You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@activemq.apache.org by "Enumudi Venkatesha, Srinivas (Nokia - IN/Bangalore)" <sr...@nokia.com> on 2020/04/16 14:56:06 UTC

Any Solution for Activemq Split brain for Master Slave

I have gone through the below page -

https://activemq.apache.org/components/artemis/documentation/latest/network-isolation.html

But, how to use this in Wildfly Application Server's Artemis activemq. Currently we have a master - slave configuration running and this scenario is occurring every month. Based on above page, Network pinger is only option available when it is of only 2 nodes.

MASTER -
<replication-master cluster-name="my-cluster" group-name="queues-cluster-group" check-for-live-server="true"/>

SLAVE -
<replication-slave cluster-name="my-cluster" group-name="queues-cluster-group" allow-failback="false" max-saved-replicated-journal-size="1"/>

Configuration for Wildfly standalone-full-ha.xml is required and help is really appreciated as we are facing this in production environment every month.

Thanks,
Srinivas

Re: Any Solution for Activemq Split brain for Master Slave

Posted by Justin Bertram <jb...@apache.org>.
The best way to mitigate split brain if you have just 2 nodes is to use
shared storage (e.g. SAN or NAS) rather than replication. In my opinion, if
you use replication with just 2 nodes you're asking for trouble, even with
the network pinger.


Justin

On Thu, Apr 16, 2020 at 9:56 AM Enumudi Venkatesha, Srinivas (Nokia -
IN/Bangalore) <sr...@nokia.com> wrote:

> I have gone through the below page -
>
>
> https://activemq.apache.org/components/artemis/documentation/latest/network-isolation.html
>
> But, how to use this in Wildfly Application Server's Artemis activemq.
> Currently we have a master - slave configuration running and this scenario
> is occurring every month. Based on above page, Network pinger is only
> option available when it is of only 2 nodes.
>
> MASTER -
> <replication-master cluster-name="my-cluster"
> group-name="queues-cluster-group" check-for-live-server="true"/>
>
> SLAVE -
> <replication-slave cluster-name="my-cluster"
> group-name="queues-cluster-group" allow-failback="false"
> max-saved-replicated-journal-size="1"/>
>
> Configuration for Wildfly standalone-full-ha.xml is required and help is
> really appreciated as we are facing this in production environment every
> month.
>
> Thanks,
> Srinivas
>
>
>