CVE-2026-126261CVSS 9FGFM authorization bypass
An unauthenticated attacker that can reach the FortiManager FGFM port can send crafted FGFM packets to execute code or commands.
A reachable FortiManager FGFM port allows unauthenticated code execution on affected builds. A separate flaw allows code execution through an enabled JSON API on FortiManager and FortiAnalyzer variants. Fortinet lists 8.0.3 as upcoming for that HTTP flaw.
The November advisories describe separate paths. Compare the exact CVE, product and branch before upgrading.
Fortinet says the HTTP code execution path requires the JSON API, which is off by default. Use the vendor CLI workaround on each administrator profile with API access enabled while waiting for a fix. Check whether disabling it affects automation and verify the effective permissions after applying the change.
config system admin profile
edit <profile name>
set rpc-permit none
end
The October hardening notice gave product-level guidance before the detailed advisories. In particular, the 8.0.2 target from that notice is still in the affected ranges for CVE-2026-126170. No verified exploitation or victim data was identified for these three CVEs.
CVE-2026-126261CVSS 9An unauthenticated attacker that can reach the FortiManager FGFM port can send crafted FGFM packets to execute code or commands.
CVE-2026-126170CVSS 9.8When the JSON API is enabled, a crafted HTTP request can allow unauthenticated code or command execution in affected FortiManager and FortiAnalyzer products, including cloud variants and FortiAnalyzer-BigData.
CVE-2026-126229CVSS 9.1An authenticated administrator can access other VDOMs via crafted HTTP requests in affected FortiManager and FortiAnalyzer products, including cloud variants.
Nov 10, 2026
Fortinet publishes FGFM, JSON API and cross-VDOM advisories
The advisories describe different access paths and patch targets.
Nov 11, 2026
Fortinet publishes the separate HTTP code execution CVE record
It lists affected 8.0 and 7.6 lines, and an affected FortiAnalyzer-BigData 7.6 line. Some fixes are listed as upcoming.
We list each claim and who made it. Primary sources (the vendor, the researcher, the registry) are marked. Where reports disagree, we grade by the worst credible case.