Thanks for the bug report, that is indeed a problem - I am not sure how the torque runner here would ever reach a complete job state. I prefer the following fix however, if you would be willing to verify it fixes the problem you are encountering I would be happy to apply it to Galaxy's stable branch.
diff -r 6a633b37c4c4 lib/galaxy/jobs/runners/cli_job/torque.py --- a/lib/galaxy/jobs/runners/cli_job/torque.py Fri Feb 28 08:52:52 2014 -0500 +++ b/lib/galaxy/jobs/runners/cli_job/torque.py Fri Feb 28 08:32:41 2014 -0600 @@ -131,4 +131,5 @@ def __get_job_state(self, state): return { 'E' : job_states.RUNNING, 'R' : job_states.RUNNING, - 'Q' : job_states.QUEUED }.get(state, state) + 'Q' : job_states.QUEUED, + 'C' : job_states.OK }.get(state, state)
-John
On Fri, Feb 28, 2014 at 12:13 AM, ngsflow@hygenomics.com ngsflow@hygenomics.com wrote:
Hi,
base: os:centos6.4 torque:4.2.7
I built a glaxy cluster today,and then I found the job completion status of torque queue is 'c',but the need state of galalxy is "ok".So I modified the torque.py file and solved this error. I do, right?
diff -r 8411a9f30feb lib/galaxy/jobs/runners/cli_job/torque.py --- a/lib/galaxy/jobs/runners/cli_job/torque.py Thu Dec 19 14:38:03 2013 -0500 +++ b/lib/galaxy/jobs/runners/cli_job/torque.py Fri Feb 28 13:02:48 2014 +0800 @@ -117,6 +117,8 @@ if id in job_ids: state = job.find('job_state').text # map PBS job states to Galaxy job states.
if state == u'C' :state = 'ok' rval[id] = self.__get_job_state(state) return rval
ngsflow@hygenomics.com
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: http://lists.bx.psu.edu/
To search Galaxy mailing lists use the unified search at: http://galaxyproject.org/search/mailinglists/