zookeeper-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Fournier, Camille F. [Tech]" <Camille.Fourn...@gs.com>
Subject RE: ClientCnxn$EventThread leak?
Date Tue, 25 Jan 2011 17:31:59 GMT
I think this might be a side-effect of https://issues.apache.org/jira/browse/ZOOKEEPER-794
which does slightly change the code in the ClientCnxn$EventThread run().

-----Original Message-----
From: Jun Rao [mailto:junrao@gmail.com] 
Sent: Tuesday, January 25, 2011 12:19 PM
To: zookeeper-user@hadoop.apache.org
Subject: ClientCnxn$EventThread leak?


I am using ZK 3.3.0 and observed a thread leaking problem. In the ZK
watcher, when there is a session expiration event, we close the old ZK
client and create a new one. I observed that every time we get a ZK session
expire event, one more ClientCnxn$EventThread thread is added. On the other
hand, there is alway 1 ClientCnxn$SendThread, no matter how many times a
session has expired. If our code caused the thread leak, I'd expect that
both ClientCnxn$EventThread and ClientCnxn$SendThread to increase over time.
Is there a known thread leaking problem in ZK? Thanks,


View raw message