[galaxy-dev] Galaxy administrators - admin_users setting

2011-05-09 Thread Peter Cock
Hi all,

To make someone an administrator on a local Galaxy install, do I just
need to add their email (login) to the comma separated setting
admin_users in universe_wsgi.ini?
https://bitbucket.org/galaxy/galaxy-central/wiki/Admin/AdminInterface

I have this working on one server, but it doesn't seem to have any
effect on a second server. Both are now running the current release
(changeset 50e249442c5a).

Is there some other setting needed to enable the admin interface?

Thanks,

Peter
___
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:

  http://lists.bx.psu.edu/


Re: [galaxy-dev] Galaxy administrators - admin_users setting

2011-05-09 Thread Peter Cock
On Mon, May 9, 2011 at 12:48 PM, Hans-Rudolf Hotz h...@fmi.ch wrote:
 Hi Peter

 First,the obvious question: Have you restarted galaxy?

Yes, and I tried hard reloads in two different browsers (both of which
are working on the primary server).

 Second: I vaguely remember a similar problem, and if I remember correctly,
 the problem we had was the space between the comma and the next e-mail
 address. Hence, make sure you write the line like this:

 admin_users = f...@bar.com,f...@bar.com,f...@bar.com

 (ie: without any blank spaces)

 I hope this helps, Hans

I'll double check that.

As an aside, it would be interesting to check if the comparison is done
case insensitively or not (emails and domain names are case insensitive).

Fingers crossed... thanks,

Peter
___
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:

  http://lists.bx.psu.edu/


[galaxy-dev] External representation of workflow

2011-05-09 Thread Frederick van Staden
Hello everyone,

I need some advice on how to represent the output of all the different tools
inside a workflow in an external application.
I am looking into using galaxy as a digital evidence processing tool. It is
very important that i can access the intermediate sets of data created by
every tool's output.
Some have suggested the actual tools should populate a database and that the
secondary(viewer) application should just retrieve the applicable data form
there.

Any suggestions?
Any input would be greatly appreciated.

Regards
Fred
___
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:

  http://lists.bx.psu.edu/

Re: [galaxy-dev] [galaxy-user] Megblast GIs

2011-05-09 Thread Nate Coraor
Peter Cock wrote:
 
 Which version of BLAST do you have?
 
 If you're using the megablast tool in Galaxy, then it will be NCBI
 legacy BLAST using the blastall binary.
 
 If you're using the (commented out by default) BLAST+ wrappers, then
 it will be using the blastn binary. There were changes to the ID
 handling in the tabular output in BLAST 2.2.25+ as I recall.

Peter,

Minor point unrelated to Douglas' original question - the BLAST+ tools
are no longer commented in the sample tool config.

--nate

 
 Another important factor is how the BLAST database was made (with or
 without the -parse_seqids flag).
 
 Peter
 
 P.S. Questions like this about local installs might be better on the
 galaxy-dev list, CC'd.
 
 ___
 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:
 
   http://lists.bx.psu.edu/
 
___
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:

  http://lists.bx.psu.edu/


Re: [galaxy-dev] Unable to set metadata in API call

2011-05-09 Thread Nate Coraor
Duddy, John wrote:
 I need to be able to set some metadata in some custom data types. For now, 
 I'm just trying to set the value of the 'misc_info' field. The client script 
 is this:
 put( sys.argv[1], sys.argv[2], { 'update_type' : 'metadata', 'misc_info' : 
 'meta data msg' } )
 and my API method is as follows. It executes fine, but the values do not show 
 on the next show() call.

Hi John,

Your method below works fine for existing metadata items like
'data_lines', so I suspect the problem is not with the update method,
but with the datatype definition.  Does your class have a 'misc_info'
MetadataElement?

--nate

 Disclaimer: Python is still quite new to me, and it's very likely I do not 
 understand the metadata model and the way library dataset associates work.
 
 Any clues will be greatly appreaciated!
 @web.expose_api
 def update( self, trans, id,  library_id, payload, **kwd ):
 
 PUT 
 /api/libraries/{encoded_library_id}/contents/{encoded_content_type_and_id}
 Sets attributes (metadata) on a library item.
 
 update_type = None
 if 'update_type' not in payload:
 trans.response.status = 400
 return Missing required 'update_type' parameter.  Please consult 
 the API documentation for help.
 else:
 update_type = payload.pop( 'update_type' )
 if update_type not in ( 'metadata' ):
 trans.response.status = 400
 return Invalid value for 'update_type' parameter ( %s ) 
 specified.  Please consult the API documentation for help. % update_type
 content_id = id
 decoded_type_and_id = trans.security.decode_string_id( content_id )
 content_type, decoded_content_id = decoded_type_and_id.split( '.' )
 if content_type not in ( 'file' ):
 trans.response.status = 400
 return Updates allowed only on files, not directories
 try:
 content = trans.sa_session.query( 
 trans.app.model.LibraryDatasetDatasetAssociation ).get( decoded_content_id )
 except:
 content = None
 if not content or ( not trans.user_is_admin() and not 
 trans.app.security_agent.can_modify_library_item( 
 trans.get_current_user_roles(), content, trans.user ) ):
 trans.response.status = 400
 return Invalid %s id ( %s ) specified. % ( content_type, str( 
 content_id ) )
 
 metadata = content.get_metadata()
 content.datatype.before_setting_metadata(content)
 
 if not metadata:
metadata = {}
 
 for name in payload:
 if name not in [ 'name', 'info', 'dbkey' ]:
 setattr( metadata, name, payload[name])
 
 content.set_metadata(metadata)
 content.datatype.after_setting_metadata( content )
 trans.sa_session.flush()
 
 return OK
 
 
 John Duddy
 Sr. Staff Software Engineer
 Illumina, Inc.
 9885 Towne Centre Drive
 San Diego, CA 92121
 Tel: 858-736-3584
 E-mail: jdu...@illumina.commailto:jdu...@illumina.com
 

 ___
 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:
 
   http://lists.bx.psu.edu/

___
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:

  http://lists.bx.psu.edu/


[galaxy-dev] shared TEMP directory location

2011-05-09 Thread Shantanu Pavgi

I am configuring galaxy to use a shared filesystem location as it's TEMP 
directory. I have modified $TEMP variable with the necessary path. Are there 
any other changes that I need to make? The cluster config  page 
https://bitbucket.org/galaxy/galaxy-central/wiki/Config/Cluster mentions about 
removing or commenting out tool runner configuration, example below: 

{{{
[galaxy:tool_runners]

biomart = local:///
encode_db1 = local:///
hbvar = local:///
microbial_import1 = local:///
ucsc_table_direct1 = local:///
ucsc_table_direct_archaea1 = local:///
ucsc_table_direct_test1 = local:///
upload1 = local:///
}}}

Should we replace it with some other value or completely remove it? Also, what 
should it be set to for any additional tools that we install?

--
Thanks,
Shantanu.


___
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:

  http://lists.bx.psu.edu/


Re: [galaxy-dev] Unable to set metadata in API call

2011-05-09 Thread Duddy, John
Silly me - 'misc_info' shows up in the show() output, but I didn't bother to 
verify that it was metadata and not some other member.


Thanks!

John Duddy
Sr. Staff Software Engineer
Illumina, Inc.
9885 Towne Centre Drive
San Diego, CA 92121
Tel: 858-736-3584
E-mail: jdu...@illumina.com

-Original Message-
From: Nate Coraor [mailto:n...@bx.psu.edu] 
Sent: Monday, May 09, 2011 8:27 AM
To: Duddy, John
Cc: galaxy-dev@lists.bx.psu.edu
Subject: Re: [galaxy-dev] Unable to set metadata in API call

Duddy, John wrote:
 I need to be able to set some metadata in some custom data types. For now, 
 I'm just trying to set the value of the 'misc_info' field. The client script 
 is this:
 put( sys.argv[1], sys.argv[2], { 'update_type' : 'metadata', 'misc_info' : 
 'meta data msg' } )
 and my API method is as follows. It executes fine, but the values do not show 
 on the next show() call.

Hi John,

Your method below works fine for existing metadata items like
'data_lines', so I suspect the problem is not with the update method,
but with the datatype definition.  Does your class have a 'misc_info'
MetadataElement?

--nate

 Disclaimer: Python is still quite new to me, and it's very likely I do not 
 understand the metadata model and the way library dataset associates work.
 
 Any clues will be greatly appreaciated!
 @web.expose_api
 def update( self, trans, id,  library_id, payload, **kwd ):
 
 PUT 
 /api/libraries/{encoded_library_id}/contents/{encoded_content_type_and_id}
 Sets attributes (metadata) on a library item.
 
 update_type = None
 if 'update_type' not in payload:
 trans.response.status = 400
 return Missing required 'update_type' parameter.  Please consult 
 the API documentation for help.
 else:
 update_type = payload.pop( 'update_type' )
 if update_type not in ( 'metadata' ):
 trans.response.status = 400
 return Invalid value for 'update_type' parameter ( %s ) 
 specified.  Please consult the API documentation for help. % update_type
 content_id = id
 decoded_type_and_id = trans.security.decode_string_id( content_id )
 content_type, decoded_content_id = decoded_type_and_id.split( '.' )
 if content_type not in ( 'file' ):
 trans.response.status = 400
 return Updates allowed only on files, not directories
 try:
 content = trans.sa_session.query( 
 trans.app.model.LibraryDatasetDatasetAssociation ).get( decoded_content_id )
 except:
 content = None
 if not content or ( not trans.user_is_admin() and not 
 trans.app.security_agent.can_modify_library_item( 
 trans.get_current_user_roles(), content, trans.user ) ):
 trans.response.status = 400
 return Invalid %s id ( %s ) specified. % ( content_type, str( 
 content_id ) )
 
 metadata = content.get_metadata()
 content.datatype.before_setting_metadata(content)
 
 if not metadata:
metadata = {}
 
 for name in payload:
 if name not in [ 'name', 'info', 'dbkey' ]:
 setattr( metadata, name, payload[name])
 
 content.set_metadata(metadata)
 content.datatype.after_setting_metadata( content )
 trans.sa_session.flush()
 
 return OK
 
 
 John Duddy
 Sr. Staff Software Engineer
 Illumina, Inc.
 9885 Towne Centre Drive
 San Diego, CA 92121
 Tel: 858-736-3584
 E-mail: jdu...@illumina.commailto:jdu...@illumina.com
 

 ___
 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:
 
   http://lists.bx.psu.edu/


___
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:

  http://lists.bx.psu.edu/


[galaxy-dev] Can I run workflow on certain nodes of the cluster?

2011-05-09 Thread Liisa Koski
Is it possible to run a specific workflow only on certain nodes of the 
cluster? Either using the API or by setting something in the config files?

Thanks,
Liisa
 ___
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:

  http://lists.bx.psu.edu/

Re: [galaxy-dev] Dynamically Specifying PBS Resources

2011-05-09 Thread Nate Coraor
Allen, Benjamin S wrote:
 I'm wondering if its possible to somehow present the user with a param that 
 lets him or her specify number of cores, nodes, memory, and wall time from 
 within the workflow for a specific tool.
 
 Other than coming up with a predictive algorithm for each tool's resource 
 needs (memory especially) how is one supposed to efficiently use Galaxy and 
 PBS other than somehow letting the user specify needed resources? Think in 
 cases like assembly, for example with Velvet depending on the input files, 
 resource requirements can be vastly different job to job.

Hi Ben,

There is not a dynamic method due to the difficulties of predetermining
resource requirements based on inputs and parameters (I have never seen
this done reliably).  A manual method could be feasible for controlled
environments, so I've created a feature request for it here:

  
https://bitbucket.org/galaxy/galaxy-central/issue/532/allow-users-to-specify-drm-resource

--nate

 
 Thanks,
 
 Ben
 ___
 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:
 
   http://lists.bx.psu.edu/
___
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:

  http://lists.bx.psu.edu/


Re: [galaxy-dev] Display BAM with IGV

2011-05-09 Thread Nate Coraor
Dave Walton wrote:
 Could someone who has successfully gotten the IGV tool to work using NGINX as 
 your proxy, tell me if there was anything specific they needed to do with 
 their NGINX or galaxy config to get it working?
Hi Dave,


It's necessary to set up the more advanced configuration under Sending
files using nginx on the nginxProxy page in the wiki.  Could you let us
know whether you're using that configuration or just a basic
pass-through proxy?

Thanks,
--nate

 
 Thanks,
 
 Dave
 
 
 On 4/27/11 9:07 PM, Dave Walton dave.wal...@jax.org wrote:
 
 I’m getting this same error and I’m running nginx as a proxy server.  Is 
 there a specfic parameter we should be aware of to allow partial gets?  I 
 don’t see anything in the wiki page for setting up the ProductionServer.
 
 Thanks,
 
 Dave
 
 
 On 4/27/11 12:33 PM, Daniel Blankenberg d...@bx.psu.edu wrote:
 
 Hi Alex,
 
 Displaying BAM files at external applications usually requires an http server 
 that supports partial gets. Unfortunately, the current lightweight http 
 server that is built into Galaxy does not support these advanced http 
 options.  However, this limitation can be overcome by using a proxy server 
 that is able to handle http1.1 requests. Please see Using a Proxy server at 
 https://bitbucket.org/galaxy/galaxy-central/wiki/Config/ProductionServer for 
 help on setting this up.
 
 Thanks for using Galaxy,
 
 Dan
 
 
 On Apr 27, 2011, at 10:06 AM, Alexander Graf wrote:
 
 Hello all,
 I just updated our local copy of Galaxy and just wanted to try the new IGV 
 display application.
 When I try to display a BAM file with the local IGV or the web IGV I 
 encounter following error complaining about the Null content-length.
 Has any of you encountered the same problem? Do I miss something?
 Thanks in advance
 
 Alex
 
 
 INFO [2011-04-27 16:00:14,440]  [TrackManager.java:225] [Thread-6]  Loading1 
 resources.
 ERROR [2011-04-27 16:00:14,589]  [SeekableHTTPStream.java:59] [Thread-6]  
 Null content-length for: 
 http://localhost:8080/display_application/a1f8a55a946ee078/igv_bam/local/454edf2520b73a7e/data/galaxy_a1f8a55a946ee078.bam
 ERROR [2011-04-27 16:00:14,590]  [TrackLoader.java:821] [Thread-6]  Error 
 loading sam track
 java.lang.RuntimeException: Error loading file: 
 http://localhost:8080/display_application/a1f8a55a946ee078/igv_bam/local/454edf2520b73a7e/data/galaxy_a1f8a55a946ee078.bam
  NULL content-length
 at 
 org.broad.igv.util.SeekableHTTPStream.init(SeekableHTTPStream.java:60)
 at 
 org.broad.igv.util.SeekableStreamFactory.getStreamFor(SeekableStreamFactory.java:42)
 at 
 org.broad.igv.sam.reader.BAMHttpQueryReader.loadHeader(BAMHttpQueryReader.java:129)
 at 
 org.broad.igv.sam.reader.BAMHttpQueryReader.getHeader(BAMHttpQueryReader.java:70)
 at 
 org.broad.igv.sam.reader.BAMHttpQueryReader.getSequenceNames(BAMHttpQueryReader.java:80)
 at 
 org.broad.igv.sam.reader.CachingQueryReader.getSequenceNames(CachingQueryReader.java:97)
 at 
 org.broad.igv.sam.AlignmentDataManager.initChrMap(AlignmentDataManager.java:75)
 at 
 org.broad.igv.sam.AlignmentDataManager.init(AlignmentDataManager.java:69)
 at 
 org.broad.igv.sam.AlignmentDataManager.getDataManager(AlignmentDataManager.java:54)
 at 
 org.broad.igv.track.TrackLoader.loadAlignmentsTrack(TrackLoader.java:761)
 at org.broad.igv.track.TrackLoader.load(TrackLoader.java:143)
 at org.broad.igv.track.TrackManager.load(TrackManager.java:271)
 at org.broad.igv.track.TrackManager.loadResources(TrackManager.java:241)
 at org.broad.igv.main.CommandExecutor.loadFiles(CommandExecutor.java:282)
 at org.broad.igv.main.CommandExecutor.hget(CommandExecutor.java:148)
 at org.broad.igv.main.CommandExecutor.execute(CommandExecutor.java:88)
 at org.broad.igv.main.CommandListener.processGet(CommandListener.java:193)
 at org.broad.igv.main.CommandListener.run(CommandListener.java:93)
 at java.lang.Thread.run(Thread.java:636)
 ERROR [2011-04-27 16:00:14,592]  [TrackLoader.java:211] [Thread-6]  
 java.lang.RuntimeException: java.lang.RuntimeException: Error loading file: 
 http://localhost:8080/display_application/a1f8a55a946ee078/igv_bam/local/454edf2520b73a7e/data/galaxy_a1f8a55a946ee078.bam
  NULL content-length
 
 
 
 ___
 
   Alexander Graf
 
 Laboratory of Functional Genome Analysis
 Gene Center, University of Munich (LMU)
 Feodor-Lynen-Strasse 25; 81377 Munich; Germany
 email: g...@lmb.uni-muenchen.de
 Internet: www.lafuga.de/genomics.htm http://www.lafuga.de/genomics.htm
  ___
 
 ___
 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:
 
   http://lists.bx.psu.edu/
 
 
 

 

Re: [galaxy-dev] Can I run workflow on certain nodes of the cluster?

2011-05-09 Thread Nate Coraor
Liisa Koski wrote:
 Is it possible to run a specific workflow only on certain nodes of the 
 cluster? Either using the API or by setting something in the config files?

Hi Liisa,

Galaxy currently only allows specification of these sorts of parameters
on a per-tool basis, so it's not possible to do it for an entire
workflow.  I agree this would be a handy feature, but it's unlikely we'd
be able to work on it any time soon.

--nate

 
 Thanks,
 Liisa
  

 ___
 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:
 
   http://lists.bx.psu.edu/

___
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:

  http://lists.bx.psu.edu/


Re: [galaxy-dev] problems previewing certain files, downloading files and with login

2011-05-09 Thread Nate Coraor
Matthew Conte wrote:
 Good afternoon,
 
 I'm having several problems with my local galaxy install which may or may
 not be related to one another.
 
 The first problem I'm having is with the Display data in browser feature.
  Some files type formats (fastqsanger, fasta) display fine.  However, other
 file type formats (png, svg, txt) simply do not display at all. I have
 debugging turned on, but do not see any error within the browser or within
 the galaxy log or web server log after clicking the Display data in
 browser eyeball icon.  It is simply a blank middle panel.
 
 The next problem I'm having has to do with downloading files.  Firefox and
 Safari will download any file from a history, but will just create an empty
 file on the client.  Again, neither log file show an error with a clue to
 the problem. Chrome will download some files, but other files give me an
 error and won't download:
 
 *This webpage is not available*
 *
 *
 *error 100 (net::ERR_CONNECTION_CLOSED): The server unexpectedly closed the
 connection.*
 
 In this case, I don't see anything in the log files either.

Hi Matt,

Are you using the sendfile/x-accel-redirect configurations of
Apache/nginx?  Are all of these issues the same if you connect directly
to Galaxy's built-in webserver instead of through a proxy server?

 
 
 Finally, I'm also having a problem logging in from the main welcome page.
  It seems to authenticate fine (I get an error when I put in the wrong
 password and don't when I put in the correct password), but it doesn't
 actually log me in and just takes me back to the main welcome page.
 
 For this problem, I do get an error in the galaxy log:
 
  *Exception happened during processing of request
 from *
 *self.wfile.flush()*
 *  File /raid/galaxy_user/lib/python2.6/socket.py, line 297, in flush*
 *self._sock.sendall(buffer(data, write_offset, buffer_size))*
 *error: [Errno 32] Broken pipe*

I see this occasionally on python 2.6, and it's usually just a harmless
intentional interrupted connection.  As above, can you test login via
the welcome page without the use of the proxy server and report back
whether there's any difference?

Thanks,
--nate

 
 
 I am able to log in via other routes, such as going to Options-Saved
 Histories which brings up an error You must be logged in to work with
 multiple histories and links to /galaxy/user/login?webapp=galaxy and from
 there I can log in fine. This happens with both admin and normal users.  It
 is not a huge problem since I have a work around, but it would be nice to
 figure out the problem so I don't have to make my users do the same.
 
 I should mention that I'm running galaxy-dist changeset 50e249442c5a on
 CentOS. The problems seem to be web server independent too as they appear
 with both Apache and Nginx. I've used the following instructions, but I'm
 just not seeing where the problem(s) is/are:
 
 https://bitbucket.org/galaxy/galaxy-central/wiki/Config/ProductionServer
 https://bitbucket.org/galaxy/galaxy-central/wiki/Config/ApacheProxy
 https://bitbucket.org/galaxy/galaxy-central/wiki/Config/nginxProxy
 https://bitbucket.org/galaxy/galaxy-central/wiki/Config/WebApplicationScaling
 
 
 Thanks,
 Matt
 
 -- 
 Matthew Conte
 Bioinformatics Scientist
 Department of Biology
 University of Maryland
 mco...@umd.edu

 ___
 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:
 
   http://lists.bx.psu.edu/

___
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:

  http://lists.bx.psu.edu/


Re: [galaxy-dev] problems previewing certain files, downloading files and with login

2011-05-09 Thread Matthew Conte
Hi Nate,


On Mon, May 9, 2011 at 3:30 PM, Nate Coraor n...@bx.psu.edu wrote:

 Matthew Conte wrote:
  Good afternoon,
 
  I'm having several problems with my local galaxy install which may or may
  not be related to one another.
 
  The first problem I'm having is with the Display data in browser
 feature.
   Some files type formats (fastqsanger, fasta) display fine.  However,
 other
  file type formats (png, svg, txt) simply do not display at all. I have
  debugging turned on, but do not see any error within the browser or
 within
  the galaxy log or web server log after clicking the Display data in
  browser eyeball icon.  It is simply a blank middle panel.
 
  The next problem I'm having has to do with downloading files.  Firefox
 and
  Safari will download any file from a history, but will just create an
 empty
  file on the client.  Again, neither log file show an error with a clue to
  the problem. Chrome will download some files, but other files give me an
  error and won't download:
 
  *This webpage is not available*
  *
  *
  *error 100 (net::ERR_CONNECTION_CLOSED): The server unexpectedly closed
 the
  connection.*
 
  In this case, I don't see anything in the log files either.

 Hi Matt,

 Are you using the sendfile/x-accel-redirect configurations of
 Apache/nginx?  Are all of these issues the same if you connect directly
 to Galaxy's built-in webserver instead of through a proxy server?


Yep, I'm using mod_xsendfile-0.12 for apache and x-accel-redirect for nginx,
but both webservers are showing the same problems.  If I use Galaxy's
built-in webserver, I don't have the above problems, so the problem is
probably related to the proxy server. I just can't seem to pinpoint the
problem.



 
 
  Finally, I'm also having a problem logging in from the main welcome page.
   It seems to authenticate fine (I get an error when I put in the wrong
  password and don't when I put in the correct password), but it doesn't
  actually log me in and just takes me back to the main welcome page.
 
  For this problem, I do get an error in the galaxy log:
 
   *Exception happened during processing of request
  from *
  *self.wfile.flush()*
  *  File /raid/galaxy_user/lib/python2.6/socket.py, line 297, in flush*
  *self._sock.sendall(buffer(data, write_offset, buffer_size))*
  *error: [Errno 32] Broken pipe*

 I see this occasionally on python 2.6, and it's usually just a harmless
 intentional interrupted connection.  As above, can you test login via
 the welcome page without the use of the proxy server and report back
 whether there's any difference?

 Thanks,
 --nate


This also works normal with the built-in webserver but not as a proxy
server.

Thanks,
Matt



 
 
  I am able to log in via other routes, such as going to Options-Saved
  Histories which brings up an error You must be logged in to work with
  multiple histories and links to /galaxy/user/login?webapp=galaxy and from
  there I can log in fine. This happens with both admin and normal users.
  It
  is not a huge problem since I have a work around, but it would be nice to
  figure out the problem so I don't have to make my users do the same.
 
  I should mention that I'm running galaxy-dist changeset 50e249442c5a on
  CentOS. The problems seem to be web server independent too as they appear
  with both Apache and Nginx. I've used the following instructions, but I'm
  just not seeing where the problem(s) is/are:
 
  https://bitbucket.org/galaxy/galaxy-central/wiki/Config/ProductionServer
  https://bitbucket.org/galaxy/galaxy-central/wiki/Config/ApacheProxy
  https://bitbucket.org/galaxy/galaxy-central/wiki/Config/nginxProxy
 
 https://bitbucket.org/galaxy/galaxy-central/wiki/Config/WebApplicationScaling
 
 
  Thanks,
  Matt
 
  --
  Matthew Conte
  Bioinformatics Scientist
  Department of Biology
  University of Maryland
  mco...@umd.edu

  ___
  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:
 
http://lists.bx.psu.edu/


___
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:

  http://lists.bx.psu.edu/

Re: [galaxy-dev] problems previewing certain files, downloading files and with login

2011-05-09 Thread Nate Coraor
Matthew Conte wrote:
 Hi Nate,
 
 Yep, I'm using mod_xsendfile-0.12 for apache and x-accel-redirect for nginx,
 but both webservers are showing the same problems.  If I use Galaxy's
 built-in webserver, I don't have the above problems, so the problem is
 probably related to the proxy server. I just can't seem to pinpoint the
 problem.

Okay, thanks for the info.  Have you enabled debug logging in the proxy
and checked the error logs for anything telling?  Same for the login
issue.

--nate

 This also works normal with the built-in webserver but not as a proxy
 server.
 
 Thanks,
 Matt
 
 
 
  
  
   I am able to log in via other routes, such as going to Options-Saved
   Histories which brings up an error You must be logged in to work with
   multiple histories and links to /galaxy/user/login?webapp=galaxy and from
   there I can log in fine. This happens with both admin and normal users.
   It
   is not a huge problem since I have a work around, but it would be nice to
   figure out the problem so I don't have to make my users do the same.
  
   I should mention that I'm running galaxy-dist changeset 50e249442c5a on
   CentOS. The problems seem to be web server independent too as they appear
   with both Apache and Nginx. I've used the following instructions, but I'm
   just not seeing where the problem(s) is/are:
  
   https://bitbucket.org/galaxy/galaxy-central/wiki/Config/ProductionServer
   https://bitbucket.org/galaxy/galaxy-central/wiki/Config/ApacheProxy
   https://bitbucket.org/galaxy/galaxy-central/wiki/Config/nginxProxy
  
  https://bitbucket.org/galaxy/galaxy-central/wiki/Config/WebApplicationScaling
  
  
   Thanks,
   Matt
  
   --
   Matthew Conte
   Bioinformatics Scientist
   Department of Biology
   University of Maryland
   mco...@umd.edu
 
   ___
   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:
  
 http://lists.bx.psu.edu/
 
 
___
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:

  http://lists.bx.psu.edu/


Re: [galaxy-dev] problems previewing certain files, downloading files and with login

2011-05-09 Thread Matthew Conte
Hi Nate,

I've turned both apache and nginx logging up to debug and still don't see
anything of relevance in the error logs for each of the problems I listed.
 Should I send you my universe_wsgi.ini?

Thanks,
Matt

On Mon, May 9, 2011 at 4:15 PM, Nate Coraor n...@bx.psu.edu wrote:

 Matthew Conte wrote:
  Hi Nate,
 
  Yep, I'm using mod_xsendfile-0.12 for apache and x-accel-redirect for
 nginx,
  but both webservers are showing the same problems.  If I use Galaxy's
  built-in webserver, I don't have the above problems, so the problem is
  probably related to the proxy server. I just can't seem to pinpoint the
  problem.

 Okay, thanks for the info.  Have you enabled debug logging in the proxy
 and checked the error logs for anything telling?  Same for the login
 issue.

 --nate

  This also works normal with the built-in webserver but not as a proxy
  server.
 
  Thanks,
  Matt
 
 
  
   
   
I am able to log in via other routes, such as going to
 Options-Saved
Histories which brings up an error You must be logged in to work
 with
multiple histories and links to /galaxy/user/login?webapp=galaxy and
 from
there I can log in fine. This happens with both admin and normal
 users.
It
is not a huge problem since I have a work around, but it would be
 nice to
figure out the problem so I don't have to make my users do the same.
   
I should mention that I'm running galaxy-dist changeset 50e249442c5a
 on
CentOS. The problems seem to be web server independent too as they
 appear
with both Apache and Nginx. I've used the following instructions, but
 I'm
just not seeing where the problem(s) is/are:
   
   
 https://bitbucket.org/galaxy/galaxy-central/wiki/Config/ProductionServer
https://bitbucket.org/galaxy/galaxy-central/wiki/Config/ApacheProxy
https://bitbucket.org/galaxy/galaxy-central/wiki/Config/nginxProxy
   
  
 https://bitbucket.org/galaxy/galaxy-central/wiki/Config/WebApplicationScaling
   
   
Thanks,
Matt
   
--
Matthew Conte
Bioinformatics Scientist
Department of Biology
University of Maryland
mco...@umd.edu
  
___
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:
   
  http://lists.bx.psu.edu/
  
  

___
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:

  http://lists.bx.psu.edu/

Re: [galaxy-dev] problems previewing certain files, downloading files and with login

2011-05-09 Thread Nate Coraor
Matthew Conte wrote:
 Hi Nate,
 
 I've turned both apache and nginx logging up to debug and still don't see
 anything of relevance in the error logs for each of the problems I listed.
  Should I send you my universe_wsgi.ini?

That'd help, and the relevant sections of your Apache and/or nginx
configurations, off-list please.

Thanks,
--nate

 
 Thanks,
 Matt
 
 On Mon, May 9, 2011 at 4:15 PM, Nate Coraor n...@bx.psu.edu wrote:
 
  Matthew Conte wrote:
   Hi Nate,
  
   Yep, I'm using mod_xsendfile-0.12 for apache and x-accel-redirect for
  nginx,
   but both webservers are showing the same problems.  If I use Galaxy's
   built-in webserver, I don't have the above problems, so the problem is
   probably related to the proxy server. I just can't seem to pinpoint the
   problem.
 
  Okay, thanks for the info.  Have you enabled debug logging in the proxy
  and checked the error logs for anything telling?  Same for the login
  issue.
 
  --nate
 
   This also works normal with the built-in webserver but not as a proxy
   server.
  
   Thanks,
   Matt
  
  
   


 I am able to log in via other routes, such as going to
  Options-Saved
 Histories which brings up an error You must be logged in to work
  with
 multiple histories and links to /galaxy/user/login?webapp=galaxy and
  from
 there I can log in fine. This happens with both admin and normal
  users.
 It
 is not a huge problem since I have a work around, but it would be
  nice to
 figure out the problem so I don't have to make my users do the same.

 I should mention that I'm running galaxy-dist changeset 50e249442c5a
  on
 CentOS. The problems seem to be web server independent too as they
  appear
 with both Apache and Nginx. I've used the following instructions, but
  I'm
 just not seeing where the problem(s) is/are:


  https://bitbucket.org/galaxy/galaxy-central/wiki/Config/ProductionServer
 https://bitbucket.org/galaxy/galaxy-central/wiki/Config/ApacheProxy
 https://bitbucket.org/galaxy/galaxy-central/wiki/Config/nginxProxy

   
  https://bitbucket.org/galaxy/galaxy-central/wiki/Config/WebApplicationScaling


 Thanks,
 Matt

 --
 Matthew Conte
 Bioinformatics Scientist
 Department of Biology
 University of Maryland
 mco...@umd.edu
   
 ___
 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:

   http://lists.bx.psu.edu/
   
   
 
___
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:

  http://lists.bx.psu.edu/