Versions Affected : All versions prior to TrueNAS 13.0-U2

---

Description

When a CAM-managed device responds to a command with an error condition, CAM may automatically retry the command following some error recovery protocol.

For instance, it may send a SCSI START UNIT command to the device before retrying the failed command. In this case, an in-memory copy of the original command is preserved for a later retry. However, a specific portion of the command state was not saved correctly, and upon a retry this could lead to memory corruption. The bug can cause kernel panics or other system-level misbehaviour.

---

Workaround

No workaround is available.

---

Mitigation

• Upgrade to TrueNAS 13.0-U2 or later.

---

Commit

• FreeBSD Revision Hash: db8082886fd8
• TrueNAS Commit
• JIRA Ticket : NAS-117683

---

Further information

• FreeBSD Errata Notice Entry