You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@activemq.apache.org by felixehm <Fe...@cern.ch> on 2016/02/10 20:23:52 UTC

request-reply broken on duplex network (5.12.1) ?

Hi,

before opening an issue, here's what we have encountered on our production
system:

Scenario:
Given two brokers A and B in a dynamic network
(http://activemq.apache.org/networks-of-brokers.html). The bridge is
configured on broker B and works in both directions (duplex) .

{code}
<networkConnectors>
          <networkConnector uri="static://(tcp://0.0.0.0:61640)"
            networkTTL="2"
            messageTTL="2"
            consumerTTL="2"
            conduitSubscriptions="false"
            decreaseNetworkConsumerPriority="false"
            duplex="true"
            suppressDuplicateQueueSubscriptions="false">
            
            <dynamicallyIncludedDestinations>
               <queue physicalName="QUEUE"/>
            </dynamicallyIncludedDestinations>
         </networkConnector>
        </networkConnectors>
{code}

A handler waits on broker B on a queue for incoming requests.
A client connects to A and sends a request to the queue and waits for an
answer.

Result:
The client never receives an answer, 

Remarks:
Although his temp-queue is generated on broker B and the server answers
correctly to it, the message is not forwarded through the network.

Question:
is this correct bahaviour ? From the docs  docs
<http://activemq.apache.org/networks-of-brokers.html>    this is not obvious
.


Thanks for clarification.
Felix





--
View this message in context: http://activemq.2283324.n4.nabble.com/request-reply-broken-on-duplex-network-5-12-1-tp4707392.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.

Re: request-reply broken on dynamic duplex network (5.12.1) ?

Posted by felixehm <Fe...@cern.ch>.
In case somebody is interested, I can send a unit test.



--
View this message in context: http://activemq.2283324.n4.nabble.com/request-reply-broken-on-dynamic-duplex-network-5-12-1-tp4707392p4707393.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.

Re: request-reply broken on duplex network (5.12.1) ?

Posted by Christopher Shannon <ch...@gmail.com>.
I believe this is currently how it is designed to work.  Because the
temporary queue is not part of the dynamicallyIncludedDestinations setting,
the remote broker wouldn't know to forward the message for the new
temporary queue.  I haven't looked too closely but most likely some work
would need to be done to enable temporary destinations to be compatible
with dynamicallyIncludedDestinations.

On Thu, Feb 11, 2016 at 3:12 AM, felixehm <Fe...@cern.ch> wrote:

> Hi David,
>
> yes. its works if you remove this. Is this a normal behavior ? I understood
> that the flag bridgeTempDestinations acts independently and request reply
> works.
> As the temp destination is created on the other broker I would have
> expected
> the correcting bridging of the message.
>
> cheers,
> Felix
>
>
>
> --
> View this message in context:
> http://activemq.2283324.n4.nabble.com/request-reply-broken-on-dynamic-duplex-network-5-12-1-tp4707392p4707408.html
> Sent from the ActiveMQ - User mailing list archive at Nabble.com.
>

Re: request-reply broken on duplex network (5.12.1) ?

Posted by felixehm <Fe...@cern.ch>.
Hi David,

yes. its works if you remove this. Is this a normal behavior ? I understood
that the flag bridgeTempDestinations acts independently and request reply
works. 
As the temp destination is created on the other broker I would have expected
the correcting bridging of the message.

cheers,
Felix



--
View this message in context: http://activemq.2283324.n4.nabble.com/request-reply-broken-on-dynamic-duplex-network-5-12-1-tp4707392p4707408.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.

Re: request-reply broken on duplex network (5.12.1) ?

Posted by Rob Davies <ra...@gmail.com>.
Be interested to see if it works if you remove the dynamicallyIncludedDestinations ?

> On 10 Feb 2016, at 19:23, felixehm <Fe...@cern.ch> wrote:
> 
> Hi,
> 
> before opening an issue, here's what we have encountered on our production
> system:
> 
> Scenario:
> Given two brokers A and B in a dynamic network
> (http://activemq.apache.org/networks-of-brokers.html). The bridge is
> configured on broker B and works in both directions (duplex) .
> 
> {code}
> <networkConnectors>
>          <networkConnector uri="static://(tcp://0.0.0.0:61640)"
>            networkTTL="2"
>            messageTTL="2"
>            consumerTTL="2"
>            conduitSubscriptions="false"
>            decreaseNetworkConsumerPriority="false"
>            duplex="true"
>            suppressDuplicateQueueSubscriptions="false">
> 
>            <dynamicallyIncludedDestinations>
>               <queue physicalName="QUEUE"/>
>            </dynamicallyIncludedDestinations>
>         </networkConnector>
>        </networkConnectors>
> {code}
> 
> A handler waits on broker B on a queue for incoming requests.
> A client connects to A and sends a request to the queue and waits for an
> answer.
> 
> Result:
> The client never receives an answer, 
> 
> Remarks:
> Although his temp-queue is generated on broker B and the server answers
> correctly to it, the message is not forwarded through the network.
> 
> Question:
> is this correct bahaviour ? From the docs  docs
> <http://activemq.apache.org/networks-of-brokers.html>    this is not obvious
> .
> 
> 
> Thanks for clarification.
> Felix
> 
> 
> 
> 
> 
> --
> View this message in context: http://activemq.2283324.n4.nabble.com/request-reply-broken-on-duplex-network-5-12-1-tp4707392.html
> Sent from the ActiveMQ - User mailing list archive at Nabble.com.