I have to ask for your help on this one:
Client has a cluster running R80.40.
Connected to the peer's network via IPSec using VTIs.
Despite being provided with single IP address for our side of the tunnel(s), TAC recommended using /29 network with vIPs to assign the tunnel IP address, claiming that the peer should not be concerned about it, since they will only see vIP.
I have seen this approach used for AWS VPN connectivity with Static Routes, but the IPs for VTIs were generated by AWS.
In the BGP via IPSec implementation guide for AWS there are no references to this approach.
Tunnel IPs on both sides are in 10.x.x.x range.
There is also a static route for the 10.0.0.0/8 pointing to the internal gateway on the cluster.
VPN is established.
We can see the 19X.XXX.XXX.0/24 networks advertised by the peer via BGP.
But in the routing table, the peer's network have the same next hop as the one defined for the 10.0.0.0/8.
I have never seen VTIs used as the cluster interfaces with vIPs, so please confirm that this is acceptable.
I would also appreciate the pointers for the reason the BGP routes having incorrect next hop.
Thank you.