One additional comment: The sec-dir reviewer said that he thought a provision for authentication MUST be present and a provision for data confidentiality service SHOULD be present. I went with his recommendation on the MUST, SHOULD text.
Hope that is OK with you. -- Maureen Maureen Stillman Nokia Services and Software Mobile: (607)229-3358 -----Original Message----- From: ext Ben Campbell [mailto:[EMAIL PROTECTED] Sent: Monday, April 28, 2008 5:34 PM To: Stillman Maureen (Nokia-S&S/MtView) Cc: [email protected]; Sengodan Senthil (NSN - US/Irving); [EMAIL PROTECTED]; [EMAIL PROTECTED]; Gopal Ram (NSN - US/San Diego); [EMAIL PROTECTED]; [EMAIL PROTECTED] Subject: Re: Gen-ART Last Call review of draft-ietf-rserpool-threats-09 Thanks for the response! This email is just in response to your emailed comments--I will respond to the update document separately. On Apr 22, 2008, at 3:19 PM, <[EMAIL PROTECTED]> <[EMAIL PROTECTED] > wrote: > I have updated the draft to reflect your comments as > draft-ietf-rserpool-threats-10 and will be sending it to you tomorrow. > I have to put in the references and clean up a few other minor nits. > > Some of your comments need more response than just changing the > document. I have below your comment for some of them followed by my > response. I also changed text in the document in response to these. > For other comments that are not called out below, I simply changed the > document. > > Section 2.3.3 > > This section needs more discussion. Is it possible for an > authenticated PE to send misinformation about another PE? Is this also > an authorization issue? > > ------- > > Each PE sends information only about itself to the ENRP server. As > you correctly point out, if PEs are allowed to speak for each other > there is a number of security issues that would be introduced. The > protocol avoids this by requiring a PE to register only for itself. Thanks! It would be helpful to mention that in the document. > > > Section 2.6.3 > > Is this advice to the implementor? If so, the document should describe > what mechanism to use to authenticate the server. Or is the intent to > say that the RSerPool protocol does not provide such a mechanism and > needs to do so? > > ------------------- > > A requirements document is just a discussion of security requirements > and does not contain advice to the implementor. The advice to the > implementor is in the ASAP and ENRP documents in the security > considerations section. The authentication method is TLS which is > described in ASAP and ENRP. As I said in my first e-mail, I can copy > the text in ASAP and ENRP into the requirements document if you want. > Historical note: both IPsec and TLS fulfill the requirements in the > threat document. After debating, we chose TLS. I don't know that its necessary to copy whole sections from the ASAP and ENRP documents, but it would be helpful if this document very briefly mentioned what mechanisms the protocols use for each requirement. > > > Section 2.8.3: > > Same question about mechanism as in my previous comment. > > --------------------- > > Same answer as above. And same response :-) > > > Section 2.9.1: > > I do not know the RSerPool details, but I would assume that if PE A > could take over for PE B for a given user, the user would have to have > the same trust relationship with PE B as for PE A. Is the point of > this section to say that RSerPool does not require this and should? > > ------------------------------ > > This trust between PU and PE is established at the application layer > so Rserpool doesn't have control over that. Rserpool is a session > layer protocol. Rserpool can notify the application when fail over > occurs so the application can take appropriate action. That is the > first option called out in the requirements list. We don't have a > mechanism to reestablish the security context transparently so I'll > delete that option. It is not really an option since we don't know > how to do it :-). Is there some guidance in one of the other documents on what the application should do when it receives such a notification? > > > Section 2.10.1: > > Is this one threat, or three different threats? > > -------------- > > Threee different threats that are all solved by an integrity > mechanism. I still find the structure confusing. It might be helpful to put a paragraph (prior to the labeled sub-items) that mentions that these threats are all just examples of data-corruption or tampering threats. > > > 2.14.1: > > s/"These Messages"/"Endpoint keep-alive messages" > > Also, Is this not really the a consequence of 2.13, rather than a > separate attack? > > --------------- > > No, this is a separate attack. See ENRP section 3.5.1. It looks to > me like I have a cut/paste error and incorrectly stated the threat as > keep-alive between PU and ENRP server. It should be keep-alive > between an ENRP server and PE which I fixed in the document. An > endpoint keep alive message is sent from an ENRP server to a PE when > the PE loses its home ENRP server. Sorry this was messed up but you > caught it. I think I did a cut/paste from the previous section and > didn't update it. OK > > > Section 3.1 and 3.2 > > It seems to me that sections 3.1 and 3.2 belong in the main body of > the document. > > ------------------------- > > OK, I moved them to the main body of the document. > OK > Thanks for your feedback. It improved the document. > Thanks! > [...] _______________________________________________ Gen-art mailing list [email protected] https://www.ietf.org/mailman/listinfo/gen-art
