On Tue, Mar 3, 2015 at 9:51 AM, Peter Cock p.j.a.cock@googlemail.com wrote:
Hello all,
Good news - automated testing on the Galaxy Tool Shed seems to be mostly working again (after a period of trouble in the New Year). Thanks Dave B. et al, I'm sure this has been a lot of effort behind the scenes. I'm looking forward to getting a clean sweep of passing tests again...
I currently have 18 repositories listed with no failing tests :)
The bad news is I currently have four Tool Shed repositories with unexpectedly failing tests (all of which I expected to pass based on past testing, including locally and on TravisCI).
Sorry, my enthusiasm got away with me. There is actually more bad news - I hadn't noticed the latest test was just over a month old - were no tests run at all in February?
I will need some assistance to debug the first three which all fail with the cryptic message "Exception: Job in error state."
(Where a tool has more than one test, I've only shown the first error)
https://toolshed.g2.bx.psu.edu/view/peterjc/blast2go
Test runs 2015-01-29 05:12:50 Automated test environment Tests that failed ...
This was last tested on the Test Tool Shed back on 2015-01-25, failing with my old enemy "Exception: History in error state."
http://toolshed.g2.bx.psu.edu/view/peterjc/clinod
2015-01-29 05:07:17 Automated test environment Tests that failed Tool id: clinod ...
Last tested on the Test Tool Shed on 2015-01-25, giving a more complex "Exception: History in error state." plus "IndexError: list index out of range" traceback.
https://toolshed.g2.bx.psu.edu/view/peterjc/effectivet3
2015-01-29 05:11:05 Automated test environment Tests that failed ...
Last tested on the Test Tool Shed on 2015-01-23, giving a complex failure with "Exception: History in error state." plus "IndexError: list index out of range" and finally: "TypeError: error_unless_accessible() takes exactly 4 arguments (3 given)"
The last error was likely fixed later that day: https://github.com/galaxyproject/galaxy/commit/cdabbe2af8e23cb7414b824b6d819...
The fourth and final failure on the main Tool Shed this morning is from: https://toolshed.g2.bx.psu.edu/view/peterjc/samtools_depad
This is different, "Converting local (test-data) bam to sam failed", already reported (27 Jan): https://lists.galaxyproject.org/pipermail/galaxy-dev/2015-January/021383.htm... http://dev.list.galaxyproject.org/test-base-twilltestcase-py-Converting-loca...
Similarly, last tested on the Test Tool Shed on 2015-01-25 with the same problem.
Peter