artemis colocated ha-policy

["jigar.parekh80" <ji...@gmail.com>]

currently I am evaluating HA-Policy for artemis and came across strange
situation where slave nodes goes missing after simple restart, below
configuration


    <ha-policy>
     <shared-store>
        <colocated>
           <request-backup>true</request-backup>
           <max-backups>1</max-backups>
           <master>
            <failover-on-shutdown>true</failover-on-shutdown>
           </master>
           <slave>
           <failover-on-shutdown>true</failover-on-shutdown>
                       <restart-backup>true</restart-backup>
           <allow-failback>true</allow-failback>
           </slave>
        </colocated>
     </shared-store>
  </ha-policy>

i follow below steps to evaluate

1) start both servers and verify both servers has master as active and
backup of each as passive
shutdown server 1, verify server 2 has master and colocated backup both
active
start server 1, verify fail-back occurred for server 1 so server has active
master
server 2 has master as active and backup of server 1 as passive
but during this process when server 1 is up it does not run any backup it
has only master active no new request for backup generated from server 2 and
it looses backup completely

not sure what could be wrong in this configuration as this could be very
basic scenario where i expect that after server restart each server have one
backup running.

I am using artemis 2.6.2 version




--
Sent from: http://activemq.2283324.n4.nabble.com/ActiveMQ-User-f2341805.html

Re: artemis colocated ha-policy

["jigar.parekh80" <ji...@gmail.com>]

Initially i guess so that there must be some voting issue, but there are few
other observations 
when i start cluster of two node it both nodes are stable with backup for
each other, problem appears after one of server gets restarted. 
The node which is up and running and not restarted does not request for
backup again to restarted node. 
Where as node which is restarted is working properly with failback. 






--
Sent from: http://activemq.2283324.n4.nabble.com/ActiveMQ-User-f2341805.html

Re: artemis colocated ha-policy

[Clebert Suconic <cl...@gmail.com>]

Its probably something around quorum. If you have a single node there is a
consensus algorithm to decide if it should activate it or not.


You could try using the network checker as some of the quorum will be
bypassed with a single node.  As it will validate if the network is up or
not before the activation check.

On Fri, Aug 31, 2018 at 2:43 AM jigar.parekh80 <ji...@gmail.com>
wrote:

> not having backup available after restart bug in colocated configuration?
>
>
>
> --
> Sent from:
> http://activemq.2283324.n4.nabble.com/ActiveMQ-User-f2341805.html
>
-- 
Clebert Suconic

Re: artemis colocated ha-policy

["jigar.parekh80" <ji...@gmail.com>]

not having backup available after restart bug in colocated configuration? 



--
Sent from: http://activemq.2283324.n4.nabble.com/ActiveMQ-User-f2341805.html