|High Availability / FAQs for HA Decisions|
The Virtual IP (VIP) is the address that represents the public network identity of the cluster.
The Virtual IP (VIP) is the address that represents the overall cluster and is intended to be used for all third-party system and device integrations. The VIP is the only address that should be shared with other systems that require integration with Vocera Platform, since it represents the public network identity of the cluster.
For an Ethernet Layer 2 Adjacent deployment, the VIP is normally assigned to the master node within a cluster. During a failover event, the VIP is assigned to another node that is taking over the master role at that time. The VIP is not intended to be permanently and irrevocably assigned to an individual server while in a cluster, but rather it is managed by the cluster; it "floats" from one node to another during failover events. In addition to the VIP, each node in the cluster also has its own unique IP address, different from the VIP, that is assigned permanently to that node. While a node is a member of a cluster, it will use the private IP address to communicate with the other nodes in the cluster in order to maintain the overall cluster function and arbitrate role assignment during failover events. Reassignment of the VIP from one server to another during a failover is an automatic action carried out by the cluster itself and does not require user intervention.
For a Non-Layer 2 Adjacent type deployment, the VIP is assigned to the ADC. The ADC itself can be deployed as either an individual server, or as its own cluster of nodes to provide robustness for the ADC services. In this scenario, the VIP is managed externally by the ADC with the local customer's HA policy, and the Vocera Platform nodes will only have their private IP assignment, regardless of their current role. In addition to managing the VIP, the ADC will redirect network connections from third-party integration systems from itself to the master role in the Vocera Platform cluster for processing. The ADC is able to determine the master role assignment by monitoring all nodes in the cluster, as the node with the master role assignment will return a positive indication. During a failover event, the master node stops sending a positive indicator, then another node will take on the master role, and then transition its monitor return value to positive. At that time, the ADC will transition its network connections from the failed master to the new master node, completing the failover event.