jiangzho opened a new pull request, #817:
URL: https://github.com/apache/flink-kubernetes-operator/pull/817
## What is the purpose of the change
This patch is to tackle the corner case where `cleanup` of CustomResource is
constantly failing due to event publish failure, and therefore cause the
`cleanup` hanging forever.
Operator attempts to publish events when reconcile changes or cleaning up
for a CustomResource. This patch allows the reconcile & clean up logic to
proceed, if and only if the event publishing fails as a result of 403
forbidden, which happens when the namespace is being deleted (in terminating
state). k8s rejects events / resource creation in a terminating namespace.
In this way, at-least-once events delivery is still guaranteed in other
cases. Event consumers need to account for the "namespace deleted" scenario,
which all resources within are implicitly deleted.
## Brief change log
- Fix event publish blocking reconcile & cleanup upon 403
## Verifying this change
This change added tests and can be verified as follows:
- Added additional scenario in EventUtilsTest and expect it to succeed
## Does this pull request potentially affect one of the following parts:
- Dependencies (does it add or upgrade a dependency): no
- The public API, i.e., is any changes to the `CustomResourceDescriptors`:
no
- Core observer or reconciler logic that is regularly executed: yes
## Documentation
- Does this pull request introduce a new feature? no
- If yes, how is the feature documented? not applicable
--
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: issues-unsubscr...@flink.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
