On 8/1/24 02:02, Sid Young via slurm-users wrote:
I've been waiting for node to become idle before upgrading them however some jobs take a long time. If I try to remove all the packages I assume that kills the slurmstep program and with it the job.
Can you be more specific about what you mean by "upgrade"? Which Slurm version are you running? Why would you want to remove all the packages?
For slurmd and slurmstepd the quick and usually OK procedure would be to simply update the RPMs while jobs are running!
There is also a more safe procedure where the nodes are first drained before upgrading slurmd, see the Wiki page https://wiki.fysik.dtu.dk/Niflheim_system/Slurm_installation/#upgrade-slurmd...
IHTH, Ole