[Iplant-api-dev] GeneSeqer-5.0 on TACC

[Rion Dooley]

I believe I’ve tracked down and fixed a couple bugs that turned up after the recent changes. The first had to do with correctly identifying user permissions if different situations. The service was being a little draconian with its interpretation of inheritance. I believe I have that bug closed out. The next had to do with guarding against wrapper scripts doing odd things. I noticed in the output logs of a few jobs that the script had somehow left the working directory. This preventing the app from finding the archive log to know which files to archive after execution. I added some code to make sure the job stays in its working tree during execution so archiving takes place as expected.

Both these changes were made to Foundation. Agave handles data staging and archiving separately from the job execution, so these issues don’t tend to creep up there. Please let me know if you see any issues now that this patch is out.

--
Rion




On Dec 6, 2013, at 4:39 PM, Barthelson, Roger A - (rogerab) <rogerab at email.arizona.edu<mailto:rogerab at email.arizona.edu>> wrote:

Several problems have shown up today with API jobs on Stampede — including the one described below. Also, I submitted a job for GeneSeqer in the DE well over an hour ago, and it is still listed as idle.  It is not in the queue on Stampede.

I also submitted a job in the DE for GMAP yesterday, and it looked like it was running on Stampede's queue yesterday afternoon. But now it is still listed in the DE as running, but it is well over the time limit, and no GMAP job is showing in the queue.

Roger

From: <Duvick>, "Jonathan P [GDCBS]" <jduvick at iastate.edu<mailto:jduvick at iastate.edu>>
Date: Friday, December 6, 2013 12:44 PM
To: Roger Barthelson <rogerab at email.arizona.edu<mailto:rogerab at email.arizona.edu>>, Nirav Merchant <nirav at email.arizona.edu<mailto:nirav at email.arizona.edu>>
Cc: Volker Brendel <vbrendel at indiana.edu<mailto:vbrendel at indiana.edu>>
Subject: GeneSeqer-5.0 on TACC

Hi Nirav and Roger,

GeneSeqer-5.0u1 is failing to return results as of today -- the Foundation jobs API reports no error, and ends with status 'archiving_finished', but it fails to deposit an output file in the archive.  (It does successfully create a job archive directory).

The same test datasets completed just fine yesterday, so this is a new problem. I've cleaned out my jobs directory so I can't attribute it to an out of space issue. And I've done enough reps to convince me there is a systemic problem not just a happenstance glitch (whatever that is).

On the other hand, the gth-lonestar app works just fine, test datasets return the correct output.

Roger, if you have a moment could you run a quick test and see if you can confirm the problem or not?

Thanks,

Jon Duvick
PlantGDB Manager
http://www.plantgdb.org/
Department of Genetics, Development and Cell Biology
2258 Molecular Biology Building
Iowa State University
Ames IA 50011

(515) 294-2360
(515) 294-6755 FAX
_______________________________________________
Iplant-api-dev Mailing List: Iplant-api-dev at iplantcollaborative.org<mailto:Iplant-api-dev at iplantcollaborative.org>
List Info and Archives: http://mail.iplantcollaborative.org/mailman/listinfo/iplant-api-dev
One-click Unsubscribe: http://mail.iplantcollaborative.org/mailman/options/iplant-api-dev/dooley%40tacc.utexas.edu?unsub=1&unsubconfirm=1

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.iplantcollaborative.org/pipermail/iplant-api-dev/attachments/20131207/73b2a61b/attachment.html