Mailing-List: contact users-help@activemq.apache.org; run by ezmlm
Precedence: bulk
Reply-To: users@activemq.apache.org
MIME-Version: 1.0
In-Reply-To: <1507283264168-0.post@n4.nabble.com>
References: <1506098972296-0.post@n4.nabble.com> <F9C244B6-AD50-44CC-B671-7569B23ACBF4@me.com>
 <56343299-B181-4408-BE07-BBEF48C6DA79@me.com> <A6DB18C5-CEF0-4B3C-9066-4B7346976B00@me.com>
 <1507112975601-0.post@n4.nabble.com> <CAKF+bspaFTw3SbEdYdpj+FDT5p4N22epDBVx0+z9URKhBsqjjA@mail.gmail.com>
 <1507283264168-0.post@n4.nabble.com>
From: Justin Bertram <jbertram@redhat.com>
Date: Fri, 6 Oct 2017 12:21:11 -0500
Message-ID: <CAF+kE=R6ZaJyv=8UvXpBLCg_Vf4AfmSHkNmf7pCfrTdsky4tMA@mail.gmail.com>
Subject: Re: 2 broker clusetr, both brokers are live
To: users@activemq.apache.org
Content-Type: multipart/alternative; boundary="94eb2c034a804510f5055ae41112"
archived-at: Fri, 06 Oct 2017 17:21:24 -0000

--94eb2c034a804510f5055ae41112
Content-Type: text/plain; charset="UTF-8"

> Please note that master and slave are on the same box.

Are they in different VMs on the same physical box?  In any event, this
sounds like it could be a problem with your environment if 2 broker
instances on the same box cannot maintain a network connection with each
other.

I assume this setup is just for testing as having the live and the backup
on the same physical machine (even if they are in separate VMs) wouldn't
provide much failure mitigation.  Can you confirm?


Justin

On Fri, Oct 6, 2017 at 4:47 AM, boris_snp <boris.godunov@spglobal.com>
wrote:

> Clebert,
>
> I've modified broker.xml for master and slave as follows:
>
>       <cluster-connections>
>          <cluster-connection name="my-cluster">
>             <address>jms</address>
>             <connector-ref>artemis</connector-ref>
>             <connection-ttl>120000</connection-ttl>
>             <reconnect-attempts>-1</reconnect-attempts>
>             <message-load-balancing>ON_DEMAND</message-load-balancing>
>             <max-hops>0</max-hops>
>             <discovery-group-ref discovery-group-name="dg-group1"/>
>          </cluster-connection>
>       </cluster-connections>
>
> and split brain still happening:
>
> Master 23:22:56,401: AMQ222092: Connection to the backup node failed,
> removing replication now
> Slave     23:24:29,380: AMQ212037: Connection failure has been detected:
> AMQ119011: Did not receive data from server for
> org.apache.activemq.artemis.core.remoting.impl.netty.
> NettyConnection@2eb88966[local=
> /11.203.148.100:47219, remote=arthost1/11.203.148.100:41410]
> [code=CONNECTION_TIMEDOUT]
> 23:25:17,667: AMQ221007: Server is now live
> Master 23:25:19,645: AMQ212034: There are more than one servers on the
> network broadcasting the same node id.
> Slave     23:25:19,660: AMQ212034: There are more than one servers on the
> network broadcasting the same node id.
>
> Please note that master and slave are on the same box. Another point,
> documentation states that reconnect-attempts is "The number of times the
> system will try to reconnect to a node in the cluster."
> Is a backup broker considered as "node"? Does reconnect-attempts apply to
> master-slave reconnection attempts?
>
> Will appreciate any help,
> Boris
>
>
>
>
> --
> Sent from: http://activemq.2283324.n4.nabble.com/ActiveMQ-User-
> f2341805.html
>

--94eb2c034a804510f5055ae41112--