I was working on this lab last night and my VPNv6 sessions dropped between the CSC customer carrier (r6,7,8). I think this happened to me before but I ran out of rack time so I saved my configs. Now I have loaded it up again and the peerings are all established again and I have fully reachability so I am confused as to what happened.
It seemed my pings were getting to the far side but the LSP on the return path was dodgy (that's what I saw before my session timed out). Has anyone had a simliar experience on this particular task? I was hoping to be able to troubleshoot it tonight.
ALso should something like this happen in the lab and I reload and it solves the problem there is no guarantee that the problem wont reoccur when the lab is being graded. Any one have any thoughts on how best to approach this?
Also another thought. I have deb ip rout and deb ipv6 rout on most of the time now cos I'm paranoid as this type of thing has happened with loops recently to try and catch it. It's almost like a sneaky script is trying to catch me out. The problem is across all unicast address families from time to time, no changes occur and all of a sudden something drops out. It's symptomatic of a bridging loops, a very slow one, building until it breaks the network.