[FLINK-8917] FlinkMiniCluster default createHighAvailabilityServices is not same as ClusterClient

• Type: Bug
• Status: Closed
• Priority: Minor
• Resolution: Won't Fix
• Affects Version/s: 1.4.0
• Component/s: Runtime / Coordination
• Assignee: Unassigned
• Reporter: jianran.tfh
• Created: 12/Mar/18 02:35
• Updated: 28/Feb/19 13:30
• Resolved: 16/Oct/18 19:56

Description

FlinkMiniCluster default createHighAvailabilityServices is not same as ClusterClient，

the FlinkMiniCluster used 

HighAvailabilityServicesUtils.createAvailableOrEmbeddedServices

but the ClusterClient used

HighAvailabilityServicesUtils.createHighAvailabilityServices，so if you use the flink-1.4 in zeppelin，the job submission will be failed with the follow msg: 

Discard message LeaderSessionMessage(00000000-0000-0000-0000-000000000000,SubmitJob(JobGraph(jobId: 33d8e7d74aa48f76a1622d4d8f78105e),EXECUTION_RESULT_AND_STATE_CHANGES)) because the expected leader session ID 87efb7ca-b761-4977-9696-d521bc178703 did not equal the received leader session ID 00000000-0000-0000-0000-000000000000.

Issue Links

• links to: GitHub Pull Request #5678
• links to: GitHub Pull Request #5678
• links to: GitHub Pull Request #5686
• links to: GitHub Pull Request #5686