Mailing-List: contact users-help@activemq.apache.org; run by ezmlm
Precedence: bulk
Reply-To: users@activemq.apache.org
Received-SPF: neutral (nike.apache.org: local policy)
Date: Mon, 20 Oct 2014 15:19:49 -0700 (PDT)
From: artnaseef <art@artnaseef.com>
To: users@activemq.apache.org
Message-ID: <1413843589787-4686566.post@n4.nabble.com>
In-Reply-To: 
 <CAMH6+awWT8ykan3kxYP5jfnSkAoCP0_oyHVWjbjPRs3au0iRyw@mail.gmail.com>
References: <1413405556270-4686426.post@n4.nabble.com>
 <CAMH6+awWT8ykan3kxYP5jfnSkAoCP0_oyHVWjbjPRs3au0iRyw@mail.gmail.com>
Subject: Re: 5.9 to 5.10 Upgrade
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Yeah, it should work fine.  Just keep in mind that messages on a broker are
only reactively moved to another broker, and they only ever live on a single
broker at a time (for the most part).

So, shutting down a broker with messages that have not drained will either
lose those messages (for non-persistent messages, including Topic and temp
dest messages) or delay their delivery (persistent messages).

To drain messages, it is possible to kick off the clients from the broker by
stopping the transport connector for the clients and then give bridges time
to drain the messages (i.e. forward them to other brokers).  However, this
most likely requires prior planning so that the transport connector for
clients and for network bridges are separated.  Otherwise, a separate tool
to bridge the messages would be needed.

Are the brokers in the network all stand-alone without H/A, or are they all
in master/slave clusters as well?  If master/slave, then upgrading the
master will result in the slave becoming active, reducing the delay in
delivering persistent messages.


--
View this message in context: http://activemq.2283324.n4.nabble.com/5-9-to-5-10-Upgrade-tp4686426p4686566.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.