DRMMA and SGE problem

16 Jul 2010

      Hi Galaxy-Dev-Team!
First of all, I would like to thank you for providing the community
with such a great framework!

I've downloaded and set up locally the latest packed release. When I
want to use Galaxy with the Sun Grid Engine, I run into problems. It
looks like Galaxy isn't able to fully control the Grid Engine and I
think it might be a version issue with DRMMA and our current SGE. Jobs
get submitted, appear in the queue and are executed, but their status
is not returned to Galaxy and therefore the job's status on the
website remains always "Job is waiting to run".

Here's a more detailed error-message:
------------------------------------------------------
[environment]
CentOS release 5.4 (Final)
SGE 6.2u5
Python 2.5.2
Jun 30th source tar.gz file
http://bitbucket.org/galaxy/galaxy-dist/get/tip.tar.gz
---------------------------------------------------------

[config]
$ cat universe_wsgi.ini |grep sge
# currently available are 'pbs' and 'sge'.
start_job_runners = sge
#default_cluster_job_runner = sge://default/mjobs.q/ cant lunch service
default_cluster_job_runner = sge:///mjobs.q/

$ grep -5 SGE_ROOT eggs.ini

[general]
repository = http://eggs.g2.bx.psu.edu/new
; these eggs must be scrambled for your local environment
no_auto = pbs_python DRMAA_python
SGE_ROOT = /home/geadmin/N1GE python scripts/scramble.py DRMAA_python

[eggs:platform]
bx_python = 0.5.0
Cheetah = 2.2.2
DRMAA_python = 0.2

---------------------------------------------------------

[Error Messages]
galaxy.jobs DEBUG 2010-07-16 17:30:24,794 job 10 dispatched
galaxy.jobs.runners.sge DEBUG 2010-07-16 17:30:27,480 (10) submitting
file /share1/home/icgc/galaxy-dist/database/pbs/galaxy_10.sh
galaxy.jobs.runners.sge DEBUG 2010-07-16 17:30:27,480 (10) command is:
perl /share1/home/icgc/galaxy-dist/tools/myTools/toolExample.pl
/share1/home/icgc/galaxy-dist/database/files/000/dataset_3.dat
/share1/home/icgc/galaxy-dist/database/files/000/dataset_10.dat
galaxy.jobs.runners.sge DEBUG 2010-07-16 17:30:27,522 (10) queued in
mjobs.q queue as 9440764
202.175.149.92 - - [16/Jul/2010:17:30:26 +0900] "POST
/root/history_item_updates HTTP/1.1" 200 -
"http://gw01.hgc.jp:8090/history" "Mozilla/5.0 (X11; U; Linux x86_64;
en-US; rv:1.9.2.6) Gecko/20100628 Ubuntu/10.04 (lucid) Firefox/3.6.6"
galaxy.jobs.runners.sge ERROR 2010-07-16 17:30:29,524 (10/9440764)
Unable to check job status
Traceback (most recent call last):
  File "/share1/home/icgc/galaxy-dist/lib/galaxy/jobs/runners/sge.py",
line 277, in check_watched_items
    state = self.ds.getJobProgramStatus( job_id )
  File "/usr/local/lib/python2.5/site-packages/DRMAA.py", line 395, in
getJobProgramStatus
    return ps
TypeError: writelines() argument must be a sequence of strings
galaxy.jobs.runners.sge WARNING 2010-07-16 17:30:29,525 (10/9440764)
job will now be errored
galaxy.jobs.runners.sge ERROR 2010-07-16 17:30:29,526 Uncaught
exception failing job
Traceback (most recent call last):
  File "/share1/home/icgc/galaxy-dist/lib/galaxy/jobs/runners/sge.py",
line 138, in run_next
    self.fail_job( obj )
  File "/share1/home/icgc/galaxy-dist/lib/galaxy/jobs/runners/sge.py",
line 338, in fail_job
    self.stop_job( self.sa_session.query( self.app.model.Job ).get(
sge_job_state.job_wrapper.job_id ) )
AttributeError: 'SGEJobRunner' object has no attribute 'sa_session'

[when I use sample code]

#!/usr/bin/env python

import DRMAA
import time
import os

def main():
    """Submit a job, and check its progress.
    Note, need file called sleeper.sh in home directory. An example:
    echo 'Hello World $1'
    sleep 30s
    """
    s=DRMAA.Session()
    s.init()

    print 'Creating job template'
    jt = s.createJobTemplate()
    jt.remoteCommand = os.getcwd() + '/sleeper.sh'
    jt.args = ['42','Simon says:']
    jt.joinFiles=True
    jt.outputPath=":"+DRMAA.JobTemplate.HOME_DIRECTORY+'/tmp/JOB_OUT'

    jobid = s.runJob(jt)
    print 'Your job has been submitted with id ' + jobid

    # Who needs a case statement when you have dictionaries?
    decodestatus = {
        DRMAA.Session.UNDETERMINED: 'process status cannot be determined',
        DRMAA.Session.QUEUED_ACTIVE: 'job is queued and active',
        DRMAA.Session.SYSTEM_ON_HOLD: 'job is queued and in system hold',
        DRMAA.Session.USER_ON_HOLD: 'job is queued and in user hold',
        DRMAA.Session.USER_SYSTEM_ON_HOLD: 'job is queued and in user
and system hold',
        DRMAA.Session.RUNNING: 'job is running',
        DRMAA.Session.SYSTEM_SUSPENDED: 'job is system suspended',
        DRMAA.Session.USER_SUSPENDED: 'job is user suspended',
        DRMAA.Session.DONE: 'job finished normally',
        DRMAA.Session.FAILED: 'job finished, but failed',
        }

    for ix in range(10):
        print ix
        print 'Checking ' + str(ix) + ' of 10 times'
        status = s.getJobProgramStatus(jobid)
        print decodestatus.get(status)
        time.sleep(5)

    print 'Cleaning up'
    s.deleteJobTemplate(jt)
    s.exit()

if __name__=='__main__':
    main()

$ ./test
Creating job template
Your job has been submitted with id 9440771
0
Checking 0 of 10 times
Traceback (most recent call last):
  File "./test", line 52, in <module>
    main()
  File "./test", line 43, in main
    status = s.getJobProgramStatus(jobid)
  File "/usr/local/lib/python2.5/site-packages/DRMAA.py", line 393, in
getJobProgramStatus
    eno, ps, estr = cDRMAA.drmaa_job_ps(jobName)
ValueError: need more than 2 values to unpack

[debug]
        print cDRMAA.drmaa_job_ps(jobName)
        eno, ps, estr = cDRMAA.drmaa_job_ps(jobName)
=> ([0, 16], '')

Thank you for taking time! Hopefully, you can give me some hints on
how to resolve this issue.

Cheers,

George Chalkidis

George Chalkidis

Nate Coraor

George Chalkidis

tags

participants (2)