[ https://issues.apache.org/jira/browse/MESOS-243?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15384410#comment-15384410 ]
Vladimir Petrovic commented on MESOS-243: ----------------------------------------- I have checked MESOS-5262 but I'm not sure that patching agent is a correct fix for the problem. In my opinion, problem is with executors which don't handle shutdown message. With my patch, driver based executor will properly shutdown when it receives shutdown message even if it has pending acknowledgements. > driver stop() should block until outstanding requests have been persisted > ------------------------------------------------------------------------- > > Key: MESOS-243 > URL: https://issues.apache.org/jira/browse/MESOS-243 > Project: Mesos > Issue Type: Bug > Affects Versions: 0.9.0, 0.10.0, 0.11.0, 0.12.0, 0.13.0, 0.14.0, 0.14.1, > 0.14.2, 0.15.0 > Reporter: brian wickman > Assignee: Vladimir Petrovic > Fix For: 1.1.0 > > > in our executor, we send a terminal status update message and immediately > call driver.stop(). it turns out that the status update is dispatched > asynchronously and races with driver shutdown, causing tasks to instead > periodically go into LOST state. -- This message was sent by Atlassian JIRA (v6.3.4#6332)