curator-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Adarsh Bhat (JIRA)" <j...@apache.org>
Subject [jira] [Created] (CURATOR-56) DistributedQueue node names grow on retries
Date Fri, 20 Sep 2013 00:50:51 GMT
Adarsh Bhat created CURATOR-56:
----------------------------------

             Summary: DistributedQueue node names grow on retries
                 Key: CURATOR-56
                 URL: https://issues.apache.org/jira/browse/CURATOR-56
             Project: Apache Curator
          Issue Type: Bug
          Components: Recipes
    Affects Versions: 2.2.0-incubating
         Environment: Zookeeper server version 3.4.5
            Reporter: Adarsh Bhat


Create a DistributedQueue with a lockPath. If QueueConsumer.consumeMessage() throws an exception,
the message gets re-queued. Every time this happens, the name of the queue item znode grows
in length. After many iterations, node names get very large and the server stops accepting
new client connections.

During a re-queue, the new znode created uses the old znode path as the base, and the SEQUENTIAL
flag is set. Zookeeper appends a sequence number to the old path (which already had a sequence
number), leading to a longer name during each re-queue.

--
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

Mime
View raw message