Is there anyone currently running Torque and Maui on OS X that could help out?
The system I am having problems with is an OS X Server 10.5.8. Currently running Torque 2.4.7 and Maui 3.3, although that was upgraded after the problems began. Will revert to earlier versions if needed. MAUI keeps getting into a state where it no longer schedules jobs, and even running "showq" lists no running jobs - however jobs are still running. The only thing that shows up is the deferred jobs. Restarting Maui corrects the problem for some amount of time. It may be 10-20 minutes, or a few days before it stops responding again. I have been looking through the log files with Level 9 (lots of logs) and the main thing I find happening is that an "INFO" log changes from one state (no PBS sched socket) to another (invalid PBS sched socket). Whilst in the second state (invalid) the system no longer shows the real queue status. The two states are listed below: I am not sure if this is a problem with Maui or Torque. It might be a problem with the configuration of the machine, but hopefully someone can point me in the right direction. Working state... 05/25 10:56:39 MRMCheckEvents() 05/25 10:56:39 INFO: no PBS sched socket connections ready 05/25 10:56:39 MSUAcceptClient(5,ClientSD,HostName,TCP) 05/25 10:56:39 INFO: accept call failed, errno: 35 (Resource temporarily unavailable) 05/25 10:56:39 INFO: all clients connected. servicing requests Broken state... 05/25 09:59:30 MRMCheckEvents() 05/25 09:59:30 INFO: invalid PBS sched socket 05/25 09:59:30 MSUAcceptClient(5,ClientSD,HostName,TCP) 05/25 09:59:30 INFO: accept call failed, errno: 35 (Resource temporarily unavailable) 05/25 09:59:30 INFO: all clients connected. servicing requests Thanks, Craig. -- Craig West Systems Manager Victorian Partnership for Advanced Computing 110 Victoria Street, Carlton South VIC 3053 P: +61 3 9925 4751 E: [email protected] http://www.vpac.org _______________________________________________ mauiusers mailing list [email protected] http://www.supercluster.org/mailman/listinfo/mauiusers
