[ https://issues.apache.org/jira/browse/HBASE-24052?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17067837#comment-17067837 ]
Bharath Vissapragada commented on HBASE-24052: ---------------------------------------------- Couple of minor comments.., I think we can add a check in the beginning for the following block.. if (master.isStopped() || master.isStopping()) return false.. {code} ServerManager sm = master.getServerManager(); 100 if (sm == null) { 101 LOG.debug("ServerManager is null; stopping={}", master.isStopping()); 102 return false; 103 } 104 sm.registerListener(this); 105 ProcedureExecutor<MasterProcedureEnv> pe = master.getMasterProcedureExecutor(); 106 if (pe == null) { 107 LOG.debug("ProcedureExecutor is null; stopping={}", master.isStopping()); 108 return false; 109 } 110 procedureEnv = pe.getEnvironment(); 111 if (this.procedureEnv == null) { 112 LOG.debug("ProcedureEnv is null; stopping={}", master.isStopping()); 113 return false; 114 } {code} Second..not totally sure if this is the right way.. technically the shutdown() should return irrespective of what happened on the server, no? This is the reason, why I was thinking of adding logging to test to see why the shutdown() fails. Thoughts? {code} () -> masterThread.getMaster().getServerManager() != null && !masterThread.getMaster().isStopping())); {code} > Add debug to TestMasterShutdown > ------------------------------- > > Key: HBASE-24052 > URL: https://issues.apache.org/jira/browse/HBASE-24052 > Project: HBase > Issue Type: Sub-task > Reporter: Michael Stack > Priority: Trivial > Attachments: 0001-HBASE-24052-Add-debug-to-TestMasterShutdown.patch > > > Temporarily add debug to TestMasterShutdown overnight to learn more about a > test failure not reproducible locally. -- This message was sent by Atlassian Jira (v8.3.4#803005)