Bad IPv6 connectivity or why not to announce more specifics (Was: IPv6 news)

Fri Oct 14 18:37:58 UTC 2005

On Fri, Oct 14, 2005 at 02:15:20PM -0400, Valdis.Kletnieks at vt.edu wrote:
> Interesting. :)  That's the first one I've seen that a sprintv6.net address isn't at
> hop number 3 or so (indicating that the person is basically directly connected to
> sprintv6.net) and also doesn't take a loop through Japan....
> 

	We (AS2914) stopped adding tunnel peers awhile ago, and only add
tunnels for customers at this point (i think).  As a tech guy, I prefer
people to connect to us native instead of any other method, it
removes all the sillyness i've seen (including v6 p-mtu issues).
It keeps the packets moving the best path and not cloaking the underlying
topology so people can find the trouble.  Same reason MPLS networks
don't always turn on/off ttl-decrement when following a (TE) tunnel.
It also makes it easier to troubleshoot and speculate where
troubles are ;-).

	- jared

-- 
Jared Mauch  | pgp key available via finger from jared at puck.nether.net
clue++;      | http://puck.nether.net/~jared/  My statements are only mine.