Skip to content

Conversation

@cmd-ntrf
Copy link
Contributor

@cmd-ntrf cmd-ntrf commented Oct 7, 2025

The number of allocated nodes was infered from metrics associated with the job. There are cases where a node's slurm-job-exporter will report no usage for a job, even if a job was allocated. This typically happens when no process for that job was launched for at least 60 seconds.

This PR makes the allocated number reflects the number of nodes allocated by Slurm and the used number, the usage value infered from prometheus metrics.

The number of allocated nodes was infered from metrics associated
with the job. There are cases where a node's slurm-job-exporter will report
no usage for a job, even if a job was allocated. This typically
happens when no process for that job was launched for at least 60 seconds.

This PR makes the allocated number reflects the number of nodes allocated
by Slurm and the used number, the usage value infered from prometheus metrics.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant