Uppsala Multidisciplinary Center for Advanced Computational Science

Problem with Slurm on Milou -- fixed

2017-04-18

There is currently a problem with Slurm on Milou. We are investinging it. You will likely not be able to submit new jobs and you may encounter strange error messages when running 'sbatch' and 'interactive'.

An example of the error when running sbatch:

" sbatch: error: Batch job submission failed: Unable to contact slurm 
controller (connect failure)"

When running jobinfo:

Clustername not found

Update at 1055 hours

Problem is still not solved. For a while we will run with a much simpler priority handlling on Milou,that at least allows job submits.

Update at 1230 hours

Slurm functions normally again, but we wait a while before declaring that it is really fixed.

Update at 1250

The problem has disappeared, without any explanation. Let us hope that it does not return, and declare that the problem is fixed.

Old System News