[6bone] Re: routing concern

[Michel Py]

> Petr Baudis wrote:

I think that Petr's post is a very good summary.

> This is wild world and natural processes mostly rule
> this world. Basically, I believe that unreliability of
> the IPv6 internet is generally caused by the fact that
> it does not run native, but tunnelled through IPv4.

It's not the only reason, though. The fact that the 6bone is free does
not help, and I have to say that I am not going to get up in the middle
of the night to troubleshoot a 6bone problem.

> And people tend to create peerings through tunnels even
> with peers they have poor latency to etc. It's not the
> 2001::/16 what saves you, it's the unwritten (?) rule
> that native links usually live inside 2001::/16 and
> tunnels inside 3ffe::/16.

This is very true.

Michel.

> As people continue with establishing of native IPv6
> links and peerings, the situation improves and the
> native peering usually tends to be much more stable and
> reliable than the peering through tunnels, especially
> when driven on some official base (and this is also
> another difference between IPv4 and IPv6, IPv4 peerings
> are protected by various contracts and agreements,
> being ran on commercial base; IPv6 usually aren't
> [altough there are obviously exceptions already]).
> And as the time goes on, people obviously tend to
> sacrifice the tunnel peerings for native ones and the
> reliability improves. The natural process. You can
> obviously push it a lot by preferring native peerings
> before the tunneled ones, and this should probably
> become another written unwritten rule.

> This has been said before already in few mails
> undirectly, this mail is meant generally as a summary
> of that view (not the only one here, obviously, just
> the one I agree with).