Fortinet warns of actively exploited FortiMail flaw


Attackers can write files without logging in, potentially enabling code execution. Some fixes remain pending, and workarounds do not remove an existing compromise.
Fortinet says attackers are exploiting a critical flaw in FortiMail, its email security platform, that lets them write files to vulnerable systems without logging in. Files placed in certain locations could let attackers run code or commands on the appliance, according to The Register’s October 2 report.
The vulnerability is tracked as CVE-2026-104286 and has a score of 9.8 under CVSS, a system for rating the severity of security flaws. An attacker can exploit it through specially crafted web requests to FortiMail’s web interface.
Fortinet reports that the following FortiMail versions are affected:
Fortinet lists fixes for several affected branches as “upcoming,” leaving customers on those branches reliant on workarounds until updates arrive. Those workarounds are available measures to limit exposure, not software fixes.
Fortinet recommends disabling the Identity Based Encryption feature if it is not required. If disabling it is not possible, administrators should prevent access to the FortiMail management interface from the internet and allow access only from trusted private networks.
Applying a workaround does not remove files attackers may already have planted or persistence mechanisms—changes that let attackers maintain access. Administrators therefore also need to check whether their systems have been compromised. Fortinet has published indicators to investigate, including suspicious files, configuration changes and IP addresses associated with the attacks.
Fortinet’s advisory does not disclose when the attacks began, who is responsible or how many customers may have been compromised.
CISA has added the flaw to its Known Exploited Vulnerabilities catalog. It has directed US federal civilian agencies to investigate possible compromise and apply mitigations by October 4.
Fortinet warned on October 1 that attackers had used the email security flaw to write files without logging in. Administrators can check for signs of compromise and use workarounds while checking for fixes.
2 min