[torqueusers] Bad ACL entry on start-up and jobs stay in Q forever

Shinta Bonnefoy shinta.bonnefoy at anu.edu.au
Thu Feb 19 18:21:19 MST 2009


Hi,

We have installed Torque 2.3 in our cluster (server running under SLES 10.1 and nodes on OSS 10.3).
It worked very fine until yesterday when we had to restart the server, it's like we "lost" part of our configuration.

Now when I start the pbs_server , I have these errors :
PBS_Server: Bad ACL entry in host list (15073) in manager_oper_chk, bad entry in acl: root at terrawulf
PBS_Server: Bad ACL entry in host list (15073) in manager_oper_chk, bad entry in acl: root at terrawulf


still and all, pbs_server seems to work.
We use pbs_sched as scheduler.


But we face another trouble, now jobs stay in QUEUE without being automatically run.
We have to manually make them run using qrun (which wasn't the case before the reboot).

Do you have any clue of those problems ?

Please advise,
Thanks and regards,
SB








More information about the torqueusers mailing list