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: <CAKF+bsqtScsdhTD-34+3yZAgmMR8Rkgnj7OpzKkxRBksvo=Bhw@mail.gmail.com>
References: <1504517355395-0.post@n4.nabble.com> <CAKF+bsqtScsdhTD-34+3yZAgmMR8Rkgnj7OpzKkxRBksvo=Bhw@mail.gmail.com>
From: Justin Bertram <jbertram@redhat.com>
Date: Thu, 7 Sep 2017 07:50:48 -0500
Message-ID: <CAF+kE=RY82563Ujp_jncyWWXyOPwzi77YAjppbpKuWZxqcvFSg@mail.gmail.com>
Subject: Re: Artemis HA cluster with replication and two slaves results in
 broken/none backup
To: users@activemq.apache.org
Content-Type: multipart/alternative; boundary="001a113cd3fcde2664055898e8bf"
archived-at: Thu, 07 Sep 2017 12:51:04 -0000

--001a113cd3fcde2664055898e8bf
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

I believe this has already been filed in JIRA [1].


Justin

[1] https://issues.apache.org/jira/browse/ARTEMIS-1285

On Wed, Sep 6, 2017 at 10:43 PM, Clebert Suconic <clebert.suconic@gmail.com=
>
wrote:

> I am not so sure multiple backup would work with shared nothing.
> I have seen that working with shared storage.
>
> We could work with you on that. I'm not sure what would take thought.
>
> On Mon, Sep 4, 2017 at 5:29 AM martk <123martin@web.de> wrote:
>
> > Hi everyone,
> >
> > I am using Artemis 2.2.0 and HA with replication and three brokers: one
> > master and two slaves (the configuration extracts, named
> > master/slave1/slave2, are below).
> >
> > Running the following scenario I am not sure if it is a bug or works as
> > designed:
> >
> > 1) start master
> > 2) start slave1 (logging =E2=80=9C=E2=80=A6AMQ221031: backup announced=
=E2=80=9D)
> > 3) start slave2 (logging =E2=80=9C=E2=80=A6waiting live to fail before =
it gets active=E2=80=9D)
> > 4) kill slave1
> >
> > The master recognize the failure and shows =E2=80=9CAMQ222092: Connecti=
on to the
> > backup node failed, removing replication now=E2=80=9D in his log. I wou=
ld assume
> > slave2 to become automatically the backup of the master but it does
> > nothing.
> >
> > To manually solve this behavior I have tested two alternatives (without
> > restarting the master or having a downtime):
> > A) restart slave2 (yes, not slave1) and then it becomes the backup of t=
he
> > master
> > B) restart slave1 and everything is as before the failure
> >
> > Regarding A the current behavior seems to be a bug. Or is there any
> > configuration to set?
> >
> >
> >
> > *Extract of master-config*
> >
> > <ha-policy>
> >    <replication>
> >       <master>
> >          <check-for-live-server>true</check-for-live-server>
> >       </master>
> >    </replication>
> > </ha-policy>
> >
> > <connectors>
> >    <connector name=3D"netty-connector">tcp://192.168.0.10:61616
> </connector>
> >    <connector
> > name=3D"netty-backup-connector-slave1">tcp://192.168.0.11:61616
> </connector>
> >    <connector
> > name=3D"netty-backup-connector-slave2">tcp://192.168.0.12:61616
> </connector>
> > </connectors>
> >
> > <acceptors>
> >    <acceptor name=3D"netty-acceptor">tcp://0.0.0.0:61616</acceptor>
> > </acceptors>
> >
> > <cluster-connections>
> >    <cluster-connection name=3D"cluster">
> >       <connector-ref>netty-connector</connector-ref>
> >       <static-connectors allow-direct-connections-only=3D"true">
> >          <connector-ref>netty-backup-connector-slave1</connector-ref>
> >          <connector-ref>netty-backup-connector-slave2</connector-ref>
> >       </static-connectors>
> >    </cluster-connection>
> > </cluster-connections>
> >
> >
> > *Extract of slave1-config*
> >
> > <ha-policy>
> >    <replication>
> >       <slave>
> >          <allow-failback>false</allow-failback>
> >       </slave>
> >    </replication>
> > </ha-policy>
> >
> > <connectors>
> >    <connector
> > name=3D"netty-live-connector">tcp://192.168.0.10:61616</connector>
> >    <connector name=3D"netty-connector">tcp://192.168.0.11:61616
> </connector>
> >    <connector
> > name=3D"netty-backup-connector-slave2">tcp://192.168.0.12:61616
> </connector>
> > </connectors>
> >
> > <acceptors>
> >    <acceptor name=3D"netty-acceptor">tcp://0.0.0.0:61616</acceptor>
> > </acceptors>
> >
> > <cluster-connections>
> >    <cluster-connection name=3D"cluster">
> >       <connector-ref>netty-connector</connector-ref>
> >       <static-connectors allow-direct-connections-only=3D"true">
> >          <connector-ref>netty-live-connector</connector-ref>
> >          <connector-ref>netty-backup-connector-slave2</connector-ref>
> >       </static-connectors>
> >    </cluster-connection>
> > </cluster-connections>
> >
> >
> > *Extract of slave2-config*
> >
> > <ha-policy>
> >    <replication>
> >       <slave>
> >          <allow-failback>false</allow-failback>
> >       </slave>
> >    </replication>
> > </ha-policy>
> >
> > <connectors>
> >    <connector
> > name=3D"netty-live-connector">tcp://192.168.0.10:61616</connector>
> >    <connector
> > name=3D"netty-backup-connector-slave1">tcp://192.168.0.11:61616
> </connector>
> >    <connector name=3D"netty-connector">tcp://192.168.0.12:61616
> </connector>
> > </connectors>
> >
> > <acceptors>
> >    <acceptor name=3D"netty-acceptor">tcp://0.0.0.0:61616</acceptor>
> > </acceptors>
> >
> > <cluster-connections>
> >    <cluster-connection name=3D"cluster">
> >       <connector-ref>netty-connector</connector-ref>
> >       <static-connectors allow-direct-connections-only=3D"true">
> >          <connector-ref>netty-live-connector</connector-ref>
> >          <connector-ref>netty-backup-connector-slave1</connector-ref>
> >       </static-connectors>
> >    </cluster-connection>
> > </cluster-connections>
> >
> >
> >
> >
> >
> > --
> > Sent from:
> > http://activemq.2283324.n4.nabble.com/ActiveMQ-User-f2341805.html
> >
> --
> Clebert Suconic
>

--001a113cd3fcde2664055898e8bf--