[
https://issues.apache.org/jira/browse/HBASE-14370?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14737789#comment-14737789
]
Enis Soztutar commented on HBASE-14370:
---------------------------------------
Sorry, I have a hard time understanding the reasoning to go back to approach v1
from v3 patch. The possible race conditions I mention above are not specific to
the Thread vs Executors, it is orthogonal to that. So v1 patch or a wait-signal
version does not buy us anything compared to v3 patch.
In the v3 patch, you are submitting a Runnable thread to the executor which
runs indefinitely everytime node data changes. The lifecycle of
ZKPermissionWatcher is different than AcccessController. I think what happens
is that the AcccessController coprocessor will be stopped everytime a region is
closed from the region server, while the ZKPermissionWatcher is cached via
TableAuthManager.
Let me attach a patch, to explain it better.
> Use separate thread for calling ZKPermissionWatcher#refreshNodes()
> ------------------------------------------------------------------
>
> Key: HBASE-14370
> URL: https://issues.apache.org/jira/browse/HBASE-14370
> Project: HBase
> Issue Type: Bug
> Affects Versions: 0.98.0
> Reporter: Ted Yu
> Assignee: Ted Yu
> Attachments: 14370-v1.txt, 14370-v3.txt, 14370-wait-nofity-v2.txt,
> 14370-wait-nofity.txt
>
>
> I came off a support case (0.98.0) where main zk thread was seen doing the
> following:
> {code}
> at
> org.apache.hadoop.hbase.security.access.ZKPermissionWatcher.refreshAuthManager(ZKPermissionWatcher.java:152)
> at
> org.apache.hadoop.hbase.security.access.ZKPermissionWatcher.refreshNodes(ZKPermissionWatcher.java:135)
> at
> org.apache.hadoop.hbase.security.access.ZKPermissionWatcher.nodeChildrenChanged(ZKPermissionWatcher.java:121)
> at
> org.apache.hadoop.hbase.zookeeper.ZooKeeperWatcher.process(ZooKeeperWatcher.java:348)
> at
> org.apache.zookeeper.ClientCnxn$EventThread.processEvent(ClientCnxn.java:519)
> at org.apache.zookeeper.ClientCnxn$EventThread.run(ClientCnxn.java:495)
> {code}
> There were 62000 nodes under /acl due to lack of fix from HBASE-12635,
> leading to slowness in table creation because zk notification for region
> offline was blocked by the above.
> The attached patch separates refreshNodes() call into its own thread.
> Thanks to Enis and Devaraj for offline discussion.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
