I’m wondering if there’s a way that I can improve how long it takes for a route to be pushed to the OpenVPN profile when a new route is introduced, but I want to clarify what we do currently for context.
Our current setup has a Fusion Hub VA hosted in our AWS VPC and all routers in the field have a PepVPN profile administered from IC2 (also hosted in AWS VPC) where the Fusion Hub is the VPN host.
Initially I had the OpenVPN profiles setup on the Fusion Hub and that was working for all of our PepVPN traffic, however, we have private cell networks as well which terminate to a Cisco firewall at our corporate office, and that traffic wasn’t accessible from OpenVPN when hosted from the Fusion Hub, so I moved the profiles to the Office LAN router which is a Balance 1350. Since then, OpenVPN can talk to all private cell traffic as well as all PepVPN profiles as desired.
I believe we had continued routing issues when the OpenVPN profile was hosted by the Fusion Hub because the Fusion Hub has a private IP Address as its WAN IP (not sure if this is normal or not).
My question stems from our operations team, which noticed that when a router is offline for an extended period of time (I believe after 12 hours down), then the route gets removed from the OpenVPN profile, so when that router comes back online we are unable to talk to it right away through OpenVPN because we (presumably) have to wait for OVPN to get the new route. Of course, the router can be reached via IC2.
Is there something I can change to force that route to update right away? Perhaps I just need to add a static route on the Balance 1350 that encapsulates the range of the LAN IP of the routers in the field and that will keep those networks accessible? Or is the TRUE fix to address the private WAN IP on the Fusion Hub because the OVPN profile is best hosted from the Fusion Hub?
Any insight is appreciated.