[
https://issues.apache.org/jira/browse/QPID-2345?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13073394#comment-13073394
]
Robbie Gemmell commented on QPID-2345:
--------------------------------------
This seems like it was raised against the 0-8 / 0-9 code path as outcome of
QPID-2262, but the 0-10 path can be seen doing similar things as shown via this
comment in QPID-3339:
https://issues.apache.org/jira/browse/QPID-3339?focusedCommentId=13073196&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-13073196
> Java Client not detecting cause of session close from client thread
> -------------------------------------------------------------------
>
> Key: QPID-2345
> URL: https://issues.apache.org/jira/browse/QPID-2345
> Project: Qpid
> Issue Type: Bug
> Components: Java Client
> Affects Versions: 0.6
> Reporter: Martin Ritchie
> Fix For: Future
>
>
> As exemplified by the SimpleACLTest when the broker closes the connection the
> client attempts to stop the current activity and remember the cause of the
> closure to report to the client. However, if the client is in the process of
> performing a operation such as session.createQueue() the closed state can be
> deteted before the cause has been set as a result the IllegalStateException
> is thrown and no cause has been set.
> This was working on 0.5. However, it is possible that the new IO layer is
> highlighting another race condition. Or that it is not correctly setting the
> state to CLOSED. Which is required to be done to prompt the client threads to
> check for a causal exception.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
---------------------------------------------------------------------
Apache Qpid - AMQP Messaging Implementation
Project: http://qpid.apache.org
Use/Interact: mailto:[email protected]
