[torqueusers] Problem with qsub depend option and Job arrarys
Glen Beane
glen.beane at gmail.com
Fri Feb 13 21:07:08 MST 2009
job arrays aren't finished so I am not surprised this does not work
(actually I think this is a well known issue at this point and I don't
have a date for when it will be fixed)
On Fri, Feb 13, 2009 at 2:44 PM, <z99621 at aol.com> wrote:
> Has anyone had a problem with using the –W depend=afterok:<jobid> not
> working with qsub of job arrarys?
>
> Example below using Torque v2.3.3.
>
> #!/bin/bash
> job1=`qsub doubletest.sh`
> echo job1 is $job1 submitted at `date`
> job2=`qsub -t 1-2 -W depend=afterok:$job1 doubletest.sh`
> echo job2 is $job2 submitted at `date`
> 0D
>
>
> $ qstat
> Job id Name User Time Use S Queue
> ------------------------- ---------------- --------------- -------- - -----
> 459.fl51u101 DoubleTest lee 00:09:59 C batch
> 460-1.fl51u101 DoubleTest-1 lee &nbs p; 00:09:58 C
> batch
> 460-2.fl51u101 DoubleTest-2 lee 00:09:59 C batch
>
> Shows jobs running concurrently.
>
> qstat –f shows dependency option
>
> Job Id: 460-1.fl51u101
> Job_Name = DoubleTest-1
> Job_Owner = lee at fl51u101
> job_state 3D R
> queue = batch
> server = fl51u101
> Checkpoint = u
> ctime = Fri Feb 13 12:21:57 2009
> depend = afterok:459.fl51u101 at fl51u101
> Error_Path = fl51u101:/user/lee/PBS/DoubleTest.e460-1
> exec_host = fl51u101/1+fl51u101/0
> Join_Path = n
> Keep_Files = n
> Mail_Points = a
> mtime = Fri Feb 13 12:21:58 2009
> Output_Path = fl51u101:/user/lee/PBS/DoubleTest.o460-1
> Priority = 0
> qtime = Fri Feb 13 12:21:58 2009
> Rerunable = False
> Resource_List.nodect = 1
> Resource_List.nodes = 1:ppn=2
> Resource_List.walltime = 01:00:00
> session_id = 11228
> substate = 42
> Variable_List = PBS_O_HOME=/user/lee,PBS_O_LANG=en_US.UTF-8,
> PBS_O_LOGNAME=lee,
>
> PBS_O_PATH=/usr/kerberos/bin:/usr/local/bin:/bin:/usr/bin:/usr/X11R6/
> & nbsp; bin,PBS_O_MAIL=/var/spool/mail/lee,PBS_O_SHELL=/bin/ksh,
> PBS_SERVER=fl51u101,PBS_O_HOST=fl51u101,PBS_O_WORKDIR=/user/lee/PBS,
> < FONT face="Courier New" size=2> PBS_O_QUEUE=batch,PBS_ARRAYID=1
> etime = Fri Feb 13 12:21:58 2009
> submit_args = -t 1-2 -W depend=afterok:459.fl51u101 doubletest.sh
> job_array_id = 1
> job_array_request = 1-2
> start_time = Fri Feb 13 12:21:58 2009
> start_count = 1
>
> Job Id: 460-2.fl51u101
> Job_Name = DoubleTest-2
> Job_Owner = lee at fl51u101
> job_state = R
> queue = batch
> server = fl51u101
> Checkpoint = u
> ctime = Fri Feb 13 12:21:57 2009
> depend = afterok:459.fl51u101 at fl51u101
> Error_Path = fl51u101:/user/lee/PBS/DoubleTest.e460-2
> &nbs p; exec_host = fl51u102/1+fl51u102/0
> Join_Path = n
> Keep_Files = n
> Mail_Points = a
> mtime = Fri Feb 13 12:21:58 2009
> Output_Path = fl51u101:/user/lee/PBS/DoubleTest.o460-2
> Priority = 0
> qtime = Fri Feb 13 12:21:58 2009
> Rerunable = False
> Resource_List.nodect = 1
> Resource_List.nodes = 1:ppn=2
> Resource_List.walltime = 01:00:00
> session_id = 3929
> substate = 42
> Variable_List = PBS_O_HOME=/user/lee,PBS_O_LANG=en_US.UTF-8,
> PBS_O_LOGNAME=lee,
>
> PBS_O_PATH=/usr/kerberos/bin:/usr/local/bin:/bin:/usr/bin:/usr/X11R6/
> bin,PBS_O_MAIL=/var/spool/mail/lee,PBS_O_SHELL=/bin/ksh,
> PBS_SERVER=fl51u101,PBS_O_HOST=fl51u101,PBS_O_WORKDIR=/user/lee/PBS,
> PBS_O_QUEUE=batch,PBS_ARRAYID=2
> etime = Fri Feb 13 12:21:58 2009
> submit_args = -t 1-2 -W depend=afterok:459.fl51u101 doubletest.sh
> job_array_id = 2
> job_array_request = 1-2
> start_time = Fri Feb 13 12:21:58 2009
> & nbsp; start_count = 1
>
> Same syntax with non-job arrays works fine.
>
>
> Thanks for any help,
> Lee A Dreger
> ________________________________
> A Good Credit Score is 700 or Above. See yours in just 2 easy steps!
> _______________________________________________
> torqueusers mailing list
> torqueusers at supercluster.org
> http://www.supercluster.org/mailman/listinfo/torqueusers
>
>
More information about the torqueusers
mailing list