[Gimle-users] SLURM upgrade on Gimle

Kent Engström kent at nsc.liu.se
Mon Apr 2 14:57:44 CEST 2012


Dear Gimle Users,

there is a bug in the version of SLURM (the resource manager) used on
Gimle at the moment, and it seems that this bug is fixed in a later
version that is available.

The bug causes problems with requeued "risk jobs" (the job starts on the
wrong node and is killed when SLURM discovers its error).

We will try to upgrade SLURM this afternoon. The upgrade should not
affect running jobs, but the squeue/sbatch commands etc could hang
during the short time while the upgrade is done.  We will tell you when
we are done.


-- 
Kent Engström, National Supercomputer Centre
kent at nsc.liu.se, +46 13 28 4444



More information about the Gimle-users mailing list