[
https://issues.apache.org/jira/browse/FLINK-4653?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15524806#comment-15524806
]
ASF GitHub Bot commented on FLINK-4653:
---------------------------------------
Github user beyond1920 commented on the issue:
https://github.com/apache/flink/pull/2524
@mxm , thanks for your review. I modified the content based on your advice:
1. The changes could compile successfully now, sorry for the level mistake.
2. AwaitJobResult method in JobClientUtils would retry upon
TimeoutException utils JobInfoTracker seems to be dead.
> Refactor JobClientActor to adapt to the new Rpc framework and new cluster
> managerment
> -------------------------------------------------------------------------------------
>
> Key: FLINK-4653
> URL: https://issues.apache.org/jira/browse/FLINK-4653
> Project: Flink
> Issue Type: Sub-task
> Components: Client
> Reporter: zhangjing
> Assignee: zhangjing
> Fix For: 1.2.0
>
>
> 1. Create a RpcEndpoint(temporary named JobInfoTracker) and
> RpcGateway(temporary named JobInfoTrackerGateway) to replace the old
> JobClientActor.
> 2. Change rpc message communication in JobClientActor to rpc method call to
> apply to the new rpc framework.
> 3. JobInfoTracker is responsible for waiting for the jobStateChange and
> jobResult util job complete. But it is not responsible for submitting new job
> because jobSubmission behavior is different in different cluster
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
