future plans and ideas


would be really nice if we could fail over tcp sessions...

on top of that, fail over BGP sessions!
BGP clusters with hitless failover
syncing BGP stuff shouldn't be too hard
lots of work... too bad real life and $dayjob leave little time