Brian,
On Aug 12, 2011, at 10:53 AM, Brian E Carpenter wrote:
> But I think it's orthogonal to the
> *operational* question of how to eliminate the residual A6
> records and how to eventually eliminate A6 queries.
If you can figure this out, perhaps the lesson can be applied to iqueries and
queries to root servers that changed IP addresses over a decade ago.
> Mark is
> completely correct that this would require a plan - but since
> it's a purely operational matter, presumably it belongs in DNSOP
> if it belongs anywhere in the IETF.
I believe the issue is folks not upgrading name servers. Mark can correct me if
I'm wrong, but the reason there are so many A6 queries is because back in the
late 90s when BINDv9 was being written, A6 was going to be the future. As a
result, A6 queries were tried by default (I think they were actually tried
before AAAA in BIND 9.{0,1,2} but not positive). When it was pointed out that
A6 had some fundamental issues (discussed in
http://tools.ietf.org/html/draft-ietf-dnsext-aaaa-a6-01 with the deprecation
codified in RFC 3363), A6 was abandoned in favor of AAAA and BINDv9 no longer
tried A6 records by default. I believe A6 was removed from the query order in
2004 or 2005.
> Marking the RFC as Historic would only be the first step, apparently.
Wouldn't hurt, but I doubt any sort of RFC action will have a significant
impact. We're in the long tail of old software getting flushed out of the
Internet.
Regards,
-drc
--------------------------------------------------------------------
IETF IPv6 working group mailing list
[email protected]
Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6
--------------------------------------------------------------------
