[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: draft-huitema-v6ops-teredo-03.txt



Margaret,

My concerns are based on the out come of one of the IETF meetings last year
when it was decided by the working group, and then confirmed by the mailing
list, that privet addresses (FEC0 etc) were "a bad thing" and were removed
from IPv6.

At the time I argued that private addresses would need to be used or we
would have another case of address space being co-opted like was eventually
done in RFC 1918. This long discussion on the list resulted after I
responded to a request for those private addresses during the meeting (which
I had not attended). After the flame war ended there was a vote in the WG
mailing list to depreciate private address spaces.

Eventually this lead to the various drafts about doing away with NAT (for
example draft-vandevelde-v6ops-nap-01).

My concerns with this teredo document is that it seems to indicate that
there may be a way to provide NAT on a private address space in IPv6. All I
am asking for is wording that shows that the NAT is still IPv4 and is not
IPv6 after the router is "upgraded" and not that there is a transition from
IPv4 to IPv6. My understanding is that this is a work around to connect IPv6
hosts to IPv4 hosts that "NATted". This is fine. It makes sense to me as
there are going to be IPv4 networks around for quite a while and the
interconnection is necessary. But please don't create false impressions that
private address spaces or NAT will exist in the IPv6 space, or change the WG
ruling and put private addresses and NAT back. I am sure that there are
people who would prefer it that way (mostly network administrators) and
those who don't want to bring them back (mostly hardware and software
developers).

All I want is a consistent message to go out from this WG as to the status
of private addresses and NAT.

Eric

----- Original Message ----- 
From: "Margaret Wasserman"
>
> Hi Eric,
>
> I am not quite sure how to respond to your post, because I am not
> sure that parts of it are applicable to Teredo at all...