I am managing over fifty Architect flows and I am now responsible for our disaster recovery strategy for our on-premise Edge servers. We currently use a single carrier for our BYOC-Premise trunks. I am concerned that a carrier outage or a local data center failure will take our entire voice network offline. I want to implement a multi-cloud trunking strategy where we have a backup carrier routed through a secondary cloud-based SBC. How do I configure the Architect flow to detect a trunk failure and automatically re-route the call to the secondary carrier without manual intervention?
Greetings. This is a critical security and compliance requirement. To achieve this, you should not rely on the Architect flow to detect the failure.
Instead, you should configure your Edge Groups with multiple trunks in a specific ‘Priority’ order. The Edge server will automatically attempt the primary trunk, and if it receives a SIP 503 or a timeout, it will immediately fail over to the secondary trunk.
This happens at the telephony layer and is completely transparent to your Architect flow logic.
That is correct. I always tell people to keep the routing logic as simple as possible. Let the hardware handle the failover.
One thing to check is your ‘Options’ ping interval on your backup trunks. You want the Edge to know the backup is alive before it needs it.
If your backup carrier is in another cloud, make sure your firewall rules allow the SIP traffic from all your Edge IPs. I have seen failover tests fail because someone forgot to whitelist the secondary SBC!