[ https://issues.apache.org/jira/browse/CASSANDRA-10058?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15242791#comment-15242791 ]
James Howe edited comment on CASSANDRA-10058 at 4/15/16 10:48 AM: ------------------------------------------------------------------ Still seeing this LEAK message from the Datastax driver (2.1.9) using Cassandra 2.2.4. was (Author: jameshowe): Still seeing this LEAK message using Cassandra 2.2.4 using Datastax driver 2.1.9. > Close Java driver Client object in Hadoop and Pig classes > --------------------------------------------------------- > > Key: CASSANDRA-10058 > URL: https://issues.apache.org/jira/browse/CASSANDRA-10058 > Project: Cassandra > Issue Type: Bug > Reporter: Alex Liu > Assignee: Alex Liu > Fix For: 2.2.4, 3.0.1, 3.1 > > Attachments: CASSANDRA-10058-2.2.txt > > > I found that some Hadoop and Pig code in Cassandra doesn't close the Client > object, that's the cause for the following errors in java driver 2.2.0-rc1. > {code} > ERROR 11:37:45 LEAK: You are creating too many HashedWheelTimer instances. > HashedWheelTimer is a shared resource that must be reused across the JVM,so > that only a few instances are created. > {code} > We should close the Client objects. -- This message was sent by Atlassian JIRA (v6.3.4#6332)