activemq-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Timothy Bish (JIRA)" <j...@apache.org>
Subject [jira] [Closed] (AMQ-3583) Failures to page in queue messages and retrieving messages caused by NullPointerExceptions
Date Fri, 07 Dec 2012 21:43:21 GMT

     [ https://issues.apache.org/jira/browse/AMQ-3583?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Timothy Bish closed AMQ-3583.
-----------------------------

    Resolution: Incomplete

There's not really enough information to investigate this one.  The best option would be to
try a newer release of the broker as 5.4.x is quite old.  Otherwise we'd need some more logs
to see what was happening on the broker at that time, and even better some unit tests that
reproduce this.  
                
> Failures to page in queue messages and retrieving messages caused by NullPointerExceptions
> ------------------------------------------------------------------------------------------
>
>                 Key: AMQ-3583
>                 URL: https://issues.apache.org/jira/browse/AMQ-3583
>             Project: ActiveMQ
>          Issue Type: Bug
>          Components: Broker
>    Affects Versions: 5.4.2
>         Environment: Network of brokers running on Windows XP SP3 machines. Java version
is Sun JRE 1.6.0_23 32 bit.
>            Reporter: Pauli Kaila
>             Fix For: AGING_TO_DIE
>
>         Attachments: exceptions.log
>
>
> Our product is composed of a server and multiple clients. The server and each of the
clients have their own internal ActiveMQ broker which are then joined in to a network of brokers.

> After having ran successfully for almost two months the server on one of our customer
installations stopped communicating with the clients. Based on the logs available (see the
attachment) the issue seems to stem from ActiveMQ, which is failing to page in messages and
retrieving messages because of NullPointerExceptions. The situation was only resolved after
we instructed the customer to restart the server process, which also restarted the ActiveMQ
broker.
> Sadly we do not have remote access to that particular installation so we could not investigate
the issue closer. The only diagnostics we have about the issue are the aforementioned log
statements.
> The broker is using Apache DerbyDB 10.6.1.0 for storage instead of the KahaDB.

--
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