TCP and ICMP soft-errors in IPv4/6 environments

[Pekka Savola <pekkas@netcore.fi>]

Hi,

(Targeted at Transport area, v6ops is there just to keep v6ops in the
loop, this has already been discussed there at some length..)

With dual-stack IPv4/6 deployments, when a TCP connection is tried to both
an IPv6 and IPv4 address, these are done in serial w/ a getaddrinfo loop
(in the typical case).  If the first destination is unreachable (e.g.
resulting in a host unreachable ICMPv6 message).  

The old host requirements document states that TCP MUST NOT abort
connections on such soft errors. However, this poses some problems in
current deployments where multiple addresses are tried sequentially, as
summarized above.

This is discussed more in section 2.3 of:

http://www.ietf.org/internet-drafts/draft-ietf-v6ops-v6onbydefault-00.txt

Has this been discussed in the transport area?  Is there something to be 
done about this, e.g. a BCP document discussing this practice and 
suggesting ICMP DU's are OK when establishing connections -- or something 
else?

Pekka
 Writing as v6ops co-chair

-- 
Pekka Savola                 "You each name yourselves king, yet the
Netcore Oy                    kingdom bleeds."
Systems. Networks. Security. -- George R.R. Martin: A Clash of Kings