activemq-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Gary Tully <>
Subject Re: PrioritizedBackup slows down producers?
Date Tue, 09 Apr 2013 19:05:48 GMT
it may be worth looking at the transportconnector (broker side)
options, updateClusterClientsOnRemove=true and
I that way the client should only be aware of the priority
availability when it returns to the cluster.

If that does not work, raise an enhancement, the sending thread should
learn about the priority broker return in an async manner and get

patches are always welcome :-)

On 9 April 2013 14:12, ByteFlinger <> wrote:
> Hi
> I have setup a network of 2 brokers connected to each other and I am
> performing some tests towards them.
> I have setup prioritizedBackup so that if broker 1 goes down and later on
> comes up again, the producers will connect themselves back to that broker
> rather than keep sending messages to broker 2.
> I have noticed that as soon as I bring broker 1 down, it seems that activemq
> will try to reconnect itself to that broker after sending every single
> message and since it takes some time to timeout (and that the same thread
> seems to be used to reconnect thus blocking sending the next message), that
> means that even though broker 2 is up and running, sending messages to it
> takes considerably more time when broker 1 is down than it does otherwise.
> Is this normal behavior? If so, is there any way to tweak that so that it
> only tries to reconnect every X messages or maybe that it uses a separate
> thread to check whether that broker is up or not? Any other suggestions on
> how one can overcome this issue are welcome.
> Regards
> ByteFlinger
> --
> View this message in context:
> Sent from the ActiveMQ - User mailing list archive at


View raw message