|
|||||||||||
|
Re: [Hipsec-rg] RE: feedback from draft-hip-applications-02
From: Pekka Nikander <pekka.nikander(at)nomadiclab.com>
Date: Thu Mar 02 2006 - 23:39:00 EST Miika, Thanks for your thoughtful comments! >> The term LSI is not described in ESP or BEET drafts, yet we found
This double-binding should not be allowed. If there are two apps, the second binding attempt should fail with EISCONN or a similar error. If there is only one app, then the situation is slightly trickier since there exist legacy server apps that want to separately bind IPv4 and IPv6 socks for the same server port. In that case I would advice for a hack that allows the double binding but directs all incoming connections to the IPv6 sock. >> The LSI and HIT point to the same public key. In this case, the
I think the rule above should be sufficient to disambiguate the situation. >> There are two choices to overcome the first problem. Either you
As I argued above, both of these might be OK, depending of the situation. >> Jan convinced me that the latter alternative sounds better, just
See above. >> The second problem with use LSIs is that the underlying
I agree with Tom. This is a no-issue. >> The third problem is about referrals, a topic which you slightly
As argued in draft-laganier-ipv6-khi-01.txt, the plan is to eventually make HITs "routable", or at least universally mappable to locators. This is a topic of active research, as you know. >> As a consequence, it seems like opportunistic HIP is perhaps the
I agree with Tom. > Anyway, I think that we discuss this issue sufficiently without
I don't think we really can make a recommendation; what is best depends on so many aspects. If you would like to propose qualitative or quantitative text on the trade-offs, that would be excellent. >> Passing an IP address as a referral can be made also mobility
There seems to be some lurking issues w.r.t. RVSses, legacy, opportunistic HIP etc. Please propose text. --Pekka Hipsec-rg mailing list Hipsec-rg@honor.cybertrust.com http://honor.cybertrust.com/mailman/listinfo/hipsec-rg Received on Thu Mar 2 23:39:48 2006 This archive was generated by hypermail 2.1.8 : Wed Aug 23 2006 - 12:42:50 EDT |
||||||||||
|
|||||||||||