activemq-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From artnaseef <...@artnaseef.com>
Subject Re: Broker Hangs after some time - or does nothing.
Date Fri, 07 Feb 2014 01:48:29 GMT
Hmm, that is very interesting.  Is the client a C++ (activemq-cpp) client?

So, first off, the "Connection reset" error means a client is losing
connection to the broker without cleaning up first - not sure why, but
that itself isn't indicative of your problem.  Keep it in mind because it
may be triggering the problem, but it's not the cause.

This looks like it could be a bug in the producer, but that's premature to
assert.  Can you look at the producer code to see how it handles failed
sends?  It looks like the producer may be getting stuck on a failure. 
Check how the producer handles connection to the broker when the
connection fails.  If the client is using the failover transport, that
shouldn't be an issue - although it depends on the transport settings.


>
>
> Thanks for reply.
>
> -I checked consumer and produce count on queue and they are 99(c) and 7(p)
> respectively.
> - I see this warning messgage randomly in broker logs: "2014-02-06
> 13:53:28,910 | WARN  | Transport Connection to: tcp://[ip]:63669 failed:
> java.net.SocketException: Connection reset |
> org.apache.activemq.broker.TransportConnection.Transport | ActiveMQ
> Transport: tcp:///[ip]:63669@61616"
>
> and on client side I see following debug messages:
>
> 14:15:32.819 [main] DEBUG o.s.jms.core.JmsTemplate - Sending created
> message: ActiveMQTextMessage {commandId = 0, responseRequired = false,
> messageId = null, originalDestination = null, originalTransactionId =
> null,
> producerId = null, destination = null, transactionId = null, expiration =
> 0, timestamp = 0, arrival = 0, brokerInTime = 0, brokerOutTime = 0,
> correlationId = null, replyTo = null, persistent = false, type = null,
> priority = 0, groupID = null, groupSequence = 0, targetConsumerId = null,
> compressed = false, userID = null, content = null, marshalledProperties =
> null, dataStructure = null, redeliveryCounter = 0, size = 0, properties =
> null, readOnlyProperties = false, readOnlyBody = false, droppable = false,
> text = <?xml version="1.0" encoding="ASCII" ?><Audit...uditMessage>}
> 14:15:33.803
> [com.commons.jms.healthcheck.listener.DefaultMessageListenerContainer-1]
> DEBUG org.apache.activemq.ActiveMQSession -
> ID:C-CP-64386-1391717730167-3:1:1 Transaction Commit :null
> 14:15:34.803
> [com.commons.jms.healthcheck.listener.DefaultMessageListenerContainer-1]
> DEBUG org.apache.activemq.ActiveMQSession -
> ID:C-CP-64386-1391717730167-3:1:1 Transaction Commit :null
> 14:15:35.803
> [com.commons.jms.healthcheck.listener.DefaultMessageListenerContainer-1]
> DEBUG org.apache.activemq.ActiveMQSession -
> ID:C-CP-64386-1391717730167-3:1:1 Transaction Commit :null
> 14:15:36.803
> [com.commons.jms.healthcheck.listener.DefaultMessageListenerContainer-1]
> DEBUG org.apache.activemq.ActiveMQSession -
> ID:C-CP-64386-1391717730167-3:1:1 Transaction Commit :null
> 14:15:37.803
> [com.gwx.commons.jms.healthcheck.listener.DefaultMessageListenerContainer-1]
> DEBUG org.apache.activemq.ActiveMQSession -
> ID:C-CP-64386-1391717730167-3:1:1 Transaction Commit :null
>
> and it just keep on going and never stops. I am not worried about
> healthcheck stuff right now. It's just there and printing logs so ignore
> that part.
>
> this behavior happens when activemq borker is frozen. I mean queue count
> is
> stuck on 230.
>
> thanks,
> chirag
>
>
> On Wed, Feb 5, 2014 at 8:07 PM, artnaseef <art@artnaseef.com> wrote:
>
>> There are many things to check.
>>
>> First off, I have a little confusion.  You metion no log messages on the
>> client or server and then show log messages.  By server, do you mean the
>> broker or something else?
>>
>> Troubleshooting steps:
>> * When flow stops, check the broker for active client connections via
>> the
>> webconsole or JMX.
>> * Check consumer and producer counts for the destinations.
>> * If the failover transport is being used, be sure to use a
>> ConnectionListener and TransportListener to receive notifications of
>> lost
>> connection to the broker.  Note that you should get a log message if the
>> failover transport loses connection, so this may not be necessary.
>> * Check broker JVM memory (especially Heap and PermGen)
>>
>> Also, if the broker is running out of memory due to having too many
>> messages
>> (i.e. slow consumer problem), that could cause the broker to stop
>> functioning.
>>
>>
>>
>>
>> --
>> View this message in context:
>> http://activemq.2283324.n4.nabble.com/Broker-Hangs-after-some-time-or-does-nothing-tp4677506p4677512.html
>> Sent from the ActiveMQ - User mailing list archive at Nabble.com.
>>
>
>
>
>
> _______________________________________________
> If you reply to this email, your message will be added to the discussion
> below:
> http://activemq.2283324.n4.nabble.com/Broker-Hangs-after-some-time-or-does-nothing-tp4677506p4677565.html
> To start a new topic under ActiveMQ - User, email
> ml-node+s2283324n2341805h3@n4.nabble.com
> To unsubscribe from ActiveMQ - User, visit
> http://activemq.2283324.n4.nabble.com/template/NamlServlet.jtp?macro=unsubscribe_by_code&node=2341805&code=YXJ0QGFydG5hc2VlZi5jb218MjM0MTgwNXwtMjA1NDcyNjY5MQ==






--
View this message in context: http://activemq.2283324.n4.nabble.com/Broker-Hangs-after-some-time-or-does-nothing-tp4677506p4677600.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.
Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message