Does any one know why these tests are failing. Is the server just slowing down to result into timeout errors? Or something else? https://ci.openmrs.org/browse/REFAPP-DUT-4250/log
@adamg, could you please look into why some tests are getting blocked at times and timing out the build?
tests are passing now but no idea what was wrong. The one thing i’m curious about is this commit, why did you delete that ?
I enabled output from tests so that travis_wait is no longer necessary and we have better insight into test logs. Travis_wait was there to keep the build alive even though there was not output.
The issue is Travis CI times out a build job after 50 minutes so we must keep it under 50 minutes. The question is why the tests run under 25 minutes, but sometimes exceed 50 minutes thus are timed out by Travis.
Here https://saucelabs.com/beta/tests/ba00cbcf90404798a8c55768a83e5a23/watch#204 it looks like the server didn’t start properly. I’ve tried toreproduce this but with no luck.
Thanks @adamg for checking… Could you please look into failing tests right away, if modules fail to start.
https://github.com/openmrs/openmrs-contrib-uitestframework/blob/master/src/main/java/org/openmrs/uitestframework/test/TestBase.java#L240 is checking for status codes, but it should also fail if the displayed page is “OpenMRS Platform … Running”, which means the refapp didn’t start up properly.
We need that, because otherwise tomcat logs do not end up in the travis-ci logs (should be appended at the end) and we can’t see why modules didn’t start. See “The log length has exceeded the limit of 4 MB (this usually means that the test suite is raising the same exception over and over).” at https://api.travis-ci.org/jobs/180338442/log.txt?deansi=true
The reference app build is currently failing.I think this is something of concern and hence thought it must be brought to notice. Here is the link : https://github.com/openmrs/openmrs-module-referenceapplication