[HBASE-18366] Fix flaky test hbase.master.procedure.TestServerCrashProcedure#testRecoveryAndDoubleExecutionOnRsWithMeta

• Type: Bug
• Status: Closed
• Priority: Major
• Resolution: Not A Problem
• Assignee: Unassigned
• Reporter: Umesh Agashe
• Created: 11/Jul/17 20:22
• Updated: 17/Jun/22 18:13
• Resolved: 25/Jun/18 18:26

Description

It worked for a few days after enabling it with HBASE-18278. But started failing after commits:

6786b2b
68436c9
75d2eca
50bb045
df93c13

It works with one commit before: c5abb6c. Need to see what changed with those commits.

Currently it fails with TableNotFoundException.

Issue Links

• is related to: HBASE-18109 Assign system tables first (priority)
• is related to: HBASE-18495 [AMv2] Make methods isServerOnline() and isServerDead() in ServerManager mutually exclusive
• is related to: HBASE-18491 [AMv2] Fail UnassignProcedure if source Region Server is not online.
• is related to: HBASE-18493 [AMv2] On region server crash do not process system table regions through AssignmentManager.checkIfShouldMoveSystemRegionAsync()
• is related to: HBASE-18492 [AMv2] Embed code for selecting highest versioned region server for system table regions in AssignmentManager.processAssignQueue()
• is related to: HBASE-18494 [AMv2] Modify LoadBalancer to consider highest versioned Region Servers as favorites for system table regions
• is related to: HBASE-18261 [AMv2] Create new RecoverMetaProcedure and use it from ServerCrashProcedure and HMaster.finishActiveMasterInitialization()
• relates to: HBASE-18278 [AMv2] Enable and fix uni test hbase.master.procedure.TestServerCrashProcedure#testRecoveryAndDoubleExecutionOnRsWithMeta
• links to: Debugging overview doc