When  I change MessageTimeout=300 , I get the following error when I start
slurmd
rsh y119 "/etc/init.d/emunge start; /etc/init.d/slurm start &"
Starting MUNGE: munged[  OK  ]
scontrol: WARNING: MessageTimeout is too high for effective fault-tolerance
starting slurmd: slurmd: WARNING: MessageTimeout is too high for effective
fault-tolerance

Slurm is still running though. I will play decreasing this value to see if
it helps.

Thanks for you help

Reply via email to