I've inherited an MPLS network, and I've not worked with MPLS before. The more I read, the more confused I am about the network I have.
I have a dozen sites, all connected to a Bellsouth (now AT&T) MPLS network. Each site has a T1 circuit, and the circuit is configured as frame-relay on each sites router. I understand this part OK.
Now to the part I don't understand. Each remote site is connected back to the main site through VPN tunnels. So, each remote site has one VPN tunnel back to the main site, and the main site has a dozen tunnels, one to each remote site. Then, each site has a dozen static routes, all pointing the their local router's WIC IP address.
Is this a typical MPLS setup? What I thought what MPLS did was not unlike what a traditional frame-relay network did; the customer's routers handed the WAN traffic to the provider's network, which through various pieces of routing magic delivered the packets to the customer's routers on the opposite end. No need to set up VPN tunnels, and you use a IGP to handle the routing.
Each frame-relay interface is point-to-point, with the IP address part of a /30 network, with one end being the local router, and the other end being the AT&T router. Is this also typical?
Most MPLS documentation seems to be geared towards how the provider sets up their network, not how the customer sets up his part. Any help would be greatly appreciated, as I have absolutely no documentation on how this setup came to be. Thanks.