[ https://issues.apache.org/jira/browse/HBASE-19753?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16322505#comment-16322505 ]
stack commented on HBASE-19753: ------------------------------- Did some quick tests this morning. Replacing the wait-on-connect with the keepalive patch had hbase-zookeeper large tests timeout occasionally. I didn't dig too much. The keepalive looks like we need it. .008 wraps it in. I ran some tests this morning on .007, no timeout addendum, and got failures like below [ERROR] Tests run: 3, Failures: 0, Errors: 1, Skipped: 0, Time elapsed: 159.169 s <<< FAILURE! - in org.apache.hadoop.hbase.zookeeper.TestReadOnlyZKClient [ERROR] testSessionExpire(org.apache.hadoop.hbase.zookeeper.TestReadOnlyZKClient) Time elapsed: 110.726 s <<< ERROR! java.util.concurrent.ExecutionException: org.apache.zookeeper.KeeperException$ConnectionLossException: KeeperErrorCode = ConnectionLoss for /test at org.apache.hadoop.hbase.zookeeper.TestReadOnlyZKClient.testSessionExpire(TestReadOnlyZKClient.java:140) Caused by: org.apache.zookeeper.KeeperException$ConnectionLossException: KeeperErrorCode = ConnectionLoss for /test [ERROR] Tests run: 3, Failures: 0, Errors: 1, Skipped: 0, Time elapsed: 204.377 s <<< FAILURE! - in org.apache.hadoop.hbase.zookeeper.TestReadOnlyZKClient [ERROR] testSessionExpire(org.apache.hadoop.hbase.zookeeper.TestReadOnlyZKClient) Time elapsed: 110.708 s <<< ERROR! java.util.concurrent.ExecutionException: org.apache.zookeeper.KeeperException$ConnectionLossException: KeeperErrorCode = ConnectionLoss for /test at org.apache.hadoop.hbase.zookeeper.TestReadOnlyZKClient.testSessionExpire(TestReadOnlyZKClient.java:140) Caused by: org.apache.zookeeper.KeeperException$ConnectionLossException: KeeperErrorCode = ConnectionLoss for /test Let me try this combined .008. Running it on cluster machines too though am less and less inclined to believe results out of these machines. > Miscellany of fixes for hbase-zookeeper tests to make them more robust > ---------------------------------------------------------------------- > > Key: HBASE-19753 > URL: https://issues.apache.org/jira/browse/HBASE-19753 > Project: HBase > Issue Type: Bug > Reporter: stack > Assignee: stack > Attachments: HBASE-19753.branch-2.001.patch, > HBASE-19753.branch-2.002.patch, HBASE-19753.branch-2.003.patch, > HBASE-19753.branch-2.004.patch, HBASE-19753.branch-2.005.patch, > HBASE-19753.branch-2.006.patch, HBASE-19753.branch-2.007.patch, > HBASE-19753.branch-2.008.patch, keepalive.diff > > > On my cluster which slows zk, tests hbase-zookeeper rarely all pass. -- This message was sent by Atlassian JIRA (v6.4.14#64029)