[mvapich-discuss] Issues running mvapich2 with slurm

Matthew Russell matthew.g.russell at gmail.com
Mon Oct 22 15:33:51 EDT 2012


>
>> Basically you'll want to make sure memlock is set to unlimited in
>> /etc/security/limits.conf and that slurm is respecting this as well.  On
>> our systems we have added `ulimit -l unlimited' into
>> /etc/sysconfig/slurm (redhat systems).
>>
>> Hope this info helps.
>>
>
I made all the changes to sshd and limits.conf on one node as prescribed in
the mvapich2 and restarted that node.. No difference.

Then, on the advice of the slurm manual, I tried to override the propagate
options, ie.,
[matt at dena]/cm/shared/apps/slurm/2.4.2% srun -N 2 --propagate=MEMLOCK
~matt/cluster_tests/ulimit.sh
slurmd[dena2]: Can't propagate RLIMIT_MEMLOCK of 'unlimited' from submit
host: Invalid argument
slurmd[dena1]: Can't propagate RLIMIT_MEMLOCK of 'unlimited' from submit
host: Invalid argument
dena2: 64
dena1: 64

I'll now try with the debug version of mvapich.

Also, for using slurm, must I always run salloc first?  The Bright Cluster
Management User Guide I have says I should only have to run srun, not first
salloc..
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.cse.ohio-state.edu/pipermail/mvapich-discuss/attachments/20121022/c0301740/attachment.html


More information about the mvapich-discuss mailing list