I've been trying to get Galaxy to submit jobs to my cluster as the
"real user" rather than the "galaxy" user.

In my galaxy.ini, I have:

outputs_to_working_directory = True
set_metadata_externally = True
drmaa_external_runjob_script = /galaxy/scripts/drmaa_external_runner.py
drmaa_external_killjob_script = /galaxy/scripts/drmaa_external_killer.py
drmaa_external_chown_script = /galaxy/scripts/external_chown_script.py

In my sudoers file I have:

galaxy ALL= (root) NOPASSWD: SETENV: /galaxy/scripts/drmaa_external_runner.py
galaxy ALL= (root) NOPASSWD: SETENV: /galaxy/scripts/drmaa_external_killer.py
galaxy ALL= (root) NOPASSWD: SETENV: /galaxy/scripts/external_chown_script.py

In my log file, I'm seeing:

galaxy.jobs.runners.drmaa DEBUG 2014-10-29 14:45:54,777 (15)
submitting file
galaxy.jobs.runners.drmaa DEBUG 2014-10-29 14:45:54,783 (15)
submitting with credentials: tjg [uid: 28369]
galaxy.jobs.runners ERROR 2014-10-29 14:45:54,784 (15) Unhandled
exception calling queue_job
Traceback (most recent call last):
  File "/galaxy/lib/galaxy/jobs/runners/__init__.py", line 96, in run_next
  File "/galaxy/lib/galaxy/jobs/runners/drmaa.py", line 189, in queue_job
    filename = self.store_jobtemplate(job_wrapper, jt)
  File "/galaxy/lib/galaxy/jobs/runners/drmaa.py", line 348, in
    f = open(filename,'w')
IOError: [Errno 2] No such file or directory: '/galaxy/database/pbs/15.jt_json'

My /var/log/messages file is not reporting any attempt to run "sudo",
so I think the process is failing before it gets to that point in the
Galaxy code.

If I comment out the "drmaa_external_*" lines, the jobs are submitted
successfully and run as the "galaxy" user.

What am I missing here?  :)


Tim Gustafson
Baskin Engineering, Room 313A
Please keep all replies on the list by using "reply all"
in your mail client.  To manage your subscriptions to this
and other Galaxy lists, please use the interface at:

To search Galaxy mailing lists use the unified search at:

Reply via email to