A server pool group consists of two server pools. You can manually switch the active server pool for traffic forwarding, or configure a health monitor for each server pool to enable automatic server pool switchover. This mode is commonly used for cross-cluster disaster recovery of virtual services. You can configure the corresponding server pools in the primary and secondary clusters, respectively. When a cluster failure occurs and a disaster recovery failover is triggered, you can choose to trigger a server pool switchover at the same time, or choose to keep the current server pool to provide greater configuration flexibility.
The server pool switchover logic is as follows:
When all members in the active server pool remain unhealthy for 1 minute continuously, the system triggers an automatic server pool switchover.
When a server pool switches frequently within a short period, all server pools are considered unhealthy, and the system stops switching the active server pool.
When a server pool recovers to a healthy state, the system automatically selects a healthy server pool as the active server pool for the virtual service. If both server pools are healthy, the primary server pool is preferred as the active server pool.
In a server pool group, server pool member modes support only load balancing mode and primary-secondary mode; multiple primaries–one secondary mode is not supported.
