[
https://issues.apache.org/jira/browse/QPID-6377?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15397423#comment-15397423
]
Keith Wall commented on QPID-6377:
----------------------------------
The defect also meant that if the Connection's first attempt to connect to the
Broker failed (i.e. the {{AMQConnection}} constructor threw an exception), the
task pool and its thread would be leaked. The application would have no way to
free this leak as it has no handle to the connection. My second commit
(r.1754392) resolves this issue. This defect could affect someone implementing
their own connection retry logic, or using a framework that does the same thing.
> Connection's taskpool not shutdown if Connection is closed from the broker
> side
> -------------------------------------------------------------------------------
>
> Key: QPID-6377
> URL: https://issues.apache.org/jira/browse/QPID-6377
> Project: Qpid
> Issue Type: Bug
> Components: Java Client
> Affects Versions: 0.18, 0.22, 0.30
> Reporter: Keith Wall
> Fix For: qpid-java-6.1, qpid-java-6.0.5
>
> Attachments: LeakedConnectionThreadsBefore.png
>
>
> If the client's connection is closed from the Broker side, the taskpool used
> to deliver unrouteable messages and some asynchronous exceptions is not
> shutdown. This will cause a small memory leak and potentially leaked
> threads too.
>
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
