, automatable, partial control | 3 days (row 2) | 14 days (row 10) | | On the KEV, automatable, total control | 3 days & forensic triage (row 1) | 3 days & forensic triage (row 9) | | On the KEV, not automatable, partial control | 14 days (row 4) | 14 days (row 12) | | Not on the KEV, automatable, partial control | 14 days (row 6) | 60 days (row 14) | ## Setting an expiry when no standard gives a number No standard cited here sets a number for a non-federal organization, so the form asks for a date and the approver chooses it. The anchors are NIST's rule that the official sets the termination date and FedRAMP's 192-day line. NIST SP 800-40 Rev. 4 supports a dated end for any stand-in: "Organizations should plan to replace emergency mitigations with permanent fixes" and "Schedules should be set and enforced for both patch deployment and mitigation removal" (Section 3.5.3). The rule below is this site's method, not a standard's requirement. Tie the expiry to an event that exists, take the earlier of that event and the organization's own cap, and write down which one applied. The worked cases assume an invented policy that caps any acceptance at 90 days with one written renewal, and a decision on 19 Nov 2026. Worked expiry choices under an invented 90-day policy; date arithmetic by this site. | Case | Event the expiry is tied to | Expiry | What happens at expiry | |---|---|---|---| | Fix waits on a vendor who promises a release by 29 Jan 2027 | The vendor release date, earlier than the 90-day cap (17 Feb 2027) | 29 Jan 2027 | Install the fix, or renew with the vendor's new written date. See [risk exception expiry](https://remediationvsmitigation.com/guides/exception-expiry-and-renewal/). | | System is scheduled for decommission on 31 Mar 2027 | The cap binds, because the decommission date is later than the cap | 17 Feb 2027 | One renewal to 18 May 2027 (17 Feb plus 90 days) is allowed only if decommissioning is still on schedule. A date that keeps moving is the failure case; see the [decommission row](https://remediationvsmitigation.com/decision-table/decommission-scheduled/). | | No fix date and no decommission date | Nothing but the cap | 17 Feb 2027 | The renewal decision must be replace, retire or accept again with new facts; a second acceptance without new facts is a silent extension. See [accepting risk on end-of-life software](https://remediationvsmitigation.com/guides/accepting-risk-on-end-of-life-software/). | Each expiry falls on or before the last milestone in field 9 of the [Risk Acceptance Form](https://remediationvsmitigation.com/risk-acceptance-form/), so the lapse forces a decision when new facts exist. For the FedRAMP requests that preceded the 192-day rule see [FedRAMP deviation requests](https://remediationvsmitigation.com/guides/fedramp-deviation-requests/). ## Sources - FedRAMP Consolidated Rules for 2026, VER-TFR-MAV, VER-TFR-MHR and effective dates: https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-evaluation-and-reporting/ (read 19 Nov 2026) - FedRAMP Consolidated Rules for 2026, VDR-TFR-PVR and VDR-TFR-KEV: https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-detection-and-response/ (read 19 Nov 2026) - FedRAMP definitions: accepted vulnerability, overdue vulnerability: https://www.fedramp.gov/2026/definitions/ (read 19 Nov 2026) - FedRAMP legacy documentation, Plans of Action and Milestones, remediation requirements: https://www.fedramp.gov/legacy/playbook/csp/authorization/poam/ (read 19 Nov 2026) - CISA, BOD 26-04, Appendix A and Table 1: https://www.cisa.gov/news-events/directives/bod-26-04-prioritizing-security-updates-based-risk (read 19 Nov 2026) - CISA, BOD 26-04 implementation guidance, waiver question: https://www.cisa.gov/news-events/directives/bod-26-04-implementation-guidance-prioritizing-security-updates-based-risk (read 19 Nov 2026) - CISA, BOD 26-02, Required Actions: https://www.cisa.gov/news-events/directives/bod-26-02-mitigating-risk-end-support-edge-devices (read 19 Nov 2026) - PCI SSC, FAQ 1597, resolving or addressing vulnerabilities: https://www.pcisecuritystandards.org/faqs/1597 (read 19 Nov 2026) - PCI SSC, FAQ 1046, approval of compensating controls: https://www.pcisecuritystandards.org/faqs/1046 (read 19 Nov 2026) - NIST SP 800-37 Rev. 2, Task R-4: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-37r2.pdf (read 19 Nov 2026) - NIST SP 800-53 Rev. 5, SI-2 (OSCAL catalog): https://csrc.nist.gov/projects/cprt/catalog#/cprt/framework/version/SP_800_53_5_2_0/home/SI-2 (read 19 Nov 2026) - NIST SP 800-40 Rev. 4, Sections 3.5.3 and 3.5.5: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-40r4.pdf (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Risk Acceptance Form: Free Template With 15 Fields](https://remediationvsmitigation.com/risk-acceptance-form/index.md) - [Risk Acceptance Form Example: Filled-In Worked Sample](https://remediationvsmitigation.com/risk-acceptance-form/example/index.md) - [Risk Exception Expiry: What Happens and How to Renew](https://remediationvsmitigation.com/guides/exception-expiry-and-renewal/index.md) - [FedRAMP Deviation Requests and Their 2026 Replacements](https://remediationvsmitigation.com/guides/fedramp-deviation-requests/index.md) --- # Who can accept risk in cybersecurity Canonical: https://remediationvsmitigation.com/risk-acceptance-form/who-may-sign/ In US federal systems only the authorizing official can accept risk, and NIST SP 800-37 says the decision cannot be delegated. PCI SSC approves nothing; an assessor judges. CISA grants no waivers. Any other organization names its signer in its own written policy, and a sample rule is below (read 19 Nov 2026). ## Four contexts, four answers Who accepts risk, by context; quotes are in the sections below; read 19 Nov 2026. | Context | Who accepts | May it be delegated? | Source | |---|---|---|---| | US federal information systems (NIST RMF) | The authorizing official; for common controls, the authorizing official for the control provider | No. Carrying out the chosen response can go to a designated representative; the acceptance cannot | [SP 800-37 Rev. 2](https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-37r2.pdf), Tasks R-3 and R-4 | | Cloud services under FedRAMP | The provider records and reports an accepted vulnerability; the agency authorizing official decides what the agency does about it | Not stated | [FedRAMP 2026 agency POA&M page](https://www.fedramp.gov/2026/agencies/use/ongoing/poams/) | | PCI DSS compensating control | The entity proposes; the Qualified Security Assessor judges whether it meets the intent | Not applicable | [PCI SSC FAQ 1046](https://www.pcisecuritystandards.org/faqs/1046) | | Any other organization | The person its own policy names, working inside the written risk tolerance | Set by that policy | [SP 800-39](https://nvlpubs.nist.gov/nistpubs/Legacy/SP/nistspecialpublication800-39.pdf), Task 3-3 | ## The authorizing official cannot hand the decision down NIST SP 800-37 Rev. 2 puts acceptance with one role in two places. Task R-4 says "The explicit acceptance of risk is the responsibility of the authorizing official and cannot be delegated to other officials within the organization." The discussion of Task R-3 adds "Because the authorizing official is the only person who can accept risk", the official reviews the assessment reports and plans of action to decide whether identified risks need mitigation before authorization. The same publication names the one thing a designated representative may not do. The representative coordinates day-to-day risk management, and "the only activity that cannot be delegated by the authorizing official to the designated representative is the authorization decision and signing of the associated authorization decision document (i.e., the acceptance of risk)". Task R-3 names the authorizing official or the designated representative as primarily responsible for identifying and carrying out the risk response, which is why a team can assign the fix or the mitigation to an engineer but not the signature. NIST defines the role as "a senior Federal official or executive with the authority to authorize (i.e., assume responsibility for) the operation of an information system" ([glossary](https://csrc.nist.gov/glossary/term/authorizing_official)). ## FedRAMP splits the record from the decision Under the 2026 rules a provider reports each accepted vulnerability with eight items, listed in [VER-RPT-AVI](https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-evaluation-and-reporting/): tracking identifier, time and source of detection, time of completed evaluation, internet-reachable or not, likely exploitable or not, the Potential Agency Impact N-rating, why it is accepted, and any supplementary information. Seven of the eight sit in fields 1, 2 (detection and evaluation times), 4, 5, 6 and 7 of the [Risk Acceptance Form](https://remediationvsmitigation.com/risk-acceptance-form/); the eighth is free text the provider chooses. FedRAMP's agency guidance gives the agency its own record: "The POA&M belongs to the agency when the action belongs to the agency," and "The agency authorizing official identifies an agency action needed to keep the system within the agency's risk tolerance." The provider does not sign for the agency, and the agency's official does not sign the provider's internal exception. ## PCI DSS and CISA directives have no signer to name PCI SSC states it cannot approve specific compensating controls, and that Qualified Security Assessors decide whether a compensating control meets the intent of a requirement ([FAQ 1046](https://www.pcisecuritystandards.org/faqs/1046)). A separate route exists for legal limits: the "Compliant but with Legal exception" option "also requires additional review from the acquirer or payment brand to whom compliance is being reported" ([FAQ 1486](https://www.pcisecuritystandards.org/faqs/1486)). For KEV actions under a CISA directive there is no one to sign, because CISA does not issue waivers or exceptions for actions required in cyber directives ([BOD 26-04 guidance](https://www.cisa.gov/news-events/directives/bod-26-04-implementation-guidance-prioritizing-security-updates-based-risk)). See the KEV row of the [Decision Table](https://remediationvsmitigation.com/decision-table/). ## Who signs outside the federal case NIST SP 800-39 gives the test but no title. Task 3-3 reads "Organizations determine acceptable degrees of residual risk based on organizational risk tolerance and the specific risk tolerances of particular decision makers." This is the SP 800-39 text; the Task R-3 discussion in SP 800-37 carries the shorter form, "based on organizational risk tolerance", without the clause about decision makers. SP 800-39 adds that "There is no correct level of organizational risk tolerance", and that the residual risk accepted by authorizing officials is "within the organizational risk tolerance, and not the individual risk tolerances of those authorizing officials." So the organization writes its tolerance down, names who may sign at each level, and the signer applies that limit and not a personal one. NIST and FedRAMP do not say whether the person who asks for an exception may also sign it, so the policy must. A requester who signs their own exception has no second check. The sample below is one reasonable arrangement, not a requirement of any standard; change the roles and levels to fit the organization. ## Sample signing rule for risk acceptance 1. A finding may be accepted only by a signed [Risk Acceptance Form](https://remediationvsmitigation.com/risk-acceptance-form/) with an expiry date, recorded in the [exception register](https://remediationvsmitigation.com/exception-register/). 2. Who signs, by residual risk after any mitigation (use the organization's own rating scale): - Low: the head of the team that owns the system. - Moderate: the business owner of the system and the head of security. - High: the executive named in the risk policy (for example the CIO or CEO), after the head of security has recorded a written opinion. 3. The person who requests the exception may not sign it. The requester's manager may sign only at the Low level. 4. A signer may not accept a risk above the written organizational risk tolerance, whatever their personal view. 5. The fix or the mitigation may be assigned to an engineer; the signature may not be assigned downward. 6. The signer's name, role and date go in field 13 of the form, and the role goes in the register's approver column. 7. A KEV due date under a CISA directive cannot be accepted by anyone; the rule does not apply to it. The risk owner named in NIST IR 8286 is a different role from the signer; [who is the risk owner](https://remediationvsmitigation.com/guides/risk-owner/) sets them side by side, and [risk acceptance versus risk tolerance](https://remediationvsmitigation.com/guides/risk-acceptance-vs-risk-tolerance/) covers the written limit the signer works within. ## Sources - NIST SP 800-37 Rev. 2, Tasks R-2, R-3, R-4 and the designated representative role: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-37r2.pdf (read 19 Nov 2026) - NIST CSRC glossary, authorizing official: https://csrc.nist.gov/glossary/term/authorizing_official (read 19 Nov 2026) - NIST SP 800-39, Task 3-3 and the risk tolerance discussion: https://nvlpubs.nist.gov/nistpubs/Legacy/SP/nistspecialpublication800-39.pdf (read 19 Nov 2026) - FedRAMP 2026, Managing POA&Ms (agencies): https://www.fedramp.gov/2026/agencies/use/ongoing/poams/ (read 19 Nov 2026) - FedRAMP 2026, VER-RPT-AVI: https://www.fedramp.gov/2026/providers/rev5/rules/vulnerability-evaluation-and-reporting/ (read 19 Nov 2026) - PCI SSC FAQ 1046: https://www.pcisecuritystandards.org/faqs/1046 (read 19 Nov 2026) - PCI SSC FAQ 1486: https://www.pcisecuritystandards.org/faqs/1486 (read 19 Nov 2026) - CISA BOD 26-04 implementation guidance, FAQ: https://www.cisa.gov/news-events/directives/bod-26-04-implementation-guidance-prioritizing-security-updates-based-risk (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Risk Acceptance Form: Free Template With 15 Fields](https://remediationvsmitigation.com/risk-acceptance-form/index.md) - [Exception Register: 15 Columns, States and Blank CSV](https://remediationvsmitigation.com/exception-register/index.md) - [Maximum Duration of a Risk Acceptance: 9 Sources](https://remediationvsmitigation.com/risk-acceptance-form/maximum-durations/index.md) - [Who Is the Risk Owner for a Security Finding?](https://remediationvsmitigation.com/guides/risk-owner/index.md) - [Risk Acceptance vs Risk Tolerance: Who Sets the Limit](https://remediationvsmitigation.com/guides/risk-acceptance-vs-risk-tolerance/index.md) --- # Terms of use Canonical: https://remediationvsmitigation.com/terms/ Data and forms on the site are licensed under Creative Commons Attribution 4.0 (CC BY 4.0), first edition 19 Nov 2026. Reuse them with the cite line below. Text quoted from CISA, NIST, FedRAMP, PCI SSC and vendors stays under its issuer's terms, and the site gives no exploit code. ## What the license covers The decision rows, the control entries, the risk acceptance fields, the date list and the forms are published under [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/). You may copy, adapt and redistribute them, including commercially, if you give credit, link the license and mark any change. License by type of material. Read 19 Nov 2026. | Material | Terms | |---|---| | Data files (decision rows, controls, risk acceptance fields, dates) | CC BY 4.0 | | Forms, including the Risk Acceptance Form in .docx and PDF | CC BY 4.0 | | Short quotes from CISA, NIST, FedRAMP, PCI SSC, OWASP, Microsoft and vendors | Remain under the issuer's terms; quoted for citation with the source URL and read date | | Third-party sites reached by links | Their own terms and policies | ## Cite line Remediation vs Mitigation, "", first edition 19 Nov 2026, remediationvsmitigation.com//. Licensed CC BY 4.0. If you reuse a table or a form, keep the read dates that come with it. A fact is correct as of its read date; the standards behind it, such as BOD 26-04 and the FedRAMP 2026 rules, can change, and the [corrections page](https://remediationvsmitigation.com/corrections/) logs what the site changed. ## Use of the pages and tools - The pages report where published guidance points. They are not an assessor's finding, a regulator's ruling or a certification. A PCI DSS compensating control is judged by a Qualified Security Assessor, per [PCI SSC FAQ 1046](https://www.pcisecuritystandards.org/faqs/1046). - The decision tool and the form run in your browser and send nothing. What you record with them is yours. - Do not submit exploit code, payloads or live vulnerability details to the editor address. The site does not publish them. - Links to outside sites are for sources. The site does not control them and does not say they are complete. ## Changes to these terms A change to these terms is dated on this page. Questions go to [editor@remediationvsmitigation.com](mailto:editor@remediationvsmitigation.com); how the site stores visits is on the [privacy page](https://remediationvsmitigation.com/privacy/), and the rules for sources are on the [editorial policy](https://remediationvsmitigation.com/editorial-policy/). ## Sources - Creative Commons, Attribution 4.0 International deed: https://creativecommons.org/licenses/by/4.0/ (read 19 Nov 2026) - PCI SSC FAQ 1046: https://www.pcisecuritystandards.org/faqs/1046 (read 19 Nov 2026) Last reviewed 19 Nov 2026 ## Related - [Privacy: Cookies, Logs and the 400-Day Limit](https://remediationvsmitigation.com/privacy/index.md) - [Editorial Policy: Sources, Labels and Corrections](https://remediationvsmitigation.com/editorial-policy/index.md) - [Corrections Log: How to Report an Error](https://remediationvsmitigation.com/corrections/index.md) - [About Remediation vs Mitigation: Scope and Hosting](https://remediationvsmitigation.com/about/index.md)