| Newsgroups |
gmane.ietf.mobike |
| Message-ID |
<[email protected]> |
Tero Kivinen writes:
>
> [email protected] writes:
>> Yes. This would work perfectly if the peer never uses the
>> src/dst addresses from the IP header for anything else except
>> transmitting the reply.
>
> I disagree on that.
Perhaps saying "works perfectly" is exaggarating slightly,
but using the latest request for path testing would certainly
be easier if the src/dst address from IP header was not
used for anything else than the reply.
>> However, the MOBIKE protocol will use them, in some cases at
>
> There is no MOBIKE protocol yet. Your MOPO might be using them,
> and it must be doing things that way, but do not generalize that
> to MOBIKE protocol which is not yet even defined.
This is not a feature of just MOPO, but any MOBIKE protocol that
works with NATs (and does not do Unilateral Self-Address Fixing).
And yes, I know we still have not formally decided whether the
MOBIKE protocol will work with NATs. But I don't think we should
even consider doing UNSAF, because that would certainly amount to
changing how IKEv2/IPsec NAT Traversal works.
Best regards,
Pasi