-
Notifications
You must be signed in to change notification settings - Fork 3.9k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Flaky test: CMS.Views.Metadata.VideoList #32690
Comments
Another recent run failed on the
|
Another failure in Initialize: https://github.com/openedx/edx-platform/actions/runs/5508776235/jobs/10040523895 |
The JS tests for CMS.Views.Metadata.VideoList are flaky and failing, often on Initialize, so the entire file is being removed. More details on the flakiness can be found here: #32690 Private ticket for fixing: https://2u-internal.atlassian.net/browse/CR-5904
The removal is in this PR: #32697 |
The JS tests for CMS.Views.Metadata.VideoList are flaky and failing, often on Initialize, so the entire file is being removed. More details on the flakiness can be found here: #32690 Private ticket for fixing: https://2u-internal.atlassian.net/browse/CR-5904
Notes:
|
@robrap @timmc-edx Jansen had two follow-up fixes to #32592:
Both fixes involved adding back in CSS that'd accidentally been removed. The flaky error message, "is rendered without opened extra videos bar", sounds to me like something that could happen if CSS is missing. My optimistic hypothesis is that the test failures were happening on branches which didn't contain both of Jansen's fixes. To test that, I've opened #32719 to see if we can get a green build. |
@kdmccormick: I'm pretty sure the test was flaky on branches that did not change commits between a failure and passing. I re-ran the tests myself on one of the failed PRs. UPDATE: See:
Both seem to be for: master b528a27 |
Of course, unless I am misunderstanding you, that's what makes it "flaky", right? I was thinking that my PR, before Jansen's fixes, could have caused the test to become flaky (ie, a true-negative when it fails, and a false-postive when it passes). That said, the build you linked has both of Jansen's commits, so the theory is bunk. |
@robrap It looks like first failing build you linked is on a branch that does not include my PR at all : https://github.com/openedx/edx-platform/commits/1db8c9793fa9ff4b58a8710b25a98832acc1880f Does anyone know whether edx-platform JS tests are run on |
Sorry @kdmccormick. I misunderstood and thought you were questioning whether it was actually flaky. I don’t know the answer to your question. However, we do use these failures to stop our deployments. Also, I feel like I’ve gotten workflow failures after merging, even if for some flaky reason. But these are just anecdotes. |
@robrap With the flaky tests removed, can this issue be closed now since we have a separate (although 2U-private) issue for fixing them? |
Works for me, but can someone link my proposed re-merge in the private issue? |
Yes. I will close the private ticket as well, since the test has been remerged. We'll need to reopen it if it (and maybe this ticket) if it continues to be flaky. Also, I'll copy the revert PR to the private ticket. |
[inform] Private ticket for fixing that has since been closed: https://2u-internal.atlassian.net/browse/CR-5904 |
This occurred again in https://github.com/openedx/edx-platform/actions/runs/5736482000/job/15546160686:
|
A/C:
Details:
We've seen the following transient failure in the js-tests CI check several times today:
Failing line:
edx-platform/cms/static/js/spec/video/transcripts/videolist_spec.js
Line 228 in 9dac65a
The text was updated successfully, but these errors were encountered: