Re: [v6ops] new draft: draft-ietf-v6ops-clatip

On Tue, May 13, 2014 at 11:07 PM, Alejandro Acosta
<alejandroacostaalamo@gmail.com> wrote:
> Hi,
>  Sorry if this question have been done before (and I guess it have been).
>  I'm know this kind of mechanism are oriented to simple devices such
> smartphones, but anyway I wonder how the prefix 192.0.0.0/29 is handle
> in a device with more than one interface.
>
>  I read in section three you mention:
>
> "   ..... It is important
>    that a host never be deployed with 2 active IPv4 continuity solutions
>    simultaneously in a way that would cause a node to have overlapping
>    address from 192.0.0.0/29."
>
>  Can I subnet this prefix in two /30 and then there won't be
> overlapping?, should I assign one /29 and then I can not used
> ds-lite/464xlat in the next interface (this is what I understand from
> the text above).
>
> Thanks,
>

Hi Alejandro,

Thanks for taking the time to read the I-D.  The goal of the I-D is
lift restrictions.  David Thaler suggested that since the 192.0.0.0/29
has fewer restriction on its use, it is possible for solutions to
overlap and cause problems on a host that has both solutions.  For
example, if a host has both a DS-lite and 464XLAT capability, and both
are active at the same time and both are assuming the full
192.0.0.0/29 can be used, this will cause a problem since the host
will have 2 connected routes to the same place.

To avoid this scenario of overlapping space (in the unlikely event a
host has 2 active IPv6 transition scenarios), the text you quoted was
added.  The text simply says "be smart" ... don't use overlapping
addresses on a host and cause yourself a problem.

That said, i find it perfectly reasonable, as you suggest, to take the
192.0.0.0/29 and create 2 /30s, 1 for DS-lite and the other for
464XLAT... if you must have this ability.  The common Android solution
for this is to just use 192.0.0.4/32, only 1 IPv4 address is needed
for 464XLAT to function on Android

https://android.googlesource.com/platform/external/android-clat/+/master/clatd.conf

>
> El 5/6/2014 8:15 AM, Fred Baker escribió:
>>
>> A new draft has been posted, at http://tools.ietf.org/html/draft-ietf-v6ops-clatip. Please take a look at it and comment.
>>
>> _______________________________________________
>> v6ops mailing list
>> v6ops@ietf.org
>> https://www.ietf.org/mailman/listinfo/v6ops
>>
>
> _______________________________________________
> v6ops mailing list
> v6ops@ietf.org
> https://www.ietf.org/mailman/listinfo/v6ops