You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Is your feature request related to a problem? Please describe.
Most of our jobs on Jenkins are configure to abort after some minutes (usually 20min) of "inactivity" (which is basically the console log not getting messages on a period of time).
The Test can uninstall the operator may get stuck and this usually means a bug, which is fine. The problem is that on being stuck, it will hit the timeout of the job, and it is not clear that the job failed because of the test. Also because the job was aborted, the test has no opportunity to print debugging messages, for example, the logs of the pod(s) pending.
The following snippet of console logs from a failed jobs, show that the job was aborted but it is hard to determine the cause. In this case, the test number 15 (uninstall the operator) is stuck and this is not displayed on the logs:
09:31:08 ok 14 [cc][agent][kubernetes][containerd] Test cannot pull an encrypted image inside the guest without decryption key
09:51:08 Build timed out (after 20 minutes). Marking the build as aborted.
09:51:13 Build was aborted
Describe the solution you'd like
Add the timeout on the test itself. This way:
it is made clear on the console logs of the job that the job failed
Is your feature request related to a problem? Please describe.
Most of our jobs on Jenkins are configure to abort after some minutes (usually 20min) of "inactivity" (which is basically the console log not getting messages on a period of time).
The Test can uninstall the operator may get stuck and this usually means a bug, which is fine. The problem is that on being stuck, it will hit the timeout of the job, and it is not clear that the job failed because of the test. Also because the job was aborted, the test has no opportunity to print debugging messages, for example, the logs of the pod(s) pending.
The following snippet of console logs from a failed jobs, show that the job was aborted but it is hard to determine the cause. In this case, the test number 15 (uninstall the operator) is stuck and this is not displayed on the logs:
Describe the solution you'd like
Add the timeout on the test itself. This way:
Additional context
The enhancement proposed here came from a discussion with @arronwy on #179 (comment)
The text was updated successfully, but these errors were encountered: