[ https://issues.apache.org/jira/browse/YARN-5128?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Wilfred Spiegelenburg resolved YARN-5128. ----------------------------------------- Resolution: Duplicate The comment was added into the CapacityScheduler only as part of YARN-3223. The discussion led to YARN-4677. I will close this as a dupe of that jira and attach a fix there. > Investigate potential race condition in the scheduler nodeUpdate() method > ------------------------------------------------------------------------- > > Key: YARN-5128 > URL: https://issues.apache.org/jira/browse/YARN-5128 > Project: Hadoop YARN > Issue Type: Task > Reporter: Ray Chiang > Assignee: Wilfred Spiegelenburg > Priority: Major > > This section of code exists in the various schedulers in the method > nodeUpdate(): > {code} > // If the node is decommissioning, send an update to have the total > // resource equal to the used resource, so no available resource to > // schedule. > // TODO: Fix possible race-condition when request comes in before > // update is propagated > if (nm.getState() == NodeState.DECOMMISSIONING) { > this.rmContext > .getDispatcher() > .getEventHandler() > .handle( > new RMNodeResourceUpdateEvent(nm.getNodeID(), ResourceOption > .newInstance(getSchedulerNode(nm.getNodeID()) > .getAllocatedResource(), 0))); > } > {code} > Investigate the TODO section. -- This message was sent by Atlassian JIRA (v7.6.3#76005) --------------------------------------------------------------------- To unsubscribe, e-mail: yarn-dev-unsubscr...@hadoop.apache.org For additional commands, e-mail: yarn-dev-h...@hadoop.apache.org