[
https://issues.apache.org/jira/browse/HBASE-8586?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
stack updated HBASE-8586:
-------------------------
Attachment: 8586.txt
Here is what Eric says:
"I noticed the zk leak test still runs a little bit slow and fails
occasionally. I suspect the test fails because zk isn’t shutdown fast enough by
the build server. Is the build server very busy? I increased the sleep time
before checking if zk is closed. I also noticed that there is a 10s
MIN_WAIT_DEAD_SERVER in org.apache.hadoop.hbase.client.ServerCallable that
overrides smaller sleep times for retry in the client. I changed retry value to
1 (which makes it not retry at all) which prevents the 10s sleep from happening
in the loop in the test."
Let me commit this and see...
> Revisit of hbase-8483, "HConnectionManager can leak ZooKeeper connections
> when using deleteStaleConnection"
> -----------------------------------------------------------------------------------------------------------
>
> Key: HBASE-8586
> URL: https://issues.apache.org/jira/browse/HBASE-8586
> Project: HBase
> Issue Type: Bug
> Reporter: stack
> Attachments: 8586.txt
>
>
> hbase-8483 did not fix testDeleteForZKConnLeak definitively. Here is a
> follow on issue.
> Eric Yu has suggested something to try. Will attach the patch.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
