> Try to start it so it does not damonize I think the option is -D but
> better check the man page, see if it core dumps.
I tried two ideas:
First, as we had done previously, I restarted everything:
/etc/init.d/slurm stop
/etc/init.d/slurmdbd stop
/etc/init.d/mysql stop
/etc/init.d/mysql start
/etc/init.d/slurmdbd start
/etc/init.d/slurm start
In this case, slurmdbd has not died, but sacctmgr does not work:
# sacctmgr -vvv
sacctmgr: Accounting storage SLURMDBD plugin loaded with AuthInfo=(null)
sacctmgr: auth plugin for Munge (http://code.google.com/p/munge/) loaded
sacctmgr: debug: slurmdbd: Sent DbdInit msg
sacctmgr: ^C
and mysqld has been working with more than 200% CPU.
Today I have restart slumdbd with -D option, as you said.
slurmdbd has not died either, but we have the same situation:
# sacctmgr -vvv
sacctmgr: Accounting storage SLURMDBD plugin loaded with AuthInfo=(null)
sacctmgr: auth plugin for Munge (http://code.google.com/p/munge/) loaded
sacctmgr: debug: slurmdbd: Sent DbdInit msg
sacctmgr: ^C
and mysqld working with more than 200% CPU.
We are obtaining, in the log file:
2013-03-30T20:59:04] debug2: DBD_JOB_SUSPEND: ID:55951 STATE:RUNNING
[2013-03-30T20:59:04] debug3: 7(as_mysql_job.c:1146) query
update "cccuam_job_table" set time_suspended=1364386896-time_suspended,
state=1 where job_db_inx=62441;update "cccuam_suspend_table" set
time_end=1364386896 where job_db_inx=62441 && time_end=0;
I think slurm is working so hard updating every job we have in the
queue system:
# squeue |wc -l
7171
but it does now work.
Any idea?
Thanks in advance,
Pablo Sanz
--
Pablo Sanz Mercado
Director Técnico
Centro de Computación Científica
Universidad Autónoma de Madrid
