Mailing List Archives

Public Access

 		[Computer Systems Lab]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Condor-users] Odd behaviour of jobs

Robin,
I would double check the PREEMPT and CONTINUE expressions on your workstation. From the log file, it appears that the PREEMPT expression is set to stop the job after 5 minutes. You may want to set this to a longer time, so that the job could reschedule on a less busy machine, or you can just set it to never preempt altogether. 

I hope this helps.

Good luck,
Rob

Robin Harrington wrote:

Hi,

In the job log below, the behaviour is not what I expected. The desired
behaviour is that jobs should be suspended when keyboard activity is
detected and then unsuspended when the system is idle again. But why are
they being evicted only a few minutes later?

Any particular job attribute I should be looking at?

All suggestions are very welcome.

Thanks
Robin

 000 (161.000.000) 05/06 09:08:19 Job submitted from host:
<132.181.5.14:1077>
...
001 (161.000.000) 05/08 00:27:44 Job executing on host:
<132.181.5.14:1080>
...
006 (161.000.000) 05/08 00:27:52 Image size of job updated: 1588
...
006 (161.000.000) 05/08 00:32:52 Image size of job updated: 1600
...
006 (161.000.000) 05/08 02:02:52 Image size of job updated: 1620
...
010 (161.000.000) 05/08 08:37:52 Job was suspended.
	Number of processes actually suspended: 1
...
011 (161.000.000) 05/08 08:39:29 Job was unsuspended.
...
010 (161.000.000) 05/08 08:42:52 Job was suspended.
	Number of processes actually suspended: 1
...
004 (161.000.000) 05/08 08:49:33 Job was evicted.
	(0) Job was not checkpointed.
		Usr 0 08:03:14, Sys 0 00:00:00  -  Run Remote Usage
		Usr 0 00:00:00, Sys 0 00:00:00  -  Run Local Usage
	0  -  Run Bytes Sent By Job
	544042  -  Run Bytes Received By Job
...
001 (161.000.000) 05/08 12:26:57 Job executing on host:
<132.181.5.14:1080>
...
006 (161.000.000) 05/08 12:32:05 Image size of job updated: 1600
...
010 (161.000.000) 05/08 13:42:05 Job was suspended.
	Number of processes actually suspended: 1
...
004 (161.000.000) 05/08 13:47:09 Job was evicted.
	(0) Job was not checkpointed.
		Usr 0 01:08:29, Sys 0 00:00:00  -  Run Remote Usage
		Usr 0 00:00:00, Sys 0 00:00:00  -  Run Local Usage
	0  -  Run Bytes Sent By Job
	544042  -  Run Bytes Received By Job
...

Robin Harrington, Advanced Technologies Manager,
Information and Communication Technology Services,
University of Canterbury, Private Bag 4800, Christchurch, New Zealand. Phone: +64 3 364 2339 Fax: +64 3 364 2332 Email: robin.harrington@xxxxxxxxxxxxxxxx 

_______________________________________________
Condor-users mailing list
To unsubscribe, send a message to condor-users-request@xxxxxxxxxxx with a
subject: Unsubscribe
You can also unsubscribe by visiting
https://lists.cs.wisc.edu/mailman/listinfo/condor-users
The archives can be found at: https://lists.cs.wisc.edu/archive/condor-users/ 

--
===================================
Rob Futrick
main: 888.292.5320

Cycle Computing, LLC
Leader in Condor Grid Solutions
Enterprise Condor Support and CycleServer Management Tools

http://www.cyclecomputing.com
http://www.cyclecloud.com