Github user tillrohrmann commented on the pull request:
https://github.com/apache/flink/pull/1500#issuecomment-172531939
I have to correct myself. This should not be an issue if you start a Yarn
cluster in HA and then use `bin/flink run ...`, because a
`LeaderRetrievalService` will be instantiated which will take care of this.
However, it should be a problem here. The `FlinkILoop` instantiates a
`ScalaShellRemoteEnvironment` which is a sub class of `RemoteEnvironment`.
Since `RemoteEnvironment` is called with `null` for the client configuration,
it cannot find a possible ZooKeeper instance where the `JobManager` is
registered in the HA case. I think you have to forward the `Configuration`
object created from `flink-conf.yaml` to the `RemoteEnvironment` in order to
make it work.
---
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.
---
