Region 1 going down and its impact

We have a set up where we have 2 regions with 9 Node setup in each region. The traffic is routed through an external load balancer and Region 2 will come into action ONLY if Region 1 is completely dead (example: due to natural disaster). Edge data replication does across the regions in active/active fashion.

If the Region 1 goes down, I assume there wouldn't be any impact to run traffic traffic as load balancer will detect this failure and route the traffic accordingly to region 2. However, some of the Apigee components like Cassandra and Zookeeper relies on quorum. How this would effect the disaster recovery. Wouldn't the zookeeper nodes needs to be reconfigured as the # of nodes of electing a leader wouldn't meet the requirement upon region 1 going down. Any help would be highly appreciated on the impact when all the Apigee components in Region 1 goes down abruptly.

0 0 172
0 REPLIES 0