We have a number of cases found that when the network card going offline, failover between cluster brokere is not happening. We have a cluster configuration setup in accordance with best practice, where broker01 running on a machine and broker01bakcup running on another. Network running CAA are connected directly. CAA Failover works fine when broker01 is reload or stopped.
My question is: why is this happening?
For more information see attached image of the configuration.
PT systemtegning v3.1-CAA.jpg
Please refer to the MQ tuning guide, section tcp tuning to allow for faster/timely failover.