chore: configure retries in pre-upgrade hook job (helm)#2710
Conversation
Signed-off-by: Anish Ramasekar <anish.ramasekar@gmail.com>
Codecov ReportPatch coverage has no change and project coverage change:
Additional details and impacted files@@ Coverage Diff @@
## master #2710 +/- ##
==========================================
+ Coverage 52.64% 52.84% +0.20%
==========================================
Files 123 123
Lines 10941 10941
==========================================
+ Hits 5760 5782 +22
+ Misses 4721 4705 -16
+ Partials 460 454 -6
Flags with carried forward coverage won't be shown. Click here to find out more. see 3 files with indirect coverage changes Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here. ☔ View full report in Codecov by Sentry. |
| helm.sh/hook-delete-policy: "hook-succeeded,before-hook-creation" | ||
| spec: | ||
| backoffLimit: 0 | ||
| backoffLimit: 3 |
There was a problem hiding this comment.
why not take the default? something less to configure?
There was a problem hiding this comment.
The default is 6 and retries are with exponential backoff (I've mentioned this in the PR description as well) which could cause the job to hang for longer in case of actual failures. The goal with configuring it to 3 was so there are some retries but still we could fail fast. Let me know what you think!
5 participants
backOffLimitto allow retries on transient failures. The default value is 6 (xref: https://kubernetes.io/docs/concepts/workloads/controllers/job/#pod-backoff-failure-policy). Setting the limit to a lower value as new jobs are created with an exponential back-off delay.fixes #2685