[
https://issues.apache.org/jira/browse/HBASE-24972?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17266393#comment-17266393
]
Prathyusha commented on HBASE-24972:
------------------------------------
Yes, ConnectionLoss is what we get when we try to use a not-yet-connected zk.
[ReadOnlyZKClient|https://github.com/apache/hbase/blob/master/hbase-client/src/main/java/org/apache/hadoop/hbase/zookeeper/ReadOnlyZKClient.java]
of hbase client uses [async
apis|https://github.com/apache/hbase/blob/master/hbase-client/src/main/java/org/apache/hadoop/hbase/zookeeper/ReadOnlyZKClient.java#L263]
of Zookeeper and works with callbacks. So this does not need to wait
explicitly for connection creation and is handled by the async setup of
connection of Zookeeper.
[RecoverableZooKeeper|https://github.com/apache/hbase/blob/master/hbase-zookeeper/src/main/java/org/apache/hadoop/hbase/zookeeper/RecoverableZooKeeper.java]
on the other hand uses [sync
apis|https://github.com/apache/hbase/blob/master/hbase-zookeeper/src/main/java/org/apache/hadoop/hbase/zookeeper/RecoverableZooKeeper.java#L323]
to fetch data from zookeeper and has to wait till the connection is created.
It does so with an [exponential
retry|https://github.com/apache/hbase/blob/master/hbase-zookeeper/src/main/java/org/apache/hadoop/hbase/zookeeper/RecoverableZooKeeper.java#L319]
if the zk conn is not yet up and
[throws|https://github.com/apache/hbase/blob/master/hbase-zookeeper/src/main/java/org/apache/hadoop/hbase/zookeeper/RecoverableZooKeeper.java#L331]
a ConnectionLoss exception once the retries are exhausted. This client is used
by region servers. Thanks.
> Wait for connection attempt to succeed before performing operations on ZK
> -------------------------------------------------------------------------
>
> Key: HBASE-24972
> URL: https://issues.apache.org/jira/browse/HBASE-24972
> Project: HBase
> Issue Type: Bug
> Reporter: Sandeep Guggilam
> Assignee: Prathyusha
> Priority: Minor
>
> {color:#1d1c1d}Creating the connection with ZK is asynchronous and notified
> via the passed in watcher about the successful connection event. When we
> attempt any operations, we try to create a connection and then perform a
> read/write
> ({color}{color:#1d1c1d}[https://github.com/apache/hbase/blob/979edfe72046b2075adcc869c65ae820e6f3ec2d/hbase-zookeeper/src/main/java/org/apache/hadoop/hbase/zookeeper/RecoverableZooKeeper.java#L323]{color}{color:#1d1c1d})
> without really waiting for the notification event
> ([https://github.com/apache/hbase/blob/979edfe72046b2075adcc869c65ae820e6f3ec2d/hbase-zookeeper/src/main/java/org/apache/hadoop/hbase/zookeeper/ZKWatcher.java#L582)]{color}
>
> {color:#1d1c1d}It is possible we get ConnectionLoss errors when we perform
> operations on ZK without waiting for the connection attempt to succeed{color}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
