gyfora opened a new pull request, #557:
URL: https://github.com/apache/flink-kubernetes-operator/pull/557
## What is the purpose of the change
Introduce config to block last-state upgrades if the checkpoint is too old
and either wait for pending checkpoint completion or use savepoints instead.
When the max-age checkpoint age is configured we will do the following if
the last completed checkpoint/savepoint is older:
- If there is any pending checkpoint triggered within the max-age, we wait
for it to complete
- If the job started within max-age we use last-state restart
- If no pending checkpoint and the job wasn't started recently, we fall
back to savepoint upgrade mode.
We also introduce a validation to only allow the max-age config if savepoint
dir is set.
As a followup improvement we could consider triggering chekpoint instead of
savepoint for Flink 1.17 in the future.
## Brief change log
- *Introduce new logic in available upgrade mode method*
- *Add test for upgrademode*
- *Add test for validation*
- *Minor TestingFlinkService refactor*
## Verifying this change
New unit tests added for new functionality and existing tests guard against
regressions.
## 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? yes
- If yes, how is the feature documented? JavaDocs
--
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: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
