[slurm-users] Unable to contact slurm controller

[Mahmood Naderan]

Hi,
It seems that squeue is broken due to the following error:

[root at rocks7 ~]# squeue
slurm_load_jobs error: Unable to contact slurm controller (connect failure)
[root at rocks7 ~]#  systemctl restart slurmd
[root at rocks7 ~]#  systemctl restart slurmctld
[root at rocks7 ~]# squeue
slurm_load_jobs error: Unable to contact slurm controller (connect failure)
[root at rocks7 ~]# ps aux | grep slurm
root      2969  0.0  0.0 343112  3268 ?        Sl   Jul07   0:12
/usr/sbin/slurmdbd
kouhika+ 22930  0.0  0.0   4348   348 pts/2    S+   Jul30   0:00
/usr/libexec/slurm-spank-x11 -t compute-0-6 -i 803.0 -cgw -s ssh -o
kouhika+ 22931  9.7  0.0 192296 20292 pts/2    S+   Jul30 145:28 ssh -Y
compute-0-6 /usr/libexec/slurm-spank-x11 -i 803.0 -c -g -w -s "ssh" -o ""
root     28532  0.0  0.0 143132  2072 ?        Sl   20:02   0:00
/usr/sbin/slurmd
root     29364  0.0  0.0 112712   964 pts/12   S+   20:03   0:00 grep
--color=auto slurm


As you can see I tried to restart slurm processes, however, has no effect.
Any thought?


Regards,
Mahmood
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.schedmd.com/pipermail/slurm-users/attachments/20180731/173546bb/attachment.html>