How to debug when wandb says run has crashed but SLURM continues to run?

W&B Help
1

Hello,

I was wondering if you had any tips on how to debug a situation where wandb says that a run has crashed but the SLURM job continues to run.

Thanks,
Nicole

2

Hi @wongnicolehy,

Thank you for writing in. Could you talk a bit about your setup, as well as your workflow?

3

Hi there Niclole, I wanted to follow up on this request. Please let us know if we can be of further assistance or if your issue has been resolved.

4

Hi Nicole, since we have not heard back from you we are going to close this request. If you would like to re-open the conversation, please let us know!

5

I’d like to re-open the conversation. What specifically do you want to know about setup and workflow?

6

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.