Mailing-List: contact activemq-users-help@geronimo.apache.org; run by ezmlm
Precedence: bulk
Reply-To: activemq-users@geronimo.apache.org
Received-SPF: pass (idunn.apache.osuosl.org: domain gmail.com designates
 66.249.92.168 as permitted sender)
DomainKey-Status: good
DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws;
        s=beta; d=gmail.com;
        h=received:mime-version:in-reply-to:references:content-type:message-id:from:subject:date:to:x-mailer;
        b=Nk1usymCnYki9RP7c/7t2aIr2nlgNWe0N9n8i4WlZm/m5zp0U3IyV/7dR0L7dZpWGjukHjRNhVaE3jpaQlmNUrkXL0gLKvj50BIcesgxf/ci/VnmLuNCZLQNXZoMU4izl5PPQn7Ahom0jLqIGGjJeI01tEqcASOe8NsE6ScKtjs=
Mime-Version: 1.0 (Apple Message framework v752.3)
In-Reply-To: <451A8120.8040001@tid.es>
References: <451A8120.8040001@tid.es>
Content-Type: multipart/alternative; boundary=Apple-Mail-10-671469995
Message-Id: <D9A546A4-0C20-417E-83C2-10D08D845954@gmail.com>
From: Rob Davies <rajdavies@gmail.com>
Subject: Re: QueueBridge and remote broker reconnections
Date: Wed, 27 Sep 2006 15:28:16 +0100
To: activemq-users@geronimo.apache.org

--Apple-Mail-10-671469995
Content-Transfer-Encoding: 7bit
Content-Type: text/plain;
	charset=US-ASCII;
	delsp=yes;
	format=flowed

Hi Manuel,

this looks like a good catch! Would mind opening a jira on this -  
just so it's easier to track - I'll look at this as soon as I can

cheers,

Rob
On 27 Sep 2006, at 14:48, Manuel Teira wrote:

> Hello.
>
> Looking at the code in DestinationBridge  
> (org.apache.activemq.network.jms), I see that when the deliver of a  
> message to the remote broker fails, there's a counter implemented  
> as the var 'attempt' that seems to be thought to mark fails and try  
> to restart the producer.
>
> But, shouldn't that variable be a member of the DestinationBridge  
> class instead of a local variable of the onMessage member method?  
> In this way, the var is always initialized to zero for every  
> onMessage call. So, restartProducer is never called:
>
>     public void onMessage(Message message) {
>         if (started.get() && message != null) {
>             int attempt = 0;
>             try {
>                 if (attempt > 0) {
>                     restartProducer();
>                 }
> ...
>
> In my tests, I've tryed changing the var 'attempt' to be an object  
> member. Now, restartProducer() is called but it seems that the new  
> connection is not being used. Looking at the code, I don't  
> understand how calling
>
> jmsConnector.restartProducerConnection()
>
>
> is really changing the environment of
>
> createProducer()
>
> in the QueueBridge subclass.
>
> For example, for the QueueBridge subclass, createProducer is using  
> the member producerConnection:
>
>     protected MessageProducer createProducer() throws JMSException{
>         producerSession=producerConnection.createQueueSession 
> (false,Session.AUTO_ACKNOWLEDGE);
>         producer = producerSession.createSender(null);
>         return producer;
>     }
>
> but I think that this is not related anymore with the  
> JmsQueueConnector outboundQueueConnection, that is the only  
> affected member in jmsConnector.restartProducerConnection(). I  
> don't ever know how or who, in the initialization is setting up the  
> QueueBridge producerConnection member, calling, I suppose,  
> setProducerConnection.
> I think that a solution to this should be to be able to change the  
> producerConnection of QueueBridge when we are restarting the  
> Producer. But for that, we should need to implement   
> restartProducer in the DestinationBridge subclasses.
>
>
> Any hint or idea? I really need to have remote bridge reconnections  
> working urgently, so please, if you need further info, make me know.
>
> Regards.
>
>


--Apple-Mail-10-671469995--