[
https://issues.apache.org/jira/browse/DRILL-4560?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15593088#comment-15593088
]
ASF GitHub Bot commented on DRILL-4560:
---------------------------------------
Github user sudheeshkatkam commented on a diff in the pull request:
https://github.com/apache/drill/pull/626#discussion_r84378620
--- Diff:
exec/java-exec/src/main/java/org/apache/drill/exec/coord/zk/ZKClusterCoordinator.java
---
@@ -158,10 +158,14 @@ public void cacheChanged() {
}
}
+ @Override
public void close() throws Exception {
- // discovery attempts to close its caches(ie serviceCache) already.
however, being good citizens we make sure to
- // explicitly close serviceCache. Not only that we make sure to close
serviceCache before discovery to prevent
- // double releasing and disallowing jvm to spit bothering warnings.
simply put, we are great!
+ // discovery attempts to close its caches(ie serviceCache) already.
however,
--- End diff --
What's changed in the comment? Formatting?
> ZKClusterCoordinator does not call DrillbitStatusListener.drillbitRegistered
> for new bits
> -----------------------------------------------------------------------------------------
>
> Key: DRILL-4560
> URL: https://issues.apache.org/jira/browse/DRILL-4560
> Project: Apache Drill
> Issue Type: Sub-task
> Components: Server
> Affects Versions: 1.6.0
> Reporter: Paul Rogers
> Assignee: Sorabh Hamirwasia
> Priority: Minor
> Fix For: 1.9.0
>
>
> ZKClusterCoordinator notifies listeners of type DrillbitStatusListener when
> drillbits disappear from ZooKeeper. The YARN Application Master (AM) also
> needs to know when bits register themselves with ZK. So, ZKClusterCoordinator
> should change to detect new Drill-bits, then call
> DrillbitStatusListener.drillbitRegistered with the new Drill-bits.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
