activemq-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Gary Tully (JIRA)" <j...@apache.org>
Subject [jira] Commented: (AMQ-2479) Channel was inactive for too long
Date Fri, 06 Nov 2009 13:28:52 GMT

    [ https://issues.apache.org/activemq/browse/AMQ-2479?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=55199#action_55199
] 

Gary Tully commented on AMQ-2479:
---------------------------------

you need to provide some more info here, like the route descriptions on #1 and #2.
In any event, #1 that does a dequeue and enqueue needs to use a dequeue transaction that spans
the enqueue  such that there is no inflight message loss.
Also, in the event of B failure, the failover transport on the jms connection will do an auto
reconnect.

How are the activemq components configured to access brokers, are the brokers persistent?
Again, attaching a test case or some variant of your routes would help.

> Channel was inactive for too long
> ---------------------------------
>
>                 Key: AMQ-2479
>                 URL: https://issues.apache.org/activemq/browse/AMQ-2479
>             Project: ActiveMQ
>          Issue Type: Bug
>          Components: Broker
>    Affects Versions: 5.3.0
>         Environment: Windows 2003 Server RC2, Enterprise Edition, SP2
>            Reporter: Christian A. Kenyeres
>            Priority: Critical
>
> We have application #1 that pulls records from a DB, puts them into messages, and sends
them to a local queue on server A. That queue routes to a 2nd queue on server B using camel
routing. 
> On Server A, the application log and AMQ log both show the messages being created and
persisted in the queue.
> On Server B, application #2 pulls, messages from its queue which were routed from the
queue on Server A.
> From application #2's log, for a 2 minute period of time, ActiveMQ threw this exception:
> 14:11:01,317 ERROR [WrapperSimpleAppMain] (application2.java:1072) - Problem creating
session... javax.jms.JMSException: Channel was inactive for too long: ServerB.foo.com/10.0.40.138:61616
> and there were also 20 other Exceptions thrown:
> 14:12:27,158 ERROR [WrapperSimpleAppMain] (application2.java:1072) - Problem creating
session... javax.jms.JMSException: Could not connect to broker URL: tcp://ServerB.foo.com:61616.
Reason: java.net.ConnectException: Connection refused: connect
> On Server B, application #2's log and the AMQ log have no record of 155 records that
were verified as created, put into messages, and sent on Server A.
> The application blew up and the wrapper service restarted it. All records & messages
before the incident were ok before and after the incident.
> The question is, what happened to the messages when the broker crashed on Server B?
> Software used:
> ActiveMQ 5.3.0
> Camel 2.0.0
> Java JDK 1.6u16
> Java Service Wrapper (tanukisoftware) v 3.3.7 (x86-32)

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


Mime
View raw message