[
https://issues.apache.org/jira/browse/HBASE-12837?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14281017#comment-14281017
]
Hudson commented on HBASE-12837:
--------------------------------
FAILURE: Integrated in HBase-0.98 #803 (See
[https://builds.apache.org/job/HBase-0.98/803/])
HBASE-12837 ReplicationAdmin leaks zk connections (stack) (enis: rev
008cf79545c017a4103adcc2d5e8bfa4868a8df7)
*
hbase-client/src/main/java/org/apache/hadoop/hbase/client/replication/ReplicationAdmin.java
> ReplicationAdmin leaks zk connections
> -------------------------------------
>
> Key: HBASE-12837
> URL: https://issues.apache.org/jira/browse/HBASE-12837
> Project: HBase
> Issue Type: Bug
> Components: Zookeeper
> Reporter: stack
> Assignee: stack
> Fix For: 1.0.0, 2.0.0, 0.98.10, 1.1.0
>
> Attachments: 12837.txt, 12837v2.txt, 12837v2.txt,
> hbase-12837-0.98.patch, hbase-12837-branch-1.patch
>
>
> One of the lads here (Chris Leroy) found issue in ReplicationAdmin where we
> never close up the zk we create in the constructor. He's using it from a
> program that is trying to give state on replication on a period so the
> leaking is pronounced in his case (there is also a System.exit in here which
> messes up hosting programs).
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
