Hi! I have a slurm 16.05.4 on fedora 23 and when i try to start slurmctld i get 
this :
Sep 03 21:17:40 sev.spacescience.ro slurmctld[8286]: slurmctld version 16.05.4 
started on cluster sev
Sep 03 21:17:40 sev.spacescience.ro audit[1]: SERVICE_START pid=1 uid=0 
auid=4294967295 ses=4294967295 msg='unit=slurmctld comm="systemd" 
exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=success'
Sep 03 21:17:40 sev.spacescience.ro slurmctld[8286]: error: 
plugin_load_from_file: dlopen(/usr/lib64/slurm/jobacct_gather_linux.so): 
/usr/lib64/slurm/jobacct_gather_linux.so: undefined symbol: proctrack_g_get_pids
Sep 03 21:17:40 sev.spacescience.ro polkitd[978]: Unregistered Authentication 
Agent for unix-process:8278:473530 (system bus name :1.34, object path 
/org/freedesktop/PolicyKit1/AuthenticationAgent, locale en_US.utf-8) 
(disconnected from bus)
Sep 03 21:17:40 sev.spacescience.ro slurmctld[8286]: error: Couldn't load 
specified plugin name for jobacct_gather/linux: Dlopen of plugin file failed
Sep 03 21:17:40 sev.spacescience.ro systemd[1]: slurmctld.service: Main process 
exited, code=exited, status=1/FAILURE
Sep 03 21:17:40 sev.spacescience.ro slurmctld[8286]: error: cannot create 
jobacct_gather context for jobacct_gather/linux
Sep 03 21:17:40 sev.spacescience.ro systemd[1]: slurmctld.service: Unit entered 
failed state.
Sep 03 21:17:40 sev.spacescience.ro audit[1]: SERVICE_STOP pid=1 uid=0 
auid=4294967295 ses=4294967295 msg='unit=slurmctld comm="systemd" 
exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=failed'
Sep 03 21:17:40 sev.spacescience.ro slurmctld[8286]: fatal: failed to 
initialize jobacct_gather plugin

any idea about this error? the slurm was packaged on the same system
so any requirements for build are still present...

also there is another problem with slurmctld.service :
the pid file is defined in the .service file so if one chose the
customize the pid file location in slurm.conf will get 
slurmctld.service: PID file /var/run/slurmctld.pid not readable (yet?) after 
start: No such file or directory
so, i think that either PIDFile= should be not setup in .service file or 
SlurmctldPidFile=
should be removed from slurm.conf (or somehow the daemon should be aware 
conflicting options
and ignore conf one (i think that the systemd pid definition is required...))

Thank you!!
Adrian

Reply via email to