> > You could use something like an IPv6 NS message that is artificially > > inflated in size for this as we have discussed in the past. But, it's > > not > > a once-and-done deal; if you're going to be probing the MTU you need > > to do it periodically so that L2 path changes don't result in black > > holes. > > ND times out pretty quickly, right? Then this happens pretty much > automatically.
FWIW RFC 2461 does not have any timeout associated with the neighbor cache entries or destination cache entries. They can be garbage collected when the node runs low on memory. And stale link-layer addresses in the neighbor cache are detected when the link-layer address is used by triggering Neighbor Unreachability Detection. Thus if you'd want some mechanisms to peridically "reprobe" the path to the neighbor you'd have to explicitly add such a mechanism. Erik -------------------------------------------------------------------- IETF IPv6 working group mailing list [EMAIL PROTECTED] Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6 --------------------------------------------------------------------
