Keeping your IP address during failover: how same-IP failover works

Two circuits, automatic switchover – and the call still drops. The reason is almost always the public IP address.

Why sessions drop on failover

Every internet connection is reachable via a public IP address issued by the provider. A TCP connection, a VPN tunnel or a VoIP call is bound to that address. When a dual-WAN router switches from provider A to provider B, it suddenly sends with a different source address. To the remote end that is a foreign connection – it discards the packets and the session drops.

The everyday consequences

  • Active calls over cloud PBXs or Microsoft Teams end.
  • Site-to-site and client VPNs have to reconnect.
  • Remote desktop and ERP sessions are disconnected.
  • Services that only admit certain IP addresses (allow-lists) block access.

The principle of same-IP failover

The solution is to decouple the public IP address from the provider. With BigLeaf, all traffic is carried through tunnels over every circuit to gateway clusters in the Bigleaf Cloud Access Network. Only there does it exit to the internet – with a public IP from a Bigleaf IP block. Which circuit is currently carrying the traffic is invisible to the remote end. If one circuit fails, the flows move to another; the address stays the same.

Prerequisites and limits

  • At least two working, ideally independent circuits (different providers and media).
  • During rollout, the firewall WAN port and any IP allow-lists at partners are switched to the Bigleaf address once.
  • If all circuits fail at the same time, no technology can carry the traffic.

Try it: interactive failover demo – Details: Same-IP failover

Sources (retrieved 7 Nov 2026)

Ready for a more reliable internet connection?

Talk to our team about your sites and circuits. We show BigLeaf in a live demo and check how it fits into your existing infrastructure.