Am debugging Dr travis on this issue though seems there is linkage on this module. Build plans are passing locally for this commit but opposite with CI. Going through the logs, the jobs are failing on un-modified file. Could there be other qa contributors facing this experience @herbert24 @sharif
cc: @k.joseph @dkayiwa
Thanks for bringing this @kdaud , Sure am en counting same issue, happily enough our automations can run successfull without issues but fails on travis, i think the hacky way is to first do with chrome driver not getting killed at runtime looking into this commit , we may also first try to kill the chrome driver and restart again still looking into this cc @k.joseph
Have dropped a comment on the pr. In the meantime am also looking into this otherwise its seeming to be a blocker. @herbert24 do you have any suggestion on this
well i get my travis too breaking remotely but locally every thing is fine
I would like to take a look at the logs your getting from your side. Do you mind sharing the url here
@herbert24 what experience do you get when you trigger an empty-commit to the remote branch with respective logs
i expect such a pr to pass if it passes locally,checkout its logs too RATEST-97:Fix AddDiagnosisToVisitNoteTest by HerbertYiga · Pull Request #334 · openmrs/openmrs-distro-referenceapplication · GitHub
This is a general concern then. Am curiously looking into this and soon raising a temporarily fix wrapped in a pr
Have worked on the fix and am going to share the pointer to it and look through it. cc: @sharif @herbert24
1 Like
i have just looked through the quick fix you pushed, i have tested one class among those and i saw it passed locally, though it breaks remotely.Am keeping the list of those test classes that a re breaking so that we can actually revisit each one by one
Am working on this and hope things to be right with the commit. Thanks @herbert24 for the outbox review
some two other tests still failing, try ignoring them
Could you take a look at Dr Travis’ report. He seems to be satisfied with the operation. cc: @herbert24 @sharif @k.joseph
@kdaud our role now will be revisiting the tests ignored for each file so as to have them remotely passing
@kdaud we can proggress if we get this merged for the mean time RATEST-108: Fix Travis failure on reference application ui-tests module by kdaud · Pull Request #340 · openmrs/openmrs-distro-referenceapplication · GitHub cc @k.joseph
the master branch should be passing, since all builds are running the same instance, you could be overloading it which is good feedback to us and OpenMRS in general. maybe to speed up things our travis instances can load local instances and test against those
does this mean we need to add what is in our test.properties file remotely
From my side this is not true with upstream encounter