You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@qpid.apache.org by "Jeff Stein (JIRA)" <qp...@incubator.apache.org> on 2009/03/26 22:12:50 UTC

[jira] Issue Comment Edited: (QPID-1770) JMS: No useful exception thrown when message is sent to full queue

    [ https://issues.apache.org/jira/browse/QPID-1770?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12689658#action_12689658 ] 

Jeff Stein edited comment on QPID-1770 at 3/26/09 2:12 PM:
-----------------------------------------------------------

I've downloaded trunk, and while I am seeing the new, fixed exceptions, the timing seems off.  If the message that exceeds the queue size is message number N, the exception appears to only get thrown when I send message number N+1.  For example, say queue limit is 100k, and the queue is current 99k.  When I send message N of 2k, no exception is thrown, even though it has exceed the queue size limit and the message is rejected by the queue.  Then, when I send the next message (N+1) of 2k, the exception is thrown.  

In contrast, the Python client seems to throw the exception at message number N, which seems more intuitive to me.

Is this a bug in trunk?

      was (Author: jmstein):
    I've downloaded trunk, and while I am seeing exceptions, the timing seems off.  If the message that exceeds the queue size is message number N, the exception appears to only get thrown when I send message number N+1.  For example, say queue limit is 100k, and the queue is current 99k.  When I send message N of 2k, no exception is thrown, even though it has exceed the queue size limit and the message is rejected by the queue.  Then, when I send the next message (N+1) of 2k, the exception is thrown.  

In contrast, the Python client seems to throw the exception at message number N, which seems more intuitive to me.

Is this a bug in trunk?
  
> JMS: No useful exception thrown when message is sent to full queue
> ------------------------------------------------------------------
>
>                 Key: QPID-1770
>                 URL: https://issues.apache.org/jira/browse/QPID-1770
>             Project: Qpid
>          Issue Type: Bug
>          Components: Java Client
>    Affects Versions: M4
>         Environment: Redhat with C++ broker
>            Reporter: Jeff Stein
>            Assignee: Rajith Attapattu
>            Priority: Critical
>             Fix For: 0.5
>
>
> In JMS, when a message is sent when a queue is already at its maximum size, nothing happens for 60 seconds.  The program just hangs there.  Then, suddenly, these two exceptions are thrown:
> org.apache.qpid.transport.SessionException: timed out waiting for session to become open (state=DETACHED)
> 	at org.apache.qpid.transport.Session.invoke(Session.java:442)
> 	at org.apache.qpid.transport.SessionInvoker.messageTransfer(SessionInvoker.java:96)
> 	at org.apache.qpid.client.BasicMessageProducer_0_10.sendMessage(BasicMessageProducer_0_10.java:160)
> 	at org.apache.qpid.client.BasicMessageProducer.sendImpl(BasicMessageProducer.java:465)
> 	at org.apache.qpid.client.BasicMessageProducer.sendImpl(BasicMessageProducer.java:420)
> 	at org.apache.qpid.client.BasicMessageProducer.send(BasicMessageProducer.java:289)
> 	at Producer.runTest(Producer.java:135)
> 	at Producer.main(Producer.java:64)
> Producer: Caught an Exception: javax.jms.JMSException: Exception when sending message
> javax.jms.JMSException: Exception when sending message
> 	at org.apache.qpid.client.BasicMessageProducer_0_10.sendMessage(BasicMessageProducer_0_10.java:173)
> 	at org.apache.qpid.client.BasicMessageProducer.sendImpl(BasicMessageProducer.java:465)
> 	at org.apache.qpid.client.BasicMessageProducer.sendImpl(BasicMessageProducer.java:420)
> 	at org.apache.qpid.client.BasicMessageProducer.send(BasicMessageProducer.java:289)
> 	at Producer.runTest(Producer.java:135)
> 	at Producer.main(Producer.java:64)
> This is in contrast to, for example, a Python qpid producer, which will throw an exception as soon as the queue is full, and it will be very specific about the problem, telling you the queue is too full, which message overfilled it, etc.
> I tested this using a hacked up version of the direct producer/consumer JMS example.  Just run the producer (without the consumer running) long enough with large enough messages so that it will reach the maximum queue size, and you should be able to see it. 

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


---------------------------------------------------------------------
Apache Qpid - AMQP Messaging Implementation
Project:      http://qpid.apache.org
Use/Interact: mailto:dev-subscribe@qpid.apache.org