On Tue, 30 Sep 2008 06:44:45 -0400 Sam Hartman <hartmans-ietf@xxxxxxx> wrote: > >>>>> "Steven" == Steven M Bellovin <smb@xxxxxxxxxxxxxxx> writes: > > Steven> On Mon, 29 Sep 2008 15:20:23 -0400 > Steven> Sam Hartman <hartmans-ietf@xxxxxxx> wrote: > > > >> Section 8 proposes that AH is the mandatory-to-implement > >> security mechanism for this option. Do we still believe that > >> is appropriate given RFC 4301's move away from AH as a > >> mandatory-to-implement service? > > Steven> As discussed way back when, when we'd agreed on what > Steven> became 2401 et seq., the IETF preferred that security > Steven> labels be part of the credential, and hence implicit in > Steven> the security association, rather than being part of the > Steven> IPsec header. Now -- here, the use of IPsec is to protect > Steven> the security label, rather than the data -- but does it > Steven> actually do that in any useful way? If the security > Steven> option is really end-to-end, we don't need it, per the > Steven> above. If it's hop-by-hop -- and it's using a v6 > Steven> hop-by-hop option -- AH doesn't provide useful protection > Steven> unless packets are digitally signed rather than MACed. > > Doesn't that sort of depend on > 1) who the attackers are > 2) who the endpoints of the SA are? I'm not sure what you mean here. I think it's reasonable to postulate an active or semi-active attacker on a nominally-secure link -- think Operation Ivy Bell, or any of a number of CIA attacks on Soviet communications systems described in a new book "Spymaster". Such an attack could record packets and resend them with a different security label; in the absence of digital signatures, this could not be detected until the receiving site, which of course will never see it -- the destination address would be changed, too. > > As far as I can tell AH would be fine for hop-by-hop SAs to protect > packets flowing over a link that cannot be trusted to preserve > integrity between two intermediate systems. > > You could potentially have both an end-to-end SA and a hop-by-hop SA. > That says that you trust intermediate systems less than you do the > endpoints, but somehow you're still trusting them not to disclose > traffic. I'd like to understand the threat model that leads to this > better. "Need to know" -- intermediate systems may be cleared very high, but they have no need to see the packet contents. > > However, I agree with you that the draft is broken. It seems clearly > like it is talking about end-to-end AH, and I agree that doesn't fit > the model well at all without digital signatures. > That's my main point. > > >> As we know, BCP 61 requires a strong mandatory-to-implement > >> security mechanism for Internet protocols. That mechanism > >> needs to be sufficient for use over the open Internet. We have > >> been very reluctant to accept weaker security mechanisms > >> because some standards-track protocol is not intended for use > >> on the open Internet; BCP 61 says we have consensus that is not > >> how we do things. I question whether AH is sufficient as a BCP > >> 61 mandatory-to-implement mechanism. The entire point of this > >> IPV6 option is to limit disclosure of confidential and > >> classified information. In order to meet that security > >> objective on the open Internet, some confidentiality mechanism > >> is required. I strongly recommend that if this TS is published > >> on the standards track,ESP with confidentially be required. > > Steven> I don't agree. The purpose of AH in this spec is to > Steven> protect certain information that's in the clear. (As > Steven> noted, though, I don't think it does it properly.) > Steven> Protection of the underlying data is the business of a > Steven> separate layer or sublayer. > > I made a number of statements and I'd like to explore our > disagreement. I'm assuming that we agree about what BCP 61 is trying > to do: it's trying to require that IETF protocols have adequate > security for the open Internet, because networks will get connected. > I seem to recall you buy into that argument:-) Yes... > > Do you disagree with my assertion that from a overall architecture > view, anyone who implements this mechanism needs confidentiality to > run their packets over the open Internet? Yes. > > If you agree confidentiality is needed somewhere, how do we get > interoperability if we don't mandate a confidentiality mechanism here? It's a different layer. The security label doesn't require confidentiality; it does require integrity. > >> The document requires that if there is a label inside and > >> outside an AH header, that these labels must be the same. Why > >> is this a valid use of a 2119 MUST? What security or > >> interoperability problem results if for example the outer label > >> dominates the inner label? As far as I can tell this > >> requirement gets in the way of tunneling arbitrary IP traffic. > > Steven> It guards against someone tampering with the outer label, > Steven> causing the packet to be misrouted perhaps. > I guess I'm having a hard time seeing why you would have both labels > if they serve the same purpose. However I think I was misreading the > text. I thought the text was intended to apply to IP within IP--for > example an inner label in a ip-in-ip tunnel. If the text is only > intended to apply to two copies of the option at the same IP layer, > then I agree they need to be the same, although I'd also like an > explanation of why you ever do that. > The model is you send a packet over some secure internal networks. It reaches a site boundary, where it's encapsulated -- say, via IPsec. The security label should be preserved. Why? I always get nervous when there are two fields describing the same data; it's too easy to sneak something past if a filter checks one but the end system relies on the other. In this case, it might be the receiving TCP which uses the inner label, but the routers that use the outer label. Of course, thinking more, an encrypted packet can often have a different label, though for traffic analysis protection one still might want an outer label that selected, say, paths that used link encryptors. I think the conclusion is your point: the document needs to be clearer on this point. > > Steven> Note 7.3.1 on > Steven> TCP considerations. (Also note that 7.3.1 disagrees with > Steven> 793 on the treatment of security labels in section 3.6 of > Steven> 793. At the least, this shoudl be noted. > > I had completely missed this. I'll call out the section to the > transport ADs > I should have added: I think the new document is in fact more correct than 793 -- the 793 scheme would permit various forms of high-bandwidth covert channels to be set up. This is an issue that was not nearly that well understood when 793 was written. That said, it is a change to TCP, and needs to be treated as such. --Steve Bellovin, http://www.cs.columbia.edu/~smb _______________________________________________ Ietf@xxxxxxxx https://www.ietf.org/mailman/listinfo/ietf