[
https://issues.apache.org/jira/browse/HIVE-14608?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15432213#comment-15432213
]
Siddharth Seth commented on HIVE-14608:
---------------------------------------
New tasks should not be scheduled on them - because scheduling is based off of
the activeInstanceSet.
For existing tasks, these will eventually timeout after communication failures.
Acting on these actively to disable the node needs to be done is a simple code
change. However it needs testing. Need to get to writing an in-proc
controllable llap test setup.
{code}
getContext().nodesUpdate(List<NodeReport>)
{code}
> LLAP: ZK registry doesn't remove nodes on kill
> -----------------------------------------------
>
> Key: HIVE-14608
> URL: https://issues.apache.org/jira/browse/HIVE-14608
> Project: Hive
> Issue Type: Bug
> Reporter: Sergey Shelukhin
> Assignee: Prasanth Jayachandran
>
> ...and presumably doesn't disable them for scheduling. I haven't looked in
> detail though, I just see some harmless killed tasks in queries after I kill
> some LLAP nodes manually between queries
> {noformat}
> public void workerNodeRemoved(ServiceInstance serviceInstance) {
> // FIXME: disabling this for now
> // instanceToNodeMap.remove(serviceInstance.getWorkerIdentity());
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
