Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Deal with batchless process data from "slurm" node that happens to be interactive #534

Open
lars-t-hansen opened this issue Jun 26, 2024 · 2 comments
Labels
component:sonalyze sonalyze/* task:bug Something isn't working

Comments

@lars-t-hansen
Copy link
Collaborator

(This is a backend problem not a dashboard issue.)

The 183K cores here makes everything be wrong. Not sure what job=0 is about either.

image

@lars-t-hansen lars-t-hansen added task:bug Something isn't working component:sonalyze sonalyze/* labels Jun 26, 2024
@lars-t-hansen lars-t-hansen self-assigned this Aug 26, 2024
@lars-t-hansen
Copy link
Collaborator Author

"Job 0" is an interactive job. gpu-3 does not actually run Slurm.

This explains the weird peak - this is bug #77 again, once again on a batchless system.

@lars-t-hansen
Copy link
Collaborator Author

Probably the reality is going to be that the admins will make nodes interactive or not without thinking about sonar. So we probably need to add functionality to Jobanalyzer or Sonar to recover. It is definitely possible to do it in Sonar - as there will not be a slurm job ID, we can assume that the job is interactive and we should use the other algorithm instead.

@lars-t-hansen lars-t-hansen changed the title Job query reports implausible data Deal with batchless process data from "slurm" node that happens to be interactive Aug 27, 2024
@lars-t-hansen lars-t-hansen removed their assignment Oct 31, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
component:sonalyze sonalyze/* task:bug Something isn't working
Projects
None yet
Development

No branches or pull requests

1 participant