August 2014 - galaxy-dev - lists.galaxyproject.org

TypeError with 'dict'
by Martin Christiansen 12 Aug '14

12 Aug '14

Hi all, I'm setting up a local galaxy server and have implemented a tool, but I can't even get it to run. When I try to run it, the terminal says: TypeError: unhashable type: 'dict' The error itself arises from a python egg: File "~/galaxy-dist/eggs/SQLAlchemy-0.7.9-py2.7-linux-x86_64-ucs4.egg/sqlalchemy/orm/identity.py", line 141 in get state = dict.get(self, key, default) Any suggestions of how to solve this? Best regards, Martin Christiansen

2 2

sql server timeout
by Shrum, Donald C 08 Aug '14

08 Aug '14

On occasion I'm getting and 'Internal Server error' on login with my production galaxy server. If I refresh the page I'm able to login just fine. This instance of galaxy has an apache frontend and a mysql server on the back end. If I look at my log files I see: OperationalError: (OperationalError) (2006, 'MySQL server has gone away') 'SELECT galaxy_session.id AS galaxy_session_id............. It looks like perhaps galaxy is trying to use a stale session or something like that. Has anyone else seen this problem and is there an easy solution aside from having users refresh the page? Thanks, Don FSU Research Computing Center

2 1

ToolShed error: samtools installation failure
by Beginner TI 07 Aug '14

07 Aug '14

2 5

"The status is being set to Never installed"
by David Hoover 07 Aug '14

07 Aug '14

When I install a tool from the toolshed that has type="set_environment" as a requirement, I get this in the main.log: Creating a new record for version None of tool dependency SCRIPT_PATH for revision 5fed38c7e063 of repository mirdeep2. The status is being set to Never installed. Consequently, nothing in the tool_dependency directory is created. The tool is shown to have missing dependencies in the 'Manage installed tool shed repositories' panel. Is there something special that must be done to allow a tool to have only an environment setting as a dependency? Is there something that can be done to fix the problem, other than going into the database and changing the tool_dependency.status from 'Never installed' to 'Installed' and manually creating the subdirectories? -------------------- David Hoover, PhD Helix Systems Staff

1 0

changes to upload.py
by Sajdak, Doris 07 Aug '14

07 Aug '14

I'm in an odd situation: I was given an older version of Galaxy with additional tools built into it by a new faculty member. He brought this from his former institutions and wanted us to run his Galaxy instance on our systems. We don't want to support several instances so we agreed to put his tools into our Galaxy implementation. However, we are running into all kinds of snags. The latest issue is that we realized changes have been made to upload.py to allow for additional data types. Here's an example of what I mean: def check_bam( temp_name ): return Bam().sniff( temp_name ) def check_sff( temp_name ): return Sff().sniff( temp_name ) def check_celstgz( temp_name ): return CelsTgz().sniff( temp_name ) While this doesn't seem like a big deal to add those changes into our installation, I'm concerned that doing so will make us unable to update Galaxy in the future. Can you either confirm or deny that changes to upload.py on our system will be overwritten with Galaxy updates? Thanks, Dori ********************************** Dori Sajdak Senior Systems Administrator State University of NY at Buffalo Center for Computational Research 701 Ellicott St Buffalo, NY 14203 Phone: (716) 881-8934 Fax: (716) 849-6656 Web: http://ccr.buffalo.edu **********************************

2 2

testtoolshed : python-2.7 installation error
by Geert Vandeweyer 07 Aug '14

07 Aug '14

Hi, I get an installation error on the python 2.7 package in the test toolshed. I used the 'contact owner' function, but wanted to mention it here too, as there hasn't been reaction so far. Sorry for double posting if so. Error: tar (child): 5.2.tar.bz2: Cannot open: No such file or directory A similar error is in the Test run outputs. I believe it is related to the following (unnecessary) line in the tool_dependency.xml: <action type="change_directory">..</action> located just after the download_file action for the 5.2.tar.bz2 file. Best, Geert -- Geert Vandeweyer, Ph.D. Department of Medical Genetics University of Antwerp Prins Boudewijnlaan 43 2650 Edegem Belgium Tel: +32 (0)3 275 97 56 E-mail: geert.vandeweyer(a)ua.ac.be http://ua.ac.be/cognitivegenetics http://www.linkedin.com/in/geertvandeweyer

3 4

condor compute cluster
by Shrum, Donald C 06 Aug '14

06 Aug '14

Our galaxy server has an apache front end with ldap authentication. I've added a condor destination... <plugin id="condor" type="runner" load="galaxy.jobs.runners.condor:CondorJobRunner"/> <destination id="condor" runner="condor"/> There are some tools I'd like to point to condor but it appears all the jobs will be sent as the user condor. Is there an existing fix that will submit jobs to our condor cluster as the logged in user? If not could someone give me a small pointer for a hack fix? Something along the lines of: galaxy-dist/lib/galaxy/jobs/runners/util/condor/__init__.py line 71: submit = Popen(('condor_submit', submit_file), stdout=PIPE, stderr=STDOUT) Could be changed to sudo -u xxx -c condor_submit... Thanks all, Don FSU Research Computing Center

2 1

August 2014 Galaxy CloudMan release
by Enis Afgan 06 Aug '14

06 Aug '14

*We just released an update to Galaxy CloudMan.* CloudMan offers an easy way to get a personal and completely functional instance of Galaxy in the cloud in just a few minutes, without any manual configuration. This is mostly an incremental bug fix release with the following summary of changes: - On AWS, updated *galaxyFS* snapshot (snap-e6e1c04a), which includes the June 2, 2014 Galaxy release with the July 30th security fix. All the tools installed via the Tool Shed have been updated and a number of new tools added, most notably: Tophat2, Bowtie2, FastQC, several FASTQ manipulation tools, several QC tools. - For AWS, added support for VPC - For OpenStack clouds, added the ability to automatically recover worker instances on cluster reboot - Added support for creating a file system based on a downloadable archive - Do not run Galaxy with multiple processes by default. This is because Tool Shed installs do not work properly in the multi-process mode. This feature can be enabled by setting user data option configure_multiple_galaxy_processes to True when launching an instance. - Set SGE slots in each queue to be equal to the number of cores on the instance - Set instance IP in the Galaxy's FTP data upload tool message - Added support for Nginx v1.4 and allow it (with the PAM module) to used as the authentication mechanism when accessing Galaxy Reports app - Fixed cluster deletion when performed via the API - No longer automatically start Hadoop and HTCondor services - On manually-invoked instance reboots, do not increment the instance reboot count that otherwise eventually leads to instance termination - Limit the size of the log message buffer used in the UI to 1000 lines. Long-running instances had issues with this log growing large and that led to poor UI performance. The complete log is still available from the Admin page (or the command line). - Automatically delete the bucket/container for Test type (*ie*, 'SGE only') clusters on cluster termination For complete details on implemented changes, please see the source code commits <https://bitbucket.org/galaxy/cloudman/commits/all?search=835%3A903> . Enjoy and please let us know what you think, Enis & Dannon & The Galaxy Team https://wiki.galaxyproject.org/Cloud

1 0

Packaging tools with language specific dependencies
by Renato Alves 06 Aug '14

06 Aug '14

Hi everyone, Is there any standard or commonly used way to package tools that have language specific dependencies. I know that with Python libraries one can use setup_virtualenv and with Java jars the JAVA_JAR_LIB strategy is used. Is there anything equivalent for R, Perl and Ruby libraries? Thanks Renato

2 6

Issues with dependencies for toolshed packages
by Iry Witham 06 Aug '14

06 Aug '14

Hi Team, I have been trying to install package_snpEff_3_5 from the Galaxy main tool shed and package_snpEff_3_6 from the Galaxy test tool shed and have the same issue. Once the installer completes it shows that the dependency snpEff.x.x was not installed. When I try to install the dependency it reports that the directory /hpcdata/galaxy-test/galaxy-setup/galaxy-dist/database/tmp/tmp-toolshed-mtdi48tQV/snpEff does not exist. I have checked to confirm that this is true and discovered that rather then a directory I find a null file named "snpEff and the snpEff_v3.x_core.zip file. I checked my shedtool dependencies folder and the directory for these tools are empty. I have since uninstalled the package. I just tried again to install the snpEff tool and am getting the followig error message: Internal Server Error Galaxy was unable to successfully complete your request URL: http://galaxy2/admin_toolshed/prepare_for_install?tool_shed_url=https://too… Module galaxy.web.framework.middleware.error:149 in __call__ >> <http://galaxy2/admin_toolshed/prepare_for_install?tool_shed_url=https://too…> app_iter = self.application(environ, sr_checker) Module paste.recursive:84 in __call__ >> <http://galaxy2/admin_toolshed/prepare_for_install?tool_shed_url=https://too…> return self.application(environ, start_response) Module paste.httpexceptions:633 in __call__ >> <http://galaxy2/admin_toolshed/prepare_for_install?tool_shed_url=https://too…> return self.application(environ, start_response) Module galaxy.web.framework.base:132 in __call__ >> <http://galaxy2/admin_toolshed/prepare_for_install?tool_shed_url=https://too…> return self.handle_request( environ, start_response ) Module galaxy.web.framework.base:190 in handle_request >> <http://galaxy2/admin_toolshed/prepare_for_install?tool_shed_url=https://too…> body = method( trans, **kwargs ) Module galaxy.web.framework:377 in decorator >> <http://galaxy2/admin_toolshed/prepare_for_install?tool_shed_url=https://too…> return func( self, trans, *args, **kwargs ) Module galaxy.webapps.galaxy.controllers.admin_toolshed:1035 in prepare_for_install >> <http://galaxy2/admin_toolshed/prepare_for_install?tool_shed_url=https://too…> raw_text = common_util.tool_shed_get( trans.app, tool_shed_url, url ) Module tool_shed.util.common_util:310 in tool_shed_get >> <http://galaxy2/admin_toolshed/prepare_for_install?tool_shed_url=https://too…> response = urlopener.open( uri ) Module urllib2:395 in open >> <http://galaxy2/admin_toolshed/prepare_for_install?tool_shed_url=https://too…> response = meth(req, response) Module urllib2:508 in http_response >> <http://galaxy2/admin_toolshed/prepare_for_install?tool_shed_url=https://too…> 'http', request, response, code, msg, hdrs) Module urllib2:427 in error >> <http://galaxy2/admin_toolshed/prepare_for_install?tool_shed_url=https://too…> result = self._call_chain(*args) Module urllib2:367 in _call_chain >> <http://galaxy2/admin_toolshed/prepare_for_install?tool_shed_url=https://too…> result = func(*args) Module urllib2:603 in http_error_302 >> <http://galaxy2/admin_toolshed/prepare_for_install?tool_shed_url=https://too…> return self.parent.open(new) Module urllib2:395 in open >> <http://galaxy2/admin_toolshed/prepare_for_install?tool_shed_url=https://too…> response = meth(req, response) Module urllib2:508 in http_response >> <http://galaxy2/admin_toolshed/prepare_for_install?tool_shed_url=https://too…> 'http', request, response, code, msg, hdrs) Module urllib2:433 in error >> <http://galaxy2/admin_toolshed/prepare_for_install?tool_shed_url=https://too…> return self._call_chain(*args) Module urllib2:367 in _call_chain >> <http://galaxy2/admin_toolshed/prepare_for_install?tool_shed_url=https://too…> result = func(*args) Module urllib2:516 in http_error_default >> <http://galaxy2/admin_toolshed/prepare_for_install?tool_shed_url=https://too…> raise HTTPError(req.get_full_url(), code, msg, hdrs, fp) HTTPError: HTTP Error 500: Internal Server Error extra data<http://galaxy2/admin_toolshed/prepare_for_install?tool_shed_url=https://too…> full traceback<http://galaxy2/admin_toolshed/prepare_for_install?tool_shed_url=https://too…> text version<http://galaxy2/admin_toolshed/prepare_for_install?tool_shed_url=https://too…> This may be an intermittent problem due to load or other unpredictable factors, reloading the page may address the problem. I appreciate any possible assistance, Regards, __________________________________ Iry T. Witham Scientific Applications Administrator Scientific Computing Group Computational Sciences Dept. The Jackson Laboratory 600 Main Street Bar Harbor, ME 04609 Phone: 207-288-6744 email: iry.witham(a)jax.org [cid:8C546FFA-BCF0-42A9-8AB5-4F9CE2ADED3C] The information in this email, including attachments, may be confidential and is intended solely for the addressee(s). If you believe you received this email by mistake, please notify the sender by return email as soon as possible.

3 4