Question: Problem Reimporting Histories
0
Alessandro Ogier • 10 wrote:
Dear galaxy users,
here at European Institute of Oncology we're deploying a brand
new
galaxy install, but facing problems with history (re)importing.
We have configured a galaxy running on one node where users took a
"sneek peek" on features while developing a real deployment on nfs and
cluster integration. Now I have to migrate histories to new galaxy
install, I'm doing that on behalf my users because we choose to
integrate with an apache proxy that does reverse proxying and ldap
authentication.
Yesterday I produced four tar.gz for a user, placed on a temporary
webserver and then proceeded with reimport on new instance: two where
reimported, two not.
All those files are relatively small (from 1MB to 13MB), and no clues
from galaxy debug log:
SUCCESSFUL ONE:
galaxy.jobs DEBUG 2013-12-18 13:29:32,718 (354) Working directory for
job is: /srv/galaxy/galaxy-dist/database/job_working_directory/000/354
galaxy.jobs.handler DEBUG 2013-12-18 13:29:32,738 (354) Dispatching to
local runner
galaxy.jobs DEBUG 2013-12-18 13:29:32,826 (354) Persisting job
destination (destination id: local)
galaxy.jobs.handler INFO 2013-12-18 13:29:32,864 (354) Job dispatched
galaxy.jobs.runners.local DEBUG 2013-12-18 13:29:33,151 (354)
executing:
export GALAXY_SLOTS="1"; python
/srv/galaxy/galaxy-
dist/lib/galaxy/tools/imp_exp/unpack_tar_gz_archive.py
http://debian.ieo.eu/drosano/dalia-37.tar.gz
/srv/galaxy/galaxy-dist/database/tmp/tmpLAAF5z --url
galaxy.jobs DEBUG 2013-12-18 13:29:33,326 (354) Persisting job
destination (destination id: local)
galaxy.jobs.runners.local DEBUG 2013-12-18 13:29:34,451 execution
finished: export GALAXY_SLOTS="1"; python
/srv/galaxy/galaxy-
dist/lib/galaxy/tools/imp_exp/unpack_tar_gz_archive.py
http://debian.ieo.eu/drosano/dalia-37.tar.gz
/srv/galaxy/galaxy-dist/database/tmp/tmpLAAF5z --url
galaxy.jobs DEBUG 2013-12-18 13:29:34,644 job 354 ended
galaxy.datatypes.metadata DEBUG 2013-12-18 13:29:34,644 Cleaning up
external metadata files
UNSUCCESSFUL ONE:
galaxy.jobs DEBUG 2013-12-18 13:30:42,824 (392) Working directory for
job is: /srv/galaxy/galaxy-dist/database/job_working_directory/000/392
galaxy.jobs.handler DEBUG 2013-12-18 13:30:42,844 (392) Dispatching to
local runner
galaxy.jobs DEBUG 2013-12-18 13:30:42,934 (392) Persisting job
destination (destination id: local)
galaxy.jobs.handler INFO 2013-12-18 13:30:42,963 (392) Job dispatched
galaxy.jobs.runners.local DEBUG 2013-12-18 13:30:43,236 (392)
executing:
export GALAXY_SLOTS="1"; python
/srv/galaxy/galaxy-
dist/lib/galaxy/tools/imp_exp/unpack_tar_gz_archive.py
http://debian.ieo.eu/drosano/dalia-138.tar.gz
/srv/galaxy/galaxy-dist/database/tmp/tmpH8buNE --url
galaxy.jobs DEBUG 2013-12-18 13:30:43,400 (392) Persisting job
destination (destination id: local)
galaxy.jobs.runners.local DEBUG 2013-12-18 13:30:47,524 execution
finished: export GALAXY_SLOTS="1"; python
/srv/galaxy/galaxy-
dist/lib/galaxy/tools/imp_exp/unpack_tar_gz_archive.py
http://debian.ieo.eu/drosano/dalia-138.tar.gz
/srv/galaxy/galaxy-dist/database/tmp/tmpH8buNE --url
galaxy.jobs DEBUG 2013-12-18 13:30:47,711 job 392 ended
galaxy.datatypes.metadata DEBUG 2013-12-18 13:30:47,712 Cleaning up
external metadata files
sorry for line wraps, I also attached logs.
I cloned your repository on my development machine to try to
recreate/debug this issue but unfortunately this time I'm unable to
import successfully any history (plain galaxy instance, sqlite, no ext
user, no job_conf, either default and stable branches).
I'm attaching the new galaxy instance universe_wsgi.ini and
job_conf.xml, as you can see they are pretty straight-forward.
What I'm asking to you : which is the correct way for debugging this
sort of things ? No luck putting some debugging code in
* lib/galaxy/tools/actions/history_imp_exp.py
* lib/galaxy/tools/imp_exp/unpack_tar_gz_archive.py
maybe I'm missing something ...
also, I'd like to contribute some patch on a couple of minor glitches
I
found, but I'll write to -dev list for that
Thank you, and thanks for this great tool :) ciao
--
Alessandro Ogier
Specialista Reti e Sistemi - Area Ricerca
Direzione ICT Gruppo IEO/CCM
Istituto Europeo di Oncologia
Via Serio, 15 - 20139 Milan, Italy
P +39 02 94375269
M +39 334 6841627 (9am-6am CET)
E alessandro.ogier@ieo.eu
W http://www.ieo.it
ADD COMMENT
• link
•
modified 5.0 years ago
by
Jennifer Hillman Jackson ♦ 25k
•
written
5.0 years ago by
Alessandro Ogier • 10
