Bad UID exception with pbs_python on galaxy 16_10 and torque 6.02

classic Classic list List threaded Threaded
1 message Options
| Threaded
Open this post in threaded view
|

Bad UID exception with pbs_python on galaxy 16_10 and torque 6.02

Vandeweyer Geert
Hi all,

We've been running a galaxy server for the last few years successfully,
using pbs_python for connection to our HPC.
Recently, we've purchased a new system, and we're trying to install
galaxy on it. However, we are having some issues.

When we try to submit jobs, galaxy returns the following error:

galaxy.jobs.runners.pbs WARNING 2017-02-15 16:55:22,178 (20) pbs_submit
failed (try 1/5), PBS error 15025: Bad UID for job execution

On the pbs_server, we see:

02/16/2017 09:14:43.631;128;PBS_Server.7156;Req;req_reject;Reject reply
code=15025(Bad UID for job execution MSG=ruserok failed validating
galaxyuser/galaxyuser from galaxy.uza.be), aux=0, type=QueueJob, from
[hidden email]


However, when we submit the generated qsub script manually, using the
same user, queue and pbs_account (-A) settings, the job runs just fine.
In our opinion this indicats an issue with galaxy somewhere. Running
jobs locally also works.


The docs state that we need to scramble a specific pbs_python version
for our recent pbs version, but the mentioned files are missing in
galaxy 16_10.


Some information:
=============

pbs_server version : 6.02

galaxy: v16_10

deprecated procedures:
https://wiki.galaxyproject.org/Admin/Config/Performance/Cluster (in PBS
: egss.ini and scripts/scramble.py are missing)


Has anybody seen this before? Are there newer docs available on
pbs_python ?


Best,

Geert
___________________________________________________________
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:
  https://lists.galaxyproject.org/

To search Galaxy mailing lists use the unified search at:
  http://galaxyproject.org/search/mailinglists/