[ckan-dev] Tests failing on master on Jenkins

Thu Apr 26 18:39:07 UTC 2012

I messed around with Jenkins today and fixed a couple of problems that
were causing tests to fail. There's still 8 tests failing on the last
run on master:

http://s031.okserver.org:8080/job/CKAN_PostgreSQL_Tests/195/

The test_parse_rfc_2822 ones are a timezone thing, those'll be fixed.

The other six are all to do with search queries coming up with the wrong
counts. Hopefully they're all due to the same one or two problems. Has
anyone got any ideas about these?

I think there are yet more tests that are failing on Jenkins
intermittently, or only the first time that the tests are run on a given
commit, etc.

It seems to be quite time consuming to get the tests to pass at all
(never mind reliably) on Jenkins and if a test ever does fail it's more
likely to be something wrong with Jenkins/the server and the scripts
etc. than with CKAN, and not always a quick fix. So I'm wondering if
it's actually worth the trouble to maintain it.

All the postgres tests are passing for me locally.

Saw this recently: http://travis-ci.org/