fix: lower log level for ScaledObject paused replicas transitioning state#7650
fix: lower log level for ScaledObject paused replicas transitioning state#7650aliaqel-stripe wants to merge 1 commit into
Conversation
✅ Snyk checks have passed. No issues have been found so far.
💻 Catch issues earlier using the plugins for VS Code, JetBrains IDEs, Visual Studio, and Eclipse. |
|
Thank you for your contribution! 🙏 Please understand that we will do our best to review your PR and give you feedback as soon as possible, but please bear with us if it takes a little longer as expected. While you are waiting, make sure to:
Once the initial tests are successful, a KEDA member will ensure that the e2e tests are run. Once the e2e tests have been successfully completed, the PR may be merged at a later date. Please be patient. Learn more about our contribution guide. |
wozniakjan
left a comment
wozniakjan
left a comment
There was a problem hiding this comment.
lgtm, just would like to wait for #7664 to merge first
rickbrouwer
added
the
waiting-on-other-pr
wozniakjan
added
the
Awaiting/2nd-approval
rickbrouwer
added
merge-conflict
|
Related PR is merged. |
…tate The "ScaledObject paused replicas are being scaled" message was returned as an error, causing the controller to log it at error level and mark the ScaledObject ReadyCondition=False. This is a transient, expected state during the normal pausing transition and should not be treated as an error. Log at warning level and return nil so the ScaledObject is not incorrectly marked as unhealthy during this transition. Fixes kedacore#6604 Signed-off-by: Ali Aqel <aliaqel@stripe.com>
aliaqel-stripe
force-pushed
the
fix/paused-replicas-log-level
branch
from
9cdcd88 to
b4ff8e7
Compare
|
Closing because the changes in #7664 make this PR obsolete. The error is no longer returned |
4 participants
When introducing the
autoscaling.keda.sh/paused-replicasannotation on a ScaledObject, there is a brief window during the first reconciliation where the annotation is set but thePausedConditionstatus hasn't been committed yet. During this window, the controller was returning a non-nil error, causing:logger.Error(...)— error-level log spam on every reconciliation during a normal, expected pausing transitionReadyCondition=Falsewith reasonScaledObjectCheckFailed— incorrectly marking the ScaledObject as unhealthyWarningKubernetes event — unnecessary noise in the event streamFix: log at info level and return
nilso this expected transitional state is not treated as a failure.Checklist
Wait for:
Fixes #6604