[ https://issues.apache.org/jira/browse/HBASE-20569?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Guanghao Zhang updated HBASE-20569: ----------------------------------- Resolution: Fixed Status: Resolved (was: Patch Available) Pushed to branch HBASE-19064. Thanks [~Apache9] for reviewing. > NPE in RecoverStandbyProcedure.execute > -------------------------------------- > > Key: HBASE-20569 > URL: https://issues.apache.org/jira/browse/HBASE-20569 > Project: HBase > Issue Type: Sub-task > Reporter: Duo Zhang > Assignee: Guanghao Zhang > Priority: Major > Attachments: HBASE-20569.HBASE-19064.001.patch, > HBASE-20569.HBASE-19064.002.patch, HBASE-20569.HBASE-19064.003.patch, > HBASE-20569.HBASE-19064.004.patch, HBASE-20569.HBASE-19064.005.patch, > HBASE-20569.HBASE-19064.006.patch, HBASE-20569.HBASE-19064.007.patch, > HBASE-20569.HBASE-19064.008.patch, HBASE-20569.HBASE-19064.009.patch, > HBASE-20569.HBASE-19064.010.patch, HBASE-20569.HBASE-19064.011.patch, > HBASE-20569.HBASE-19064.012.patch, HBASE-20569.HBASE-19064.013.patch, > HBASE-20569.HBASE-19064.013.patch, HBASE-20569.HBASE-19064.014.patch, > HBASE-20569.HBASE-19064.015.patch > > > We call ReplaySyncReplicationWALManager.initPeerWorkers in INIT_WORKERS state > and then use it in DISPATCH_TASKS. But if we restart the master and the > procedure is restarted from state DISPATCH_TASKS, no one will call the > initPeerWorkers method and we will get NPE. -- This message was sent by Atlassian JIRA (v7.6.3#76005)