placeholder: can kapp-controller signal kapp to transiently suppress failures when success is expected to take a few cycles? #349
Labels
carvel-accepted
This issue should be considered for future work and that the triage process has been completed
dk-think-more
enhancement
This issue is a feature request
priority/awaiting-more-evidence
Lowest priority. Possibly useful, but not yet enough support to actually get it done.
priority/unprioritized-backlog
Higher priority than priority/awaiting-more-evidence but not planned. Contributions are welcome.
Describe the problem/challenge you have
Users of kapp-controller, including UI dashboards that rely on kapp-controller under the hood, often see a "reconcile failed" message which resolves into a "reconcile succeeded" message if they wait for some tens of seconds. This false failure undermines user confidence and presents rough edges to end-users.
Describe the solution you'd like
Are there case where kapp-controller "knows" that it's very likely to take a few retry cycles, e.g. when we first have to resolve image pull secrets? If we can recognize those cases, can we invoke kapp with a flag along the lines of "show_failures_as_retries=5" and then kapp will either leave the status as "Reconciling" or update to e.g. "Retrying Reconciling" (might want to discuss with users whether intermediate state would be useful in their UX) until after N failures.
Anything else you would like to add:
[Additional information that will assist in solving the issue.]
Vote on this request
This is an invitation to the community to vote on issues, to help us prioritize our backlog. Use the "smiley face" up to the right of this comment to vote.
👍 "I would like to see this addressed as soon as possible"
👎 "There are other more important things to focus on right now"
We are also happy to receive and review Pull Requests if you want to help working on this issue.
The text was updated successfully, but these errors were encountered: