javax.jms.JMSException: Cannot send, channel has already failed: tcp://127.0.0.1:61616

[Julián <jg...@summan.com>]

Hi all,

I currently have a Jboss Fuse 6.2.1 with 5 root containers and 6 child
containers each.

We have a replicated mode broker profile deployed on mq-01, mq-02, mq-03,
mq-04, mq-05. 

One of the 6 child containers is a mq-XX the same applies to the other child
containers.

In order to find the master we make use of the gateway-mq profile, and this
profile is assigned to gateway child containers and are part of the 6 child
containers ej: gateway-01, gateway-02, gateway-03, gateway-04, gateway-05.


This way our Mediations the only thing they need to do to find the broker is
call localhost:61616

And the gateway-XX child container then sends the petition to the correct
mq-XX. (Master)

The problem I'm seeing is that, sometimes a random gateway-XX child
container, throws an exception:

javax.jms.JMSException: Cannot send, channel has already failed:
tcp://127.0.0.1:61616


And we are forced to restart that container, it's not systematic, it happens
from time to time, randomly.


Has any of you face a similar problem?



--
View this message in context: http://activemq.2283324.n4.nabble.com/javax-jms-JMSException-Cannot-send-channel-has-already-failed-tcp-127-0-0-1-61616-tp4723695.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.

Re: javax.jms.JMSException: Cannot send, channel has already failed: tcp://127.0.0.1:61616

[Tim Bain <tb...@alumni.duke.edu>]

I suspect that multiple things can cause these symptoms. One of them
(according to
https://serverfault.com/questions/550735/activemq-cannot-send-channel-has-already-failed-every-2-seconds)
is having a TCP port check of the sort you might use to monitor whether a
broker is alive. I expect that there are probably others.

If your symptoms aren't being caused by a TCP port check, I recommend you
describe your symptoms in detail, because with your current description
there's not enough detail for anyone to guess what the root cause is.

Tim

On Sep 17, 2017 5:39 PM, "Madhu" <mc...@incomm.com> wrote:

> We are looking for a solution for this issue, appreciate your help
>
> Thanks,
> Madhu
>
>
>
> --
> Sent from: http://activemq.2283324.n4.nabble.com/ActiveMQ-User-
> f2341805.html
>

Re: javax.jms.JMSException: Cannot send, channel has already failed: tcp://127.0.0.1:61616

[Madhu <mc...@incomm.com>]

We are looking for a solution for this issue, appreciate your help

Thanks,
Madhu



--
Sent from: http://activemq.2283324.n4.nabble.com/ActiveMQ-User-f2341805.html

Re: javax.jms.JMSException: Cannot send, channel has already failed: tcp://127.0.0.1:61616

[Madhu <mc...@incomm.com>]

I am facing the same issue, were you able to fix this.

Thanks,
Madhu



--
Sent from: http://activemq.2283324.n4.nabble.com/ActiveMQ-User-f2341805.html

Re: javax.jms.JMSException: Cannot send, channel has already failed: tcp://127.0.0.1:61616

[Raffi <ra...@gmail.com>]

We see the same error, sporadically, we feel it's caused by load-balancers
momentarily directing JMS clients to slaves on initial connect during a
master failover situation, at least that's our best guess.



--
Sent from: http://activemq.2283324.n4.nabble.com/ActiveMQ-User-f2341805.html