Test Tool Shed, nightly tests: Error 'unicode' object has no attribute 'get'
Hi guys, Yesterday I made a small tweak to the BLAST+ packages to fix a bash syntax error in the fall back action: https://github.com/peterjc/galaxy_blast/commit/4b2fa1fbf485c4630cb6dc0beaf72... I uploaded this update to the Test Tool Shed packages (all four, BLAST+ 2.2.26, 2.2.27, 2.2.28 and 2.2.29), most of which now unfortunately all show up under "Latest revision: installation errors" (along with the BLAST+ wrappers): Name↓Latest Installable RevisionOwnerncbi_blast_plus<http://testtoolshed.g2.bx.psu.edu/repository/browse_my_writable_repositories_with_install_errors?sort=name&operation=view_or_manage_repository&id=c1542e8b1988898c> 39:22b7cdcf4960 (2014-02-20) peterjcpackage_blast_plus_2_2_26<http://testtoolshed.g2.bx.psu.edu/repository/browse_my_writable_repositories_with_install_errors?sort=name&operation=view_or_manage_repository&id=2f85dfc5735b60ae> 5:ebde5fa60194 (2014-02-20) iucpackage_blast_plus_2_2_28<http://testtoolshed.g2.bx.psu.edu/repository/browse_my_writable_repositories_with_install_errors?sort=name&operation=view_or_manage_repository&id=bda6192e64fc6e5d> 2:6f6a8be93479 (2014-02-20) iucpackage_blast_plus_2_2_29<http://testtoolshed.g2.bx.psu.edu/repository/browse_my_writable_repositories_with_install_errors?sort=name&operation=view_or_manage_repository&id=236d3184a5d7f542> 1:c021862e9ea8 (2014-02-20) iuc The BLAST+ 2.2.26 package doesn't seem to show an error: Automated tool test results Test runs 2014-02-20 07:58:47 Tool dependencies TypeNameVersion blast+ package 2.2.26+ Error None The BLAST+ 2.2.27 package seems to have no test results at all. However, the BLAST+ 2.2.28 package has this strange unicode error: Automated tool test results Test runs 2014-02-20 12:48:52 This repository Error 'unicode' object has no attribute 'get' Error 'unicode' object has no attribute 'get' Error 'unicode' object has no attribute 'get' Error 'unicode' object has no attribute 'get' And similarly for BLAST+ 2.2.29, Automated tool test results Test runs 2014-02-20 14:37:58 This repository Error 'unicode' object has no attribute 'get' Error 'unicode' object has no attribute 'get' Error 'unicode' object has no attribute 'get' Error 'unicode' object has no attribute 'get' This is puzzling, since the changes I made to the packages seemed innocuous - making me wonder if something changed in the Test Tool Shed itself? Regards, Peter
Hello Peter, On Feb 21, 2014, at 4:53 AM, Peter Cock <p.j.a.cock@googlemail.com> wrote:
Hi guys,
Yesterday I made a small tweak to the BLAST+ packages to fix a bash syntax error in the fall back action:
https://github.com/peterjc/galaxy_blast/commit/4b2fa1fbf485c4630cb6dc0beaf72...
The BLAST+ 2.2.27 package seems to have no test results at all.
The above issue was due to a bug that was introduced into the install and test framework late yesterday, causing the test to not run at all. The bug has been corrected in preparation for tonight's test run. The Test environment entry for the above repository was produced by the preparation script.
However, the BLAST+ 2.2.28 package has this strange unicode error:
Automated tool test results
Test runs 2014-02-20 12:48:52 This repository Error 'unicode' object has no attribute 'get' Error 'unicode' object has no attribute 'get' Error 'unicode' object has no attribute 'get' Error 'unicode' object has no attribute 'get'
And similarly for BLAST+ 2.2.29,
Automated tool test results
Test runs 2014-02-20 14:37:58 This repository Error 'unicode' object has no attribute 'get' Error 'unicode' object has no attribute 'get' Error 'unicode' object has no attribute 'get' Error 'unicode' object has no attribute 'get'
This is puzzling, since the changes I made to the packages seemed innocuous - making me wonder if something changed in the Test Tool Shed itself?
The above 2 issues were the result of a bug in the Install and Test Framework that was exposed only with the changes you made to you repositories. The bug resulted in invalid information being included in the test results database column, and the "unicode object has not attribute get" error you're seeing above is due to the exception handling for displaying the invalid data in the Tool Shed repository's Test runs container. We've corrected the data in the database as well as the bug that produced it, so tonight's test run should not result in this behavior for these repositories. Thanks very much! Greg Von Kuster
On Fri, Feb 21, 2014 at 6:54 PM, Greg Von Kuster <greg@bx.psu.edu> wrote:
Hello Peter,
On Feb 21, 2014, at 4:53 AM, Peter Cock <p.j.a.cock@googlemail.com> wrote:
Hi guys,
Yesterday I made a small tweak to the BLAST+ packages to fix a bash syntax error in the fall back action:
https://github.com/peterjc/galaxy_blast/commit/4b2fa1fbf485c4630cb6dc0beaf72...
The BLAST+ 2.2.27 package seems to have no test results at all.
The above issue was due to a bug that was introduced into the install and test framework late yesterday, causing the test to not run at all. The bug has been corrected in preparation for tonight's test run. The Test environment entry for the above repository was produced by the preparation script.
However, the BLAST+ 2.2.28 package has this strange unicode error:
Automated tool test results ... Error 'unicode' object has no attribute 'get'
And similarly for BLAST+ 2.2.29, ... Error 'unicode' object has no attribute 'get'
This is puzzling, since the changes I made to the packages seemed innocuous - making me wonder if something changed in the Test Tool Shed itself?
The above 2 issues were the result of a bug in the Install and Test Framework that was exposed only with the changes you made to you repositories. The bug resulted in invalid information being included in the test results database column, and the "unicode object has not attribute get" error you're seeing above is due to the exception handling for displaying the invalid data in the Tool Shed repository's Test runs container. We've corrected the data in the database as well as the bug that produced it, so tonight's test run should not result in this behaviour for these repositories.
Thanks very much!
Greg Von Kuster
Hi Greg, Good progress - no sign of the unicode error :) I now have the following three repositories shown under "Latest revision: installation errors" http://testtoolshed.g2.bx.psu.edu/view/peterjc/ncbi_blast_plus (Seems to list both a failed and a successful BLAST+ 2.2.29 install?) http://testtoolshed.g2.bx.psu.edu/view/iuc/package_blast_plus_2_2_26 (No error?) http://testtoolshed.g2.bx.psu.edu/view/iuc/package_blast_plus_2_2_28 (No error?) On the other hand, http://testtoolshed.g2.bx.psu.edu/view/iuc/package_blast_plus_2_2_27 (No test/install results?) http://testtoolshed.g2.bx.psu.edu/view/iuc/package_blast_plus_2_2_29 Looks good. So it seems there is still something not quite right here... Thanks, Peter
Hello Peter, On Feb 24, 2014, at 3:32 AM, Peter Cock <p.j.a.cock@googlemail.com> wrote:
Hi Greg,
Good progress - no sign of the unicode error :)
I now have the following three repositories shown under "Latest revision: installation errors"
http://testtoolshed.g2.bx.psu.edu/view/peterjc/ncbi_blast_plus (Seems to list both a failed and a successful BLAST+ 2.2.29 install?)
http://testtoolshed.g2.bx.psu.edu/view/iuc/package_blast_plus_2_2_26 (No error?)
http://testtoolshed.g2.bx.psu.edu/view/iuc/package_blast_plus_2_2_28 (No error?)
On the other hand,
http://testtoolshed.g2.bx.psu.edu/view/iuc/package_blast_plus_2_2_27 (No test/install results?)
http://testtoolshed.g2.bx.psu.edu/view/iuc/package_blast_plus_2_2_29 Looks good.
So it seems there is still something not quite right here…
We've eliminated the use of fabric for installing tool dependencies on the test tool shed, and with this change I've seen the test tool shed fail at the same point the past 3 runs. It fails attempting to compile boost where it times out ( we have abuildbot timeout set at 36000 seconds if nothing occurs). This error occurs about half way through Stage 1 of the test run, so about 1/2 of the tool dependency definition repositories are not tested and no tests are run for repositories with tools. The following log shows the repeaated compilation error. We'll be investigating approaches for handling issues like this. Now that we've eliminated fabric we'll have many more options for handling installation issues like this. Of course, if it turns out that our new code that eliminated fabric caused this, we'll get a fix in. Thanks, Greg Von Kuster fabric_util.STDOUT DEBUG 2014-02-23 20:20:14,740 -- NUMBER_OF_JOBS: 2 (maximal number of concurrent compile jobs) fabric_util.STDOUT DEBUG 2014-02-23 20:20:14,741 -- Extracting BOOST .. fabric_util.STDOUT DEBUG 2014-02-23 20:20:20,006 -- Extracting BOOST .. done fabric_util.STDOUT DEBUG 2014-02-23 20:20:20,006 -- Bootstrapping Boost libraries (./bootstrap.sh --prefix=/var/opt/buildslaves/buildslave-ec2-2/buildbot-install-test-test-tool-shed-py27/build/test/install_and_test_tool_shed_repositories/tool_dependency_definitions/tmp/tmpShf1jI/tmp-toolshed-mtdq8JEyP/OpenMS-1.11.1/contrib --with-libraries=date_time,iostreams,math,regex) ... fabric_util.STDOUT DEBUG 2014-02-23 20:20:37,231 -- Bootstrapping Boost libraries (./bootstrap.sh --prefix=/var/opt/buildslaves/buildslave-ec2-2/buildbot-install-test-test-tool-shed-py27/build/test/install_and_test_tool_shed_repositories/tool_dependency_definitions/tmp/tmpShf1jI/tmp-toolshed-mtdq8JEyP/OpenMS-1.11.1/contrib --with-libraries=iostreams,math,date_time,regex) ... done fabric_util.STDOUT DEBUG 2014-02-23 20:20:37,232 -- Installing Boost libraries (./bjam -j 2 -sZLIB_SOURCE=/var/opt/buildslaves/buildslave-ec2-2/buildbot-install-test-test-tool-shed-py27/build/test/install_and_test_tool_shed_repositories/tool_dependency_definitions/tmp/tmpShf1jI/tmp-toolshed-mtdq8JEyP/OpenMS-1.11.1/contrib/src/zlib-1.2.5 -sBZIP2_SOURCE=/var/opt/buildslaves/buildslave-ec2-2/buildbot-install-test-test-tool-shed-py27/build/test/install_and_test_tool_shed_repositories/tool_dependency_definitions/tmp/tmpShf1jI/tmp-toolshed-mtdq8JEyP/OpenMS-1.11.1/contrib/src/bzip2-1.0.5 link=static cxxflags=-fPIC install --build-type=complete --layout=tagged --threading=single,multi) ... fabric_util.STDOUT DEBUG 2014-02-23 20:36:11,619 -- Installing Boost libraries (./bjam -j 2 -sZLIB_SOURCE=/var/opt/buildslaves/buildslave-ec2-2/buildbot-install-test-test-tool-shed-py27/build/test/install_and_test_tool_shed_repositories/tool_dependency_definitions/tmp/tmpShf1jI/tmp-toolshed-mtdq8JEyP/OpenMS-1.11.1/contrib/src/zlib-1.2.5 -sBZIP2_SOURCE=/var/opt/buildslaves/buildslave-ec2-2/buildbot-install-test-test-tool-shed-py27/build/test/install_and_test_tool_shed_repositories/tool_dependency_definitions/tmp/tmpShf1jI/tmp-toolshed-mtdq8JEyP/OpenMS-1.11.1/contrib/src/bzip2-1.0.5 link=static cxxflags=-fPIC install --build-type=complete --layout=tagged --threading=single,multi) ... failed command timed out: 36000 seconds without output, attempting to kill process killed by signal 9
Thanks,
Peter
On Mon, Feb 24, 2014 at 12:03 PM, Greg Von Kuster <greg@bx.psu.edu> wrote:
Hello Peter,
We've eliminated the use of fabric for installing tool dependencies on the test tool shed, and with this change I've seen the test tool shed fail at the same point the past 3 runs. It fails attempting to compile boost where it times out ( we have abuildbot timeout set at 36000 seconds if nothing occurs). This error occurs about half way through Stage 1 of the test run, so about 1/2 of the tool dependency definition repositories are not tested and no tests are run for repositories with tools.
Ah. So the apparently random split in behaviour for the BLAST+ packages may just be down to if they were in the first or second half of the repositories?
The following log shows the repeaated compilation error. We'll be investigating approaches for handling issues like this. Now that we've eliminated fabric we'll have many more options for handling installation issues like this. Of course, if it turns out that our new code that eliminated fabric caused this, we'll get a fix in.
If push comes to shove, you could declare BOOST an expected library to be pre-installed? https://wiki.galaxyproject.org/Admin/Config/ToolDependenciesList Peter
Hi, I got an error in “Join two Datasets" (version 2.0.2). /home/galaxy-dist/lib/galaxy/__init__.py:5: UserWarning: Module psyco_full was already imported from /home/galaxy-dist/lib/psyco_full.py, but /usr/lib64/python2.6/site-packages is being added to sys.path __import__( "pkg_resources" ).declare_namespace( __name__ ) Do you know, what could be the reason? Thanks, Alper Kucukural, PhD Bioinformatics Core, University of Massachusetts Medical School 368 Plantation St.Room AS4.2067 Worcester, MA 01605-2324 Phone: 774-312-4493 E-mail: alper@kucukural.com
Hi Alper, as the error suggests you have two versions of psyco_full in your PYTHONPATH, is it rseqc? You can get rid of any of these versions or run Galaxy inside a virtualenv if you like: wget https://raw.github.com/pypa/virtualenv/master/virtualenv.py python ./virtualenv.py --no-site-packages galaxy_env . ./galaxy_env/bin/activate Cheers, Bjoern
Hi, I got an error in “Join two Datasets" (version 2.0.2).
/home/galaxy-dist/lib/galaxy/__init__.py:5: UserWarning: Module psyco_full was already imported from /home/galaxy-dist/lib/psyco_full.py, but /usr/lib64/python2.6/site-packages is being added to sys.path __import__( "pkg_resources" ).declare_namespace( __name__ )
Do you know, what could be the reason? Thanks,
Alper Kucukural, PhD Bioinformatics Core, University of Massachusetts Medical School 368 Plantation St.Room AS4.2067 Worcester, MA 01605-2324 Phone: 774-312-4493 E-mail: alper@kucukural.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/
To search Galaxy mailing lists use the unified search at: http://galaxyproject.org/search/mailinglists/
participants (4)
-
Alper Kucukural -
Björn Grüning -
Greg Von Kuster -
Peter Cock