ietf
[Top] [All Lists]

Re: one data point regarding native IPv6 support

2011-06-10 18:41:48

As for Geoff figures.  No one as far as I know has done measurements
of 6to4 failure of explictly enabled 6to4.  Yes Geoff figures seem
scary but if vendors were to make 6to4 users do something to turn
it on I truly think the failure rate would be in the noise.

Lots of the perceived issues with broken 6to4 go away with reasonable
application support for multi-homed servers.  Just yesterday I had
a dual homed site that was reachable over IPv4 but not over IPv6,
routing SNAFU in Canada.  Safari made me wait 30 odd seconds to
connect.  With Chrome I couldn't tell the site was not reachable
over IPv6.  This wasn't a 6to4 issue as neither end was running
6to4.

However it is this delay, which is completely avoidable on the
applications part, that is partially driving this push to move 6to4
to historic.

Also the kill 6to4 crowd push back on anything that would make 6to4
work better for those that can't get native IPv6, whether there is
a technical problem in the proposal or not.  If there is not a
technical problem then shut up.  You may not want to work on it but
others might.

Making 6to4 historic will also stop us working on things that make
6to4 behave better.  Whenever something is proposed in the future
you will hear the kill 6to4 crowd say "6to4 is historic, we can't
work on that."

Mark
-- 
Mark Andrews, ISC
1 Seymour St., Dundas Valley, NSW 2117, Australia
PHONE: +61 2 9871 4742                 INTERNET: marka(_at_)isc(_dot_)org
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www.ietf.org/mailman/listinfo/ietf