potiuk commented on PR #26105: URL: https://github.com/apache/airflow/pull/26105#issuecomment-1238046386
> @potiuk @uranusjr 3 Helm tests failed during last CI run, are they flaky or should I fix them first? @kokrin - your goal is to make it green. And whether to do it is something that you shoud be able to assess yourself I think 1) assume it's you first and look at the actual log/error if you can figure if it is related 2) if this is a once-off failure among multiple similar jobs and seems unrelated/networking problem - commit --amend, push and check if it is reproduced 3) if other PRs and main are failing with similar problems - assume it is a shared/main problem and raise awareness (here or in #development slack) 3) it the log seems releated and failure repeats in several similar jobs - reproduce locally and fix (the K8S tests have now very nice, reproducible path that you can run locally - see the exact recipe in TESTING.rst 4) if you are lost after doing all the above - ping us :) I think this is a good recipe and I wonder if following it will help in this case (I am using this a bit of a testing ground and if it works I will try to make contributors aware of this "approach" they should use - I am looking for a ways to scale the PR process better. Let me know @kokorin - and maybe if you have other suggestions how to improve the "process" that woudl be awesome. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
