Live › vulnerability › CVE-2026-65165
NVIDIA Slurm, the cluster job scheduler, lets a user alter the nodelist passed with the --distribution=arbitrary option. A user can use this to reach compute nodes outside their own job allocation.
Summary written from the sources below. Check the originals before acting.
First seen by InfraStack 2026-11-03 16:21 UTC. All times on this page are UTC.
Clusters that run NVIDIA Slurm and let ordinary users submit jobs are the ones in scope. The sources do not name the affected versions, so the team has to check the installed version against NVD. Single-user clusters where every user already owns all nodes are unlikely to be affected.
No product ranges from the sources yet.
sinfo --version or slurmctld -V, then compare it to the versions listed in the NVD entry.No fix yet, per the sources.
Each row is the first time that source carried this signal. The analyst’s rows are its own findings.
NVIDIA Slurm contains a vulnerability where a user could deliberately alter the provided nodelist alongside the --distribution=arbitrary option to gain access to compute nodes outside of the current job allocation.
This page as .md · .json · in your agent: get_signal("CVE-2026-65165")
NVD data is provided by NIST and is public domain. Posts quoted belong to their authors.