Have you looked at the network health check
https://activemq.apache.org/components/artemis/documentation/latest/network-isolation.html If you cannot add more nodes, but are getting split brain issues, this can help reduce issues. Basically you set some remote network ping addresses (loadbalancer, firewall some core switch) ideally a couple of hops away, and should a node not be able to ping them, it will assumes it is the isolated node and will stop it from promoting itself. Get Outlook for Android On Mon, Jul 1, 2019 at 5:50 AM +0100, "yw yw" <wy96...@gmail.com> wrote: Do you mean you would use jdbc storage and ha replicaiton? If so, the split-brain issue are still there IMO since storage type is not related to vote mechanism which checks whether node is alive. blue_nekton 于2019年6月29日周六 上午12:46写道: > Hi All, > > We are noticing split-brain issues (Network isolation) in Single pair HA > Replication. Due to limitations, we will not be able to use multiple > pairs. > > So I am exploring the JDBC storage option for the Artemis High availability > (HA) with a single pair of master and slave. > > Have anyone used jdbc with single pair HA successfully? did you face any > split-brain issue in past? > please advise. I will also try disrupt the network connections between > brokers and DB and analysis. > > Thanks > > > > -- > Sent from: > http://activemq.2283324.n4.nabble.com/ActiveMQ-User-f2341805.html >