[sunset4] New draft: draft-yourtchenko-ipv6-disable-ipv4-proxyarp-00

[Andrew Yourtchenko]

Hi all,

we've posted a new draft:

http://tools.ietf.org/html/draft-yourtchenko-ipv6-disable-ipv4-proxyarp-00

This could be an intermediate step before draft-perreault-sunset4-noipv4 is
widely implemented, which is simple enough to implement and does not
require coordination between the multiple parties on the network.

would be interesting to hear your feedback.

--a

[Simon Perreault]

Post by Andrew Yourtchenko
http://tools.ietf.org/html/draft-yourtchenko-ipv6-disable-ipv4-proxyarp-00
This could be an intermediate step before draft-perreault-sunset4-noipv4
is widely implemented, which is simple enough to implement and does not
require coordination between the multiple parties on the network.

Interesting...

Do you know what real-world OSes implement the "ARP for everything"
behaviour by default?

Thanks,
Simon

[Tillmann Karras]

Post by Simon Perreault
Interesting...
Do you know what real-world OSes implement the "ARP for everything"
behaviour by default?
Thanks,
Simon

RFC 3927 Introduction: "Microsoft Windows 98 (and later) and Mac OS 8.5
(and later) already support this capability."

According to [0], it is enabled by default at least on Windows 98
through Server 2003 and I have seen these addresses on our network
(apparently, not everyone considers them to be non-routable...)

Regards,
Tillmann


[0] http://support.microsoft.com/kb/220874

PS: Sorry Simon, I forgot to use reply-to-all.

[Andrew Yourtchenko]

Post by Tillmann Karras

Post by Simon Perreault
Interesting...
Do you know what real-world OSes implement the "ARP for everything"
behaviour by default?
Thanks,
Simon

RFC 3927 Introduction: "Microsoft Windows 98 (and later) and Mac OS 8.5
(and later) already support this capability."
According to [0], it is enabled by default at least on Windows 98 through
Server 2003 and I have seen these addresses on our network (apparently, not
everyone considers them to be non-routable...)

Originally I had noticed this behaviour on the iPhone/iPad on the networks
I dealt with. I can also confirm MacOS X Lion for sure, but can not say
much about Win* - but given the original RFC authorship, highly likely.

Would be very interesting to know about Android/Linux behavior - I did not
do much tests with it; though I had a feeling that Linux did not do "ARP
for everything" - take the latter with about 30% worthiness, I did not
spend any time to doublecheck before typing this, it can be
incorrect/outdated.

--a

Post by Tillmann Karras
Regards,
Tillmann
[0] http://support.microsoft.com/**kb/220874<http://support.microsoft.com/kb/220874>
PS: Sorry Simon, I forgot to use reply-to-all.
______________________________**_________________
sunset4 mailing list
https://www.ietf.org/mailman/**listinfo/sunset4<https://www.ietf.org/mailman/listinfo/sunset4>

[Simon Perreault]

Post by Andrew Yourtchenko
Would be very interesting to know about Android/Linux behavior - I did
not do much tests with it; though I had a feeling that Linux did not do
"ARP for everything" - take the latter with about 30% worthiness, I did
not spend any time to doublecheck before typing this, it can be
incorrect/outdated.

It's a per-distro thing. The distro I use most often (Fedora) does not
use 169.254/16 addresses by default. I don't know about Android.

This problem will be added to the gap analysis draft.

Simon