You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@activemq.apache.org by "Justin Bertram (JIRA)" <ji...@apache.org> on 2016/08/23 18:41:21 UTC
[jira] [Resolved] (ARTEMIS-689) INTERNAL_ERROR reported on
ClientSession createTemporaryQueue()
[ https://issues.apache.org/jira/browse/ARTEMIS-689?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Justin Bertram resolved ARTEMIS-689.
------------------------------------
Resolution: Cannot Reproduce
Assignee: Justin Bertram
I've attempted to reproduce this on my own, but without a description of the use-case I'm really just guessing. Unsurprisingly, none of my guesses resulted in any kind of failure. I'm closing this until we can get a reproducible test-case.
> INTERNAL_ERROR reported on ClientSession createTemporaryQueue()
> ---------------------------------------------------------------
>
> Key: ARTEMIS-689
> URL: https://issues.apache.org/jira/browse/ARTEMIS-689
> Project: ActiveMQ Artemis
> Issue Type: Bug
> Components: Broker
> Affects Versions: 1.1.0
> Environment: Redhat Linux
> Reporter: Ruben Cala
> Assignee: Justin Bertram
>
> artemis.log message:
> WARN [org.apache.activemq.artemis.core.server] Sending unexpected exception to client: java.lang.NullPointerException
> at org.apache.activemq.artemis.core.postoffice.impl.WildcardAddressManager.addBinding(WildcardAddressManager.java:101) [artemis-server-1.1.0-jar:1.1.0]
> This happened after several successful calls to createTemporaryQueue() using the same address, using a test client running in Eclipse to consume messages. The address used has no wildcards.
> When the broker is in this state, I noticed in JConsole that the Address entry for it remains with failed queue names in the QueueNames and BindingNames attributes, yet there are no Queue entries for them. When I delete the queues (using JConsole), the Address entry goes away, and the createTemporaryQueue() call works again.
> Also, when in this state, the createTemporaryQueue() call works using a wildcard address.
> I suspect that running the test client in Eclipse is preventing a clean connection break from the broker, which is causing the broker to not clean up the queue/address from the test client correctly.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)