Re: [Condor-users] a job submited to condor-g never stops running

[Dan Bradley <dan@xxxxxxxxxxxx>]

I don't know of a specific cause for this sort of behavior.  You may  
want to compare how it works both with and without the "gridmonitor"  
enabled in your condor configuration.  Example:

ENABLE_GRID_MONITOR = False

--Dan

On Jul 11, 2006, at 5:03 PM, Olga Kornievskaia wrote:

> Hi,
>
> Has anybody ever had a problem like this. I submit the following  
> script:
>
> universe = grid
> grid_type = gt2
> globusscheduler = zen.citi.umich.edu/jobmanager-condor
> executable = sh_loop
> arguments = 600
> x509userproxy = /tmp/x509_proxy_aglo
> should_transfer_files = true
> whentotransferoutput = on_exit
> MyProxyHost     = yoga.citi.umich.edu:7512
> MyProxyPassword = foobar
> MyProxyServerDN = /C=US/ST=Michigan/L=Ann Arbor/O=University of
> Michigan/OU=CITI Production
> KCA/CN=myproxy/yoga.citi.umich.edu/emailAddress=aglo@xxxxxxxxxxxxxx
> MyProxyNewProxyLifetime = 5
> MyProxyRefreshThreshold = 180
> error = script6.err
> output = script6.out
> log = script6.log
> queue
>
> The job is successfully starts and finishes in condor. StarterLog has:
>
> 7/11 17:19:33 Starting a VANILLA universe job with ID: 42.0
> 7/11 17:19:33 IWD: /home/aglo/gram_scratch_UW0W4vAkqk
> 7/11 17:19:33 Output file:
> /home/aglo/.globus/job/zen.citi.umich.edu/28687.1152652478/stdout
> 7/11 17:19:33 Error file:
> /home/aglo/.globus/job/zen.citi.umich.edu/28687.1152652478/stderr
> 7/11 17:19:33 Using wrapper /usr/local/bin/spkm3-wrapper to exec
> condor_exec.exe
> /home/aglo/.globus/.gass_cache/local/md5/32/ 
> 0e2d7cf69a8d3a612c2a49374d3087/md5/5a/53a49f318032335ea7bf518346b3f4/ 
> data
> 600
> 7/11 17:19:33 Create_Process succeeded, pid=28812
> 7/11 17:29:41 Process exited, pid=28812, status=0
> 7/11 17:29:42 Got SIGQUIT.  Performing fast shutdown.
> 7/11 17:29:42 ShutdownFast all jobs.
> 7/11 17:29:42 **** condor_starter (condor_STARTER) EXITING WITH STATUS  
> 0
>
> Yet. the job's status in condor's queue never changes from "running".
> GridmanagerLog happily spews out a continum of messages that contain
> what looks like queries for results:
>
> 7/11 17:59:30 [28682] Using constraint
> ((Owner=?="aglo"&&JobUniverse==9)) && ((Managed =!= "ScheddDone")) &&
> (JobStatus == 3 || JobStatus == 4 || (JobStatus == 5 && Managed =?=
> "External"))
> 7/11 17:59:30 [28682] Fetched 0 job ads from schedd
> 7/11 17:59:30 [28682] leaving doContactSchedd()
> 7/11 17:59:33 [28682] GAHP[28683] <- 'RESULTS'
> 7/11 17:59:33 [28682] GAHP[28683] -> 'S' '0'
>
> The job never gets done. Any ideas why? Thanks.
> _______________________________________________
> 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 either
> https://lists.cs.wisc.edu/archive/condor-users/
> http://www.opencondor.org/spaces/viewmailarchive.action?key=CONDOR