[c-nsp] Troubleshooting dropped packets in a PPPoE session via L2TP tunnel

[Joe Freeman]

Greetings all-

I have a customer for whom we've built a Remote Access DSL network. We
recieve the PPPoE sessions from Bell, and using the authenticate before
forward feature, have radius setup to return attributes that cause our LNS
to forward that customer's session to them via another L2TP tunnel, with the
customer terminating the actual PPPoE session.

The path between my LNS is via an ATM DS3 to an ATM/Frame switch, ATM to
another switch, and frame from that switch to the customer.

The customer has a total of three frame relay circuits to which we deliver
these sessions. Each frame relay circuit has a different L2TP tunnel.

The customer is complaining about 'dropped pings' on sessions in one
specific tunnel. The sessions in the other tunnels appear to be fine. The
customer is of course blaming our network, since it worked fine when the
connections were PPPoA and switched in our ATM switch.

I've checked all the L1/L2 stats in all the switches... no discards, no
fecns/becns, no invalid/discarded cells, no sequence errors in the L2TP
tunnels.

Is there anything else I can check? I truly think the problem is in the
customer's network at this point, but I need to make sure.

Joe