As part of the GCP migration of our hosted VMware Edge Cloud Orchestrators, a Disaster Recovery (DR) synchronization is required from the Active Orchestrator running on AWS to the Standby Orchestrator running on GCP prior to the actual cutover. During this step, some Edges may not successfully deliver heartbeat messages to the Standby Orchestrator. A cutover is only regarded as successful when all active Edges are heartbeating to the Standby Orchestrator, because the Standby Orchestrator cannot be promoted to Active until this occurs. If one or more Edges fail to successfully heartbeat with the Standby Orchestrator, this may require the Broadcom team to perform a manual recovery of those Edges after the cutover to reestablish heartbeats to the DR Standby GCP Orchestrator.
VMware VeloCloud SD-WAN
Orchestrators on GCP use a different URL from the original Orchestrator on AWS.
In some circumstances, the Edge may not be able to properly identify and route DNS requests made by the Edge’s management plane. This would result in the Standby Orchestrator not being able to receive heartbeat messages from the affected Edges.
The Broadcom team will implement backend updates to adjust the DNS references. During this time, you may notice configuration updates being pushed to the Edges. However, there would be no impact on user traffic. The backend changes will be reverted once the DNS updates are activated, and we expect the entire process to take less than 10 minutes. This process may be repeated during the cutover of the Active Orchestrator on AWS to the Standby Orchestrator on GCP. The Broadcom team will contact you if further assistance is needed. Throughout this process, communication between the VeloCloud Edges and the Active Orchestrator would remain unaffected. No action is required from the customer during the DNS reference update process.