I recently switch from Site to Site EZVPN connection to a fiber connection for my remote site back to the home network and had some issues getting the route learned through EIGRP instead of that EZVPN site to site.
Now that's all working, but now when users on our AnyConnect VPN try to access machines on that remote network they can't. When I connect to the vpn then try to ping the remote network, pings drop. When I traceroute it traces out to the internet like it doesn't recognize the route.
Packet tracer doesn't help because both ways I put the addresses in it says ALLOW. packet-tracer input inside icmp <VPN address> 8 0 <remote network switch address> and packet-tracer input inside icmp <remote network switch address> 8 0 <VPN address>. I even tried it using DMZ (since our VPN is in our DMZ) as the source and I still get allow both ways.
The main firewall is learning the route correctly as far as I can see: Firewall# sh eigrp topology | inc <remote network>
P <remote network> 255.255.255.0, 1 successors, FD is 3584
I'm not sure why it's not pinging out right. Any help is appreciated.
If I run a sh ip route on the remote site's switch I can see my computer on VPN's IP address. But I can't ping it. And when I try to run a traceroute it hits my home location's switch but then stops there.
Testing from another remote site that's been connected for a while, I can't ping back to a VPN address either but everything else works fine. I'm trying to devise a test to help me nail down what the issue is, i.e. use packet tracer with port 3389 since RDP seems to be the thing VPN from one remote site can do and the other can't. But packet tracer has been pretty useless on this front.
Found the solution, shows how long since I've had to deal with the AnyConnect VPN. I had to add the remote site back into my split tunnel acl.