shuai.xu created FLINK-8938:
-------------------------------
Summary: Not remove job graph during job master failover
Key: FLINK-8938
URL: https://issues.apache.org/jira/browse/FLINK-8938
Project: Flink
Issue Type: Bug
Components: Cluster Management
Affects Versions: 1.5.0
Reporter: shuai.xu
Assignee: shuai.xu
When job master failover， the new dispatcher should not delete the job graph if
failed to start the job manager runner, or else the other dispatchers can not
recover it.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
