Hello,
Jobs on our galaxy server are not being run. When a new workflow is run, only a single job in the workflow is executed. All later jobs are not run. If the server is restarted, then the 2nd job in the workflow will be executed, but all later jobs will not be run.
I am logging at the debug level and see the following error message repeated many times in the logs
UnboundExecutionError: Instance <Job at 0x5203290> is not bound to a Session; attribute refresh operation cannot proceed galaxy.jobs ERROR 2010-09-09 13:36:03,985 Exception in monitor_step Traceback (most recent call last): File "/labkey/galaxy/galaxy_dist/lib/galaxy/jobs/__init__.py", line 138, in __monitor self.__monitor_step() File "/labkey/galaxy/galaxy_dist/lib/galaxy/jobs/__init__.py", line 212, in __monitor_step log.exception( "failure running job %d" % job.id ) File "/labkey/galaxy/galaxy_dist/eggs/SQLAlchemy-0.5.6_dev_r6498-py2.6.egg/sqlalchemy/orm/attributes.py", line 158, in __get__ return self.impl.get(instance_state(instance), instance_dict(instance)) File "/labkey/galaxy/galaxy_dist/eggs/SQLAlchemy-0.5.6_dev_r6498-py2.6.egg/sqlalchemy/orm/attributes.py", line 377, in get value = callable_() File "/labkey/galaxy/galaxy_dist/eggs/SQLAlchemy-0.5.6_dev_r6498-py2.6.egg/sqlalchemy/orm/state.py", line 185, in __call__ attr.impl.key in unmodified File "/labkey/galaxy/galaxy_dist/eggs/SQLAlchemy-0.5.6_dev_r6498-py2.6.egg/sqlalchemy/orm/mapper.py", line 1863, in _load_scalar_attributes "attribute refresh operation cannot proceed" % (state_str(state)))
I have not been able to successfully completely run a workflow yet on this server (this is a new install)
Specifics about the server are
- Galaxy version: changeset: 4246:f09915c8da94 - Operating System: Ubuntu 10.04 - Python version: 2.6.5 - Database: postgres 8.4
Has anyone seen an error like this before. I have searched the mailing lists and the wiki and could not find a reference.
Thank you in advance for the assistance.
Brian Connolly
Brian and Branden,
What you are experiencing sounds like an issue we encountered a while back. In your instance configurations (universe_wsgi.ini), do you happen to have track_jobs_in_database set to True?
Dannon
Dannon,
This is commented out in the file. Do you know if this is set to True or False by default?
Thank you,
Brian Connolly
On Thu, Sep 9, 2010 at 7:38 PM, Dannon Baker dannonbaker@me.com wrote:
Brian and Branden,
What you are experiencing sounds like an issue we encountered a while back. In your instance configurations (universe_wsgi.ini), do you happen to have track_jobs_in_database set to True?
Dannon
I believe this is a bug we've fixed since the revision you're using, but if you uncomment it and set it to True, it should solve your problems until you update again.
-Dannon
On Sep 10, 2010, at 10:13 AM, Brian Connolly wrote:
Dannon,
This is commented out in the file. Do you know if this is set to True or False by default?
Thank you,
Brian Connolly
On Thu, Sep 9, 2010 at 7:38 PM, Dannon Baker dannonbaker@me.com wrote: Brian and Branden,
What you are experiencing sounds like an issue we encountered a while back. In your instance configurations (universe_wsgi.ini), do you happen to have track_jobs_in_database set to True?
Dannon
That seems to have done it. Thank you for your assistance.
- Brian
On Fri, Sep 10, 2010 at 7:16 AM, Dannon Baker dannonbaker@me.com wrote:
I believe this is a bug we've fixed since the revision you're using, but if you uncomment it and set it to True, it should solve your problems until you update again.
-Dannon
On Sep 10, 2010, at 10:13 AM, Brian Connolly wrote:
Dannon,
This is commented out in the file. Do you know if this is set to True or
False by default?
Thank you,
Brian Connolly
On Thu, Sep 9, 2010 at 7:38 PM, Dannon Baker dannonbaker@me.com wrote: Brian and Branden,
What you are experiencing sounds like an issue we encountered a while
back. In your instance configurations (universe_wsgi.ini), do you happen to have track_jobs_in_database set to True?
Dannon
galaxy-dev@lists.galaxyproject.org
-
Brian Connolly -
Dannon Baker