Re: [Int-area] Comments on INT Area Tunnels draft: aft-ietf-intarea-tunnels-09

[Gorry Fairhurst <[e-mail]>]

Thanks Joe, see below.

On 16/04/2019, 16:12, Joe Touch wrote:
> Hi, Gorry,
>
> Thank you for the deep feedback - I’ve been waiting for this sort of 
> input for quite a long time (from the community as a whole)...
>
>> On Apr 16, 2019, at 4:28 AM, gorry Fairhurst <[e-mail] 
>> <mailto:[e-mail]>> wrote:
>>
>>
>> I have read draft-ietf-intarea-tunnels-09, because it is cited from 
>> another document that I am reviewing. I have a number of comments 
>> that I’d like to pass to the WG/editors for consideration, I also 
>> have a few more technical questions, that I will post later.
>>
>> --- My first top-level comment is that I fear a small part of the 
>> language in the document could put some readers off because it seems 
>> to go-back-in-time to cite early ways of describing networks, which I 
>> think could be easily tuned to avoid this and make it much more 
>> accessible to younger folk - who probably are a really important 
>> audiance (see NiTs below).
>
> I’ll look; it certainly is important to start with Internet terms 
> (gateway, datagram, etc.) before jumping into more recent terms.
>
>> --- There are places where IPv4 language is used, and I'd prefer to 
>> make sure that IPv6 is regarded as equally in all places (see NiTs).
>
> Agreed,
>
>> --- It would be good to refer to and clarify the recommendations in 
>> [I-D.ietf-tsvwg-ecn-encap-guidelines], 
>> [I-D.ietf-tsvwg-rfc6040update-shim] and 
>> [I-D.ietf-tsvwg-datagram-plpmtud] witin TSVWG, and also IP 
>> Fragmentation Considered Fragile [I-D.ietf-intarea-frag-fragile-09].
>
> Certainly - thanks for noting that. The doc has been in development 
> over an absurdly long time (due to waxing and waning IETF interest), 
> but I do hope we can wrap things up now that there seems to be broader 
> interest.
>
> Further notes below...
>
>>
>> Best wishes,
>>
>> Gorry
>>
>> —
>
>> "The opening sentence of the Introduction states “The Internet 
>> layering architecture is loosely based on the ISO seven
>> layer stack, in which data units traverse the stack by being wrapped
>> inside data units of the next layer down [Cl88][Zi80]”.
>>
>> - While I totally agree with the observation, I wonder about the 
>> prudence in starting the Intro with this sentence. For many years 
>> Universities (at least where I have been) have taught the ISO model 
>> as a formal protocol stack as well as an architectural model. That’s 
>> not common anymore, and starting with a historical fact can actually 
>> deter “younger” readers.
> - this remains the dominant model for teaching (top-down or bottom-up)
> - I know of only one teaching approach and textbook that do not use 
> that model (and it’s mine and under development ;-)
>
You may not be as alone as you feel on this.
>> - The text could you say something like:
>>
>> “The Internet layering architecture follows a layered model, in which 
>> data units traverse a stack by being wrapped inside data units of the 
>> next layer down [Cl88][Zi80]”.
>
> It’s important to refer to the ISO 7-layer stack because it remains 
> dominant in protocol stack descriptions (even the idea of a protocol 
> stack, inaccurate as it is, is based thereon).
>
>>
>> - and later, this strikes me as somewhat old-fashioned:
>> “ Such descriptions can help explain
>> behavior, as when the OSI seven-layer model is used as a teaching
>> example [Zi80].”
>>
>> - and again:
>> “An Internet gateway is an OSI Layer 3 router when it
>> transits IP datagrams but it acts as an OSI Layer 2 host as it
>> sources or sinks Layer 2 messages on attached links to accomplish
>> this transit capability. “
>
> See above.
>
>> - That’s pretty old-fashioned language for a reader in 2019, and 
>> definitely not the language that anyone brought up with IPv6 would
>> recognise. Since those early days “gateway” has had different 
>> interpretations by different readers - and that’s even more so today.
>
> Agreed; I’ll provide a translation table up front and use more modern 
> terms.
>
The stack of protocols have long since died... the reference model 
wasn't bad... encapsulation and layering remain. Anyway, to me the 
really important thing fr is that people find and then read the 
RFC-to-be and read past the start of the introduction to take away the 
important recommendations. I suspect some checking can separate 
historical context from providing a motivation to read.
>>
>> ---
>> - Similarly, this seems quaint:
>> “There is essentially no difference between a tunnel and
>> the conventional layering of the ISO stack (i.e., by this
>> definition, Ethernet is can be considered tunnel for IP).”
>> - Is it talking about protocol encapsulation? In 2019 do we need to 
>> mention the ISO Stack?
>
> We do - the stack is an ISO concept.
>
> Yes, this refers to protocol encapsulation.
>
> Note that this sentence is fundamental to much of the rest of the 
> document - i.e., if you get tunneling right, you can’t tell the 
> difference between running over a tunnel vs. running over the next 
> protocol layer down.
>
>> -------------------------------
>>
>> - The multicast discussion is split between two parts fo the intro. 
>> The phrases:
>> “Tunnels allowed multicast
>> packets to transit efficiently between multicast-capable routers over
>> paths that did not support native link-layer multicast.”
>> ... introduce multicast before the list of protocols, but the 
>> discussion of multicast continues after more current examples in
>> “The first attempt to use large-scale tunnels was to transit multicast
>> traffic across the Internet in 1988 “
>> - Could these two be combined?
>
> Probably - I’ll check.
>
>> --
>> - I suspect the sentence mentioning IPv6 transition will be more 
>> familiar than mulkticast to many, maybe it could be a separate para?
>> “Similar
>> techniques have been used to support incremental deployment of other
>> protocols over legacy substrates, such as IPv6 [RFC2546].”
>
> Sure.
>
>> ---
>> “Link packet: a link layer message, which can carry an IP datagram
>> as a payload”
>> - I would expect many know this as a link frame, should this also be 
>> mentioned?
>
> I’ll add that to the terms; some links use frames (Ethernet, SONET); 
> others use cells (ATM), and others use packets. I
>
>> ---
>> In bullet “Ingress”:
>> - Is “physical link” defined? The use of physical and virtual may be 
>> important concepts, and could benefit from being defined, or a few 
>> words added here.
>
> Will do.
>
>> ---
>> - The statement:
>> “To the extent that
>> the Internet has a single, coherent interpretation, its architecture
>> is defined by its core protocols “
>> - The meaning seems clear, but the following list omits mention of 
>> the IPv6 Stack completely. That strikes me as really needing to be at 
>> least mentioned as *the* architecture for the Internet.
>
> Will do.
>
>> ---
>> In section 3.3:
>> “so both are viewed as hosts on network N (Figure 7).
>> Consequently [RFC1122]”
>> - is this the same in IPv6? (I think so, please can we cite an RFC?)
>
> Yes - thankfully there now is such a doc (RFC6434).
>
> (there wasn’t when I started this doc in 2008)
>
>> and:
>> “Because routers, not links, alter hop counts [RFC1812],”
>> - please add IPv6 language and reference.
>
> Will do - though this is only a draft right now 
> (draft-ietf-v6ops-ipv6rtr-reqs) and it expired last year.
>
>> “it must satisfy all
>> requirements expected of a link [RFC1122][RFC3819].”
>> - and IPv6 reference?
>
> See above...
>
>> ---
>> In section 4.1.1:
>> “When feedback is received from these fields,”
>> - Please explain “feedback” here, this was not clear by what 
>> mechanism and at what point along the path.
>
> Will do.
>
>> ---
>> In section 4.1.1:
>> “There are exceptions to this rule that are explicitly intended to
>> relay signals from inside the tunnel to the network outside the
>> tunnel,”
>> - I think this refers to an action at the tunnel egress.
>
> Both ingress and egress. Will clarify.
Ah yes - that should be written about.
>
>> ---
>> In section 4.1.2:
>> - When referring to [RFC6040] it would be useful to also reference 
>> this BCP-to-be (to be WGLC’ed shortly in tsvwg:
>> [I-D.ietf-tsvwg-ecn-encap-guidelines]
>> and
>> [I-D.ietf-tsvwg-rfc6040update-shim]
>> Which is a standards track specification that will update existing 
>> IP-shim-(L2)-IP protocols.
>>
>> - Although there consider more than this document, referencing them 
>> ensures that readers do not overlook them
>
> Will do.
>
>> .
>> ---
>> In section 4.1.2:
>> “interfaces of the router [RFC1122].”
>> - Please add IPv6 node reference.
>
> Will do (see above).
>
>> ---
>> In section 4.1.3:
>> “A router is
>> required to decrement the TTL by at least one or the number of
>> seconds the packet is delayed, whichever is larger [RFC1812].”
>> - Please insert “IPv4” router. I’d actually personally prefer the 
>> text to say that RFC 1812 required that an IPv4 router..”
>
> WIll do, though I tend to not use RFCs in the text (only as references).
>
>> ---
>> “(link-local discovery and related protocols [RFC4861]).”
>> - Please cite The Generalized TTL Security Mechanism (GTSM)
>> RFC 5082 as the reference and ND as an example.
>
> Will do.
>
>> ---
>> “ The uniqueness of the IP ID is a known problem for high speed nodes,
>> because it limits the speed of a single protocol between two
>> endpoints [RFC4963]. “
>> - I’d quibble - I think it is more accurate to say it *can* limit the 
>> speed when this field is used to uniquely identify packets in flight 
>> (e.g.
>> when fragmentation is enabled).
>
> Will do - again, this should refer to RFC6864, which didn’t exist when 
> we started this trek...
>
>> ---
>> In section 4.1.5:
>> - I’m cautious about the text below, because the other danger is to 
>> any other receiving process on the same node receiving corrupted packets:
>> “For this reason, it is safe to use UDP with a zero checksum for
>> atomic tunnel link packets only;”
>> - Could we cite [RFC6936] again after this to be sure the context is 
>> understood?
>
> Will do.
>
>> ---
>> “(PLMTUD) [RFC4821]”
>> - I think we should also cite: ietf-tsvwg-datagram-plpmtud.
>
> Will do.
>
>> ---
>> In section 4.2.2:
>> “ As a
>> link in network M, transit packets might be fragmented before they
>> reach the tunnel..”
>> - I found that sentence hard to parse and rather long. Maybe it was 
>> because it starts with “As a...” - is it possible to rephrase?
>
> Will do.
>
>> ---
>> “A router
>> attempting to process such a message would already have generated an
>> ICMP "packet too big" and the transit packet would have been dropped
>> before entering into this algorithm.”
>> - Not quite my understanding because of the wording, would this text 
>> be OK?:
>> “A router
>> attempting to process such a packet should already have
>> generated an
>> ICMP "packet too big" message and the transit packet would have
>> been dropped before entering into this algorithm.”
>
> I’ll check and make sure it’s more clear.
>
>> ---
>> - NiT:
>> “ Some messages have detailed specifications for relaying between the
>> tunnel link packet and transit packet, including Explicit Congestion
>> Notification (ECN [RFC6040]) and multicast (IGMP, e.g.).”
>> - Is it necessary to add IGMP as an example here? (I’m really unsure).
>
> It is, IMO - it is a signaling protocol that we tend to ignore and one 
> of the ways in which tunnels can break where (real) links work if we 
> don’t get it right.
>
>> - If you think this is useful can you relate to RFC 7450 - because 
>> that is being deployed and is a PS for tunnels.
>
> Will do.
>
>> “IGMP snooping enables IP multicast and also please add "and MLD” 
>> after IGMP.
>
> Will do.
>
>> ---
>> In section 4.3.3:
>> - It would be good to add a reference to RFC 8085 for multicast CC, 
>> citing section 4 of RFC 8085 contains guidance on congestion control 
>> for multicast tunnels.
>
> Will do.
>
>> ---
>> - NiT:
>> “It is useful to relay network congestion notification between the
>> tunnel link and the tunnel transit packets.”
>> - Please cite the RFCs and IDs on this - e.g. ECN RFCs cited earlier. 
>> I think you’ll find there are requirements, rather than “useful” 
>> comments.
>
> There are some, but not all the ones that are needed IMO. And frankly 
> I haven’t checked to make sure that all the docs in this space are 
> actually correct — taking a look at the list of errors that were not 
> noticed in previous standards, I’d want to either cite them as “like 
> these docs” or need to check them to make sure of the details before 
> citing them as recommended by **this* doc.
>
>> -----------
>>
>>
Gorry