Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Activity
Martin Cech
@martenson
yeah, should log to
lol I mean tool_shed_webapp.log
galaxybot
@galaxybot
[dave_b] yeah, that's the default location
Martin Cech
@martenson
@davebx any luck with reproducing the revision issue?
Dave B.
@davebx
I just got my toolshed installation running, so I haven't been able to try reproducing until now
Marius van den Beek
@mvdbeek
interesting, can we access that file on jenkins ?
or during local ts tests ?
Martin Cech
@martenson
I don't know that
Dave B.
@davebx
error confirmed, working on a fix now
Martin Cech
@martenson
@mvdbeek I think we use this to setup TS for tests https://github.com/galaxyproject/galaxy/blob/dev/test/base/driver_util.py#L534
Marius van den Beek
@mvdbeek
so we could inject an env var and log to an additional file ?
I'd be curious to know if we can at least spot the error during testing
Martin Cech
@martenson
I think we should have these logs available in workspace after test.
Marius van den Beek
@mvdbeek
Yes, that would be great
Martin Cech
@martenson
(modify things so we have them that is)
Dave B.
@davebx
Marius van den Beek
@mvdbeek
that only makes sense for the galaxy app anyway
Dave B.
@davebx
indeed
since the toolshed app has no jobs, dataset collections, or tags
a quick fix would be to wrap that in a if self.app.name == 'galaxy':
Martin Cech
@martenson
Dave B.
@davebx
confirmed. The if statement eliminates the message and metadata behaves as desired
Marius van den Beek
@mvdbeek
@davebx sounds reasonable to me
@martenson we could set this based on an env var at that location, right ?
Jennifer Hillman-Jackson
@jennaj
@mvdbeek Could you explain how the BAM indexes work now? Always coordinate sorted? What if a tool requires queryname sort?
And I think that is breaking the samtools sort -n tool
Marius van den Beek
@mvdbeek
nothing changed
yes, always coordinate sorted
if a tool needs queryname sorted it needs to do this on it's own
Jennifer Hillman-Jackson
@jennaj
Hum, ok, could you look at dataset 3 in this history - it is green, looks intact, but has messages about the index not being created: https://usegalaxy.org/u/jen/h/test-history-sort-sam-bam
Marius van den Beek
@mvdbeek
So yeah, dataset 3 should be in error, that is probably a tool bug
Jennifer Hillman-Jackson
@jennaj
Ok - that's what it looks like to me, but why would we have a tool that queryname sorts a bam when that will always fail?
Marius van den Beek
@mvdbeek
also samtools sort -n should generate BamNative (for now, or better yet a new queryname-sorted datatype)
Well, that should never have worked
Jennifer Hillman-Jackson
@jennaj
right - those are the changes I'm referring to.
I think we need a ticket for this with all the details. can you help with that since you have ideas about how this should work?
Marius van den Beek
@mvdbeek
I can do that yes
Jennifer Hillman-Jackson
@jennaj
right now all the help I have for users (FAQs, forums) and the help on some tools - or the actual tool function itself, like samtools - doesn't capture this
Marius van den Beek
@mvdbeek
but also note that usegalaxy.org still runs 17.09
so no changes there
Jennifer Hillman-Jackson
@jennaj
will 18.01 handle the datatype differently?
I'll just wait for the ticket and starting tracking/doing the updates for support faqs and help text on tools. not a big deal, just want to be make sure it is all clear and current
Marius van den Beek
@mvdbeek
it will have the new BamNative datatype, that makes no assumptions about sort order, and BAM will be stricter -- if an index can't be generated the dataset will be unusable
so we should update the samtools sort tool to output BamNative
Jennifer Hillman-Jackson
@jennaj
agree - thanks for explaining all of this. I've been following in GH but didn't "get" the big picture until now, thx!
Marius van den Beek
@mvdbeek
sure
Dave B.
@davebx
@mvdbeek @martenson galaxyproject/galaxy#5429
Martin Cech
@martenson
thanks @davebx - could you please retarget that to 18.01?
Jie-lan
@Jie-lan
stick MAF blocks, but MAF no option available. How can I fix it?
MiguelJulia
@MiguelJulia

@natefoo Hi! I tried what you said, removing the first server { } block and adding the lines
listen 8080;
server_name localhost;
to the server with the proxying configuration. After restarting, if I try to execute galaxy, run.sh returns this error:
Starting server in PID 22910.
Traceback (most recent call last):

File "./scripts/paster.py", line 26, in <module>
serve.run()
File "/home/galaxy_user/galaxy/lib/galaxy/util/pastescript/serve.py", line 1055, in run
invoke(command, command_name, options, args[1:])
File "/home/galaxy_user/galaxy/lib/galaxy/util/pastescript/serve.py", line 1061, in invoke
exit_code = runner.run(args)
File "/home/galaxy_user/galaxy/lib/galaxy/util/pastescript/serve.py", line 226, in run
result = self.command()
File "/home/galaxy_user/galaxy/lib/galaxy/util/pastescript/serve.py", line 676, in command
serve()
File "/home/galaxy_user/galaxy/lib/galaxy/util/pastescript/serve.py", line 653, in serve
server(app)
File "/home/galaxy_user/galaxy/lib/galaxy/util/pastescript/loadwsgi.py", line 233, in server_wrapper
context.local_conf)
File "/home/galaxy_user/galaxy/lib/galaxy/util/pastescript/loadwsgi.py", line 91, in fix_call
val = callable(*args,
kw)
File "/home/galaxy_user/galaxy/.venv/lib/python2.7/site-packages/paste/httpserver.py", line 1361, in server_runner
serve(wsgi_app, *kwargs)
File "/home/galaxy_user/galaxy/.venv/lib/python2.7/site-packages/paste/httpserver.py", line 1310, in serve
request_queue_size=request_queue_size)
File "/home/galaxy_user/galaxy/.venv/lib/python2.7/site-packages/paste/httpserver.py", line 1153, in init
request_queue_size=request_queue_size)
File "/home/galaxy_user/galaxy/.venv/lib/python2.7/site-packages/paste/httpserver.py", line 1132, in init
request_queue_size=request_queue_size)
File "/home/galaxy_user/galaxy/.venv/lib/python2.7/site-packages/paste/httpserver.py", line 345, in init
HTTPServer.init(self, server_address, RequestHandlerClass)
File "/usr/local/lib/python2.7/SocketServer.py", line 420, in init
self.server_bind()
File "/usr/local/lib/python2.7/BaseHTTPServer.py", line 108, in server_bind
SocketServer.TCPServer.server_bind(self)
File "/usr/local/lib/python2.7/SocketServer.py", line 434, in server_bind
self.socket.bind(self.server_address)
File "/usr/local/lib/python2.7/socket.py", line 228, in meth
return getattr(self._sock,name)(
args)
socket.error: [Errno 98] Address already in use
galaxy.jobs.handler INFO 2018-02-01 11:45:08,165 sending stop signal to worker thread
galaxy.jobs.handler INFO 2018-02-01 11:45:08,165 job handler queue stopped
galaxy.jobs.runners INFO 2018-02-01 11:45:08,165 LocalRunner: Sending stop signal to 5 worker threads
galaxy.jobs.handler INFO 2018-02-01 11:45:08,166 sending stop signal to worker thread
galaxy.jobs.handler INFO 2018-02-01 11:45:08,166 job handler stop queue stopped

I traced it back to the listen directive, that's the one causing this error to happen. I tried commenting the lines in the galaxy.ini config file refering to he port to serve content and listen, just in case those ones where the ones blocking the address, but the error is still there.

Doing a little more research, I discovered that the files uploaded with nginx are actually saved in database/tmp as tmpXXXXX, but not indexed in the database. When trying to download a file with nginx, I get an empty file in my disk. Does it rings a bell?