Document: draft-cheshire-dnsext-dns-sd-07.txt Reviewer: Stéphane Bortzmeyer Review Date: 2010-12-06 IETF LC End Date: 2010-11-23 IESG Telechat date: Unknown Summary: The draft is long, containing a lot of interesting material, but not always well organized. As a general rule, I am a big fan of "rationale" sections, which let you know why a MUST or a SHOULD was used. But, here, they are intermixed with the normative content and it is often difficult to know when standard stops and background material begins. Examples are sections 4.4 or 6.2. In the same way, user interface discussions (section 4.2 for instance) are very interesting and useful but should be more clearly separated (may be in a distinct document). It also raises a major issue when you put it in perspective with other DNS-related documents: non-standard terminology and strange ways to describe the rest of the DNS world. It does not seem suitable for a Standards-Track document. Major issues: There are non-standard terms which may create confusion with existing technologies. My main concern is with "unicast DNS". There is no such thing as unicast DNS, there is the DNS and there are other protocols (like LLMNR or Apple Bonjour name protocol) which are not DNS. Even worse is "unicast DNS domain name" (unicast refers to a name resolution method, but a name exists independently of a resolution method). Using "local", outside of the procedures of RFC 2606 and 2826 and similar is a bad idea. I do not elaborate here since the problem is actually in draft-cheshire-dnsext-multicastdns-12.txt. The idea in section 7 of using _udp for every SRV records (but TCP) is a bad one, which you cannot find in RFC 2782 which states clearly that other values than _tcp and _udp can be used. The fact (section 18) that there is no IANA registry for these values is not, in my opinion, a sufficient reason. The idea of "IANA secret registrations" (section 18) would be, it seems, a first in the IETF world and should be rejected. The entire point of IANA registries is to document publicly the *technical* identifiers. Trademarks and marketing terms do not have to be in the IANA registry and therefore do not need special rules for them. Minor issues: For name resolution on the local network, only mDNS (a misnomer since it is not DNS) is mentioned, not LLMNR (RFC 4795) which seems it would work as well. Section 4.1 says that the name must be in NFC. Why not referring to its subset defined in RFC 5198? Among the practical differences are the fact that RFC 5198 handles the case of end-of-lines, which is ignored by the draft. "commonly-used 16-bit Unicode space" is a misnomer. If one wants to talk about the Basic Multilingual Plane (which stores only a minority of Unicode code points), it should use the standard term . The "standard DNS message size" is no longer 512 bytes since RFC 2671 (section 6.1). Yes, EDNS0 sometimes is blocked by broken middleboxes but for a technique used only in local networks, this is not a too serious issue. It seems that the possibility of several TXT in a RRset (not several strings in a TXT, which is covered in 6.1) is not mentioned. Is it legal and, if yes, how to use it? The advice of section 13.1 to stuff several other records (which may be in different bailiwicks) in the DNS answer is at odds with security practices (RFC 2181, section 5.4.1 and RFC 5452, section 6). Nits/editorial comments: Long quotes of existing RFCs (section 4.1) are unecessary since the other RFCs are available elsewhere.