Executive brief
Keylime is a security tool used to monitor the integrity of remote computers using specialized hardware (TPM). A flaw was found where the system uses a predictable security code instead of a random one when verifying a computer's health. This allows an attacker who already has administrative access to a monitored machine to trick the central monitoring system into believing the machine is still secure even after it has been compromised.
Technical details
A vulnerability exists in the Keylime verifier's TPM quote attestation process when using the push model deployment. The verifier utilizes a hardcoded challenge nonce rather than a cryptographically random value. An attacker with root privileges on an enrolled monitored machine can exploit this by using tools like tpm2_quote to stockpile valid TPM quotes while the system is in a clean state. These quotes can subsequently be replayed to the verifier to maintain a 'trusted' status even after the system has been compromised, effectively bypassing runtime integrity monitoring. This issue does not affect pull model deployments. Red Hat has released patches for RHEL 10 to address this flaw.
Affected products
- Keylime Keylime Affected in push model deployment
- Red Hat Red Hat Enterprise Linux 9
- Red Hat Red Hat Enterprise Linux 10 Fixed in 7.14.1-5.el10_2.1
Timeline
- 2026-04-16: other: Reported to Red Hat Bugzilla
- 2026-05-06: disclosed: NVD Published Date
- 2026-06-23: patched: Red Hat released RHSA-2026:28582 for RHEL 10