[slurm-users] srun: error: Timed out waiting for job step to complete

petitbois.jean at laposte.net petitbois.jean at laposte.net
Mon Nov 27 03:35:47 MST 2017


Hello when i run job with srun and if my node reboot i have this error after 1 minute : 
srun: error: Timed out waiting for job step to complete 

it's possible to change timeout to 3600seconds for example ? to force slurmctld to wait that my node reboot ? 
i try 
srun --wait 3600 

or i add in my slurm.conf : 
resumetimeout=3600 

but i have no effect, can you help me ? 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.schedmd.com/pipermail/slurm-users/attachments/20171127/7819ae7c/attachment.html>


More information about the slurm-users mailing list