Ok, I added

track_jobs_in_database = True

 

to universe_wsgi.ini and now workflows appear to go to completion.

 

Another problem I was having was that steps in a workflow would stay gray even though examination of the log revealed the steps had completed and produced correct output (as verified by directly opening the dat files).  This problem also seems to have been fixed by the above.

 

Mark

 

 

Hi All

 

On my local install of galaxy I’m having  problems executing workflows.  I will upload data into a new history and then select to run a saved workflow.  This initiates fine but, after the completion of a variable number of steps, the workflow just stops after the successful completion of a step.  No errors appear in the web interface and all I can see in the paster.log is something like

 

 

90 HTTP/1.1" 200 - "http://127.0.0.1:8081/history" "Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:18.0) Gecko/20100101 Firefox/18.0"

127.0.0.1 - - [28/Jan/2013:12:29:07 -0400] "GET /api/histories/82b264d8c3d11790 HTTP/1.1" 200 - "http://127.0.0.1:8081/history" "Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:18.0) Gecko/20100101 Firefox/18.0"

galaxy.jobs.runners.local DEBUG 2013-01-28 12:29:08,621 execution finished: python /home/rosema1/Projects/galaxy-dist/tools/fastq/fastq_trimmer_by_quality.py '/home/rosema1/Projects/galaxy-dist/database/files/000/dataset_590.dat' '/home/rosema1/Projects/galaxy-dist/database/files/000/dataset_592.dat' -f 'sanger' -s '5'      -t '1' -e '53' -a 'mean' -x '0' -c '>=' -q '20.0'

galaxy.jobs DEBUG 2013-01-28 12:29:08,838 Tool did not define exit code or stdio handling; checking stderr for success

galaxy.tools DEBUG 2013-01-28 12:29:09,135 Error opening galaxy.json file: [Errno 2] No such file or directory: '/home/rosema1/Projects/galaxy-dist/database/job_working_directory/000/418/galaxy.json'

galaxy.jobs DEBUG 2013-01-28 12:29:09,439 job 418 ended

galaxy.jobs DEBUG 2013-01-28 12:29:10,041 (420) Working directory for job is: /home/rosema1/Projects/galaxy-dist/database/job_working_directory/000/420

galaxy.jobs.handler DEBUG 2013-01-28 12:29:10,042 dispatching job 420 to local runner

galaxy.jobs.handler INFO 2013-01-28 12:29:10,200 (420) Job dispatched

galaxy.jobs.runners.local DEBUG 2013-01-28 12:29:10,385 Local runner: starting job 420

galaxy.jobs.runners.local DEBUG 2013-01-28 12:29:10,759 executing: python /home/rosema1/Projects/galaxy-dist/tools/fastq/fastq_filter.py /home/rosema1/Projects/galaxy-dist/database/files/000/dataset_592.dat /home/rosema1/Projects/galaxy-dist/database/job_working_directory/000/420/tmph0Ll2V /home/rosema1/Projects/galaxy-dist/database/files/000/dataset_594.dat /home/rosema1/Projects/galaxy-dist/database/job_working_directory/000/420/dataset_594_files 'sanger'

127.0.0.1 - - [28/Jan/2013:12:29:12 -0400] "GET /api/histories/82b264d8c3d11790 HTTP/1.1" 200 - "http://127.0.0.1:8081/history" "Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:18.0) Gecko/20100101 Firefox/18.0"

127.0.0.1 - - [28/Jan/2013:12:29:12 -0400] "GET /api/users/f2db41e1fa331b3e HTTP/1.1" 200 - "http://127.0.0.1:8081/history" "Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:18.0) Gecko/20100101 Firefox/18.0"

127.0.0.1 - - [28/Jan/2013:12:29:12 -0400] "GET /api/histories/82b264d8c3d11790/contents?ids=3da459a26c8f00b4%2C736af8f0a76a2e71 HTTP/1.1" 200 - "http://127.0.0.1:8081/history" "Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:18.0) Gecko/20100101 Firefox/18.0"

 

127.0.0.1 - - [28/Jan/2013:12:29:16 -0400] "GET /api/histories/82b264d8c3d11790 HTTP/1.1" 200 - "http://127.0.0.1:8081/history" "Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:18.0) Gecko/20100101 Firefox/18.0"

 

Though I seem to see this even when a step completes and the workflow proceeds.

 

Moreover, if I then begin executing the next steps in the workflow, they work, but create new entries in the history.

 

So, some questions:

 

Why is this happening and what can I do to get workflows to run to completion?

 

Also,  is it possible to reinitiate the workflow from the premature stopping point without having to launch each step individually and point each at new inputs rather than the ones dictated in the workflow?

 

Thanks

 

Mark

 

 

 

This message may contain confidential information. If you are not the designated recipient, please notify the sender immediately, and delete the original and any copies. Any use of the message by you is prohibited.