Thanks John, That really helped: Exception: History in error state. -------------------- >> begin captured stdout << --------------------- History with id adb5f5c93f827949 in error - summary of datasets in error below. -------------------------------------- | 1 - filter1_test5.tab (HID - NAME) | Dataset Blurb: | error | Dataset Info: | uploaded tabular file | /usr/local/lib/python2.7/dist-packages/setuptools-2.2-py2.7.egg/pkg_resources.py:991: UserWarning: /home/travis/.python-eggs is writable by group/others and vulnerable to attack when used with get_resource_filename. Consider a more | Dataset Job Standard Output: | *Standard output was empty.* | Dataset Job Standard Error: | /usr/local/lib/python2.7/dist-packages/setuptools-2.2-py2.7.egg/pkg_resources.py:991: UserWarning: /home/travis/.python-eggs is writable by group/others and vulnerable to attack when used with get_resource_filename. Consider a more secure location (set with .set_extraction_path or the PYTHON_EGG_CACHE environment variable). | -------------------------------------- --------------------- >> end captured stdout << ---------------------- Fixed here: https://travis-ci.org/saketkc/galaxy_tools/jobs/23282261#L1402 with: before_install: - mkdir "$HOME/.python-eggs" - chmod 700 "$HOME/.python-eggs" The test is failing, but that I expected it to anyway ;) Thanks everyone! Especially John and Peter! Saket On 18 April 2014 19:15, John Chilton <jmchilton@gmail.com> wrote:
Pushed out this changeset https://bitbucket.org/galaxy/galaxy-central/commits/183668ed911ca0ad937f3ffe... which should allow better reporting of dataset errors in histories.
Make sure export GALAXY_TEST_VERBOSE_ERRORS=True is executed before your tests and then check the output of the API driven version - it should have a breakdown of all datasets in error when it encounters a history that is not ok. This should include the standard error and standard output from the upload tool - which in turn should tell us the exact problem (fingers crossed).
On Fri, Apr 18, 2014 at 8:13 AM, Saket Choudhary <saketkc@gmail.com> wrote:
* fork your repository.
On 18 April 2014 18:42, Saket Choudhary <saketkc@gmail.com> wrote:
On 18 April 2014 18:06, Peter Cock <p.j.a.cock@googlemail.com> wrote:
On Fri, Apr 18, 2014 at 11:47 AM, Saket Choudhary <saketkc@gmail.com> wrote:
On 18 April 2014 14:59, Peter Cock <p.j.a.cock@googlemail.com> wrote:
Very strange. Adding "ls test-data" to the TravisCI list worked, and showed the input file condel_input.tabular was present: https://github.com/saketkc/galaxy_tools/commit/c04ff946db0182de9aca4f6dc36ff... https://s3.amazonaws.com/archive.travis-ci.org/jobs/23239476/log.txt https://s3.amazonaws.com/archive.travis-ci.org/jobs/23239477/log.txt
How about replacing that with "ls -l test-data/condel_input.tabular" to verify the ownership and permissions (although if that was the upload problem, Galaxy could give a clearer error).
Thanks, tried it: https://travis-ci.org/saketkc/galaxy_tools/jobs/23270454#L530
I see, that looks fine:
$ ls -l test-data/condel_input.tabular -rwxrwxr-x 1 travis travis 110 Apr 18 10:28 test-data/condel_input.tabular
This happens irrespective of using ftype. and the plce where ERROR occurs is this:
...
so my dataset state is set to ERROR, essentially upload failed. But why, is what it doesn't tell me.
If this were a systematic problem other people should have seen it - but I have tests using tabular input files which still work fine.
I wonder where exactly in the process it is failing... perhaps in the metadata processing which would be consistent with it being some oddity in your sample file which is causing trouble?
Can you upload this file condel_input.tabular into Galaxy via the UI without error?
Yes, the upload works fine with 'Auto-Detect'. It sets the type to be 'interval'. I am also able to upload if I manually set it to 'tabular', which is anyway expected.
Just to double check I tested it with 'filter1_test5.tab' that comes with galaxy-central and that is failing too: https://travis-ci.org/saketkc/galaxy_tools/jobs/23277155#L1283
I am going to clone your repository and try pushing in my tools to the clone to see if that helps.
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/
To search Galaxy mailing lists use the unified search at: http://galaxyproject.org/search/mailinglists/