Github user tillrohrmann commented on the pull request:
https://github.com/apache/flink/pull/1230#issuecomment-145846664
For the sake of completeness I repost what I've just commented on the last
commit:
I think it would be better to move the `JobManager` retrieval logic into
the `JobClientActor`. Then the `JobClientActor` would be responsible for
finding the current leader and resubmit the `JobGraph`. The `Client` would
simply wait for the `JobExecutionResult` or an `Exception`.
That way, we would unify the retry logic in one place which would reduce a
little bit of complexity.
Since @uce is quite overloaded right now, I can take a shot at it. Does the
PR depends on anything from the PRs #1227 and #1153?
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
