Re: comments on draft-ike-mobike-design-03.txt
Shinta Sugimoto <[email protected]>
| Newsgroups | gmane.ietf.mobike |
|---|---|
| Message-ID | <[email protected]> |
On Mon, 29 Aug 2005 15:08:02 +0200 Francis Dupont <[email protected]> wrote: > In your previous mail you wrote: > > I read draft-ike-mobike-design-03.txt and had a few comments/questions > as below: > > Technical Comments: > > - Section 4, fourth paragraph, it says "Using this API, the MOBIKE > daemon can create entries in the Security Association (SAD) and > Security Policy Databases (SPD)." I believe "this API" refers to > PF_KEY which is initially designed as an interface to SAD. So, the > draft should be accurate in this respect, IMHO. > > => PF_KEY can't offically be used to manage the SPD so IMHO the > current wording should be kept. That's the point that I wanted to make. The current texts could be read in a way that both SADB and SPD can be accessed by the API (PF_KEY). Figure 3 shows an example protocol interaction between a pair of MOBIKE peers. MOBIKE interacts with the IPsec engine using the PF_KEY API [RFC2367]. Using this API, the MOBIKE daemon can create entries in the Security Association (SAD) and Security Policy Databases (SPD). So just thought that it would be bettter to say: Figure 3 shows an example protocol interaction between a pair of MOBIKE peers. MOBIKE interacts with the IPsec engine using certain API. PF_KEY [RFC2367] can be used by the MOBIKE daemon to access the Security Association Database (SAD). Regards, Shinta > > - Section 5.8 (Scope of SA changes), I think this is very important > issue. Separate management of IKE SA and IPsec SA in terms of address > update allows us more flexible and (potentially) user-friendly > operation although there will be more complexity and overhead to > do so. > > => I agree on both: it is an important issue and the current decision is bad. > > Regards > > [email protected]