Re: [Roll] [6man] #8 (draft-thubert-6man-flow-label-for-rpl): draft-thubert-6man-flow-label-for-rpl-03 relation with draft-bormann-6lo-rpl-mesh-00
"6man issue tracker" <trac+6man@trac.tools.ietf.org> Thu, 14 August 2014 07:23 UTC
Return-Path: <trac+6man@trac.tools.ietf.org>
X-Original-To: roll@ietfa.amsl.com
Delivered-To: roll@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 523F11A08FC; Thu, 14 Aug 2014 00:23:51 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: 1.331
X-Spam-Level: *
X-Spam-Status: No, score=1.331 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, FB_CIALIS_LEO3=3.899, RP_MATCHES_RCVD=-0.668] autolearn=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 5ePIuIQSdZzK; Thu, 14 Aug 2014 00:23:44 -0700 (PDT)
Received: from zinfandel.tools.ietf.org (zinfandel.tools.ietf.org [IPv6:2001:1890:123a::1:2a]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 72A741A08F5; Thu, 14 Aug 2014 00:23:44 -0700 (PDT)
Received: from localhost ([::1]:57842 helo=zinfandel.tools.ietf.org) by zinfandel.tools.ietf.org with esmtp (Exim 4.82_1-5b7a7c0-XX) (envelope-from <trac+6man@trac.tools.ietf.org>) id 1XHpNv-0002o4-GP; Thu, 14 Aug 2014 00:23:43 -0700
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 8bit
From: 6man issue tracker <trac+6man@trac.tools.ietf.org>
X-Trac-Version: 0.12.3
Precedence: bulk
Auto-Submitted: auto-generated
X-Mailer: Trac 0.12.3, by Edgewall Software
To: pthubert@cisco.com, mariainesrobles@gmail.com
X-Trac-Project: 6man
Date: Thu, 14 Aug 2014 07:23:43 -0000
X-URL: http://tools.ietf.org/wg/6man/
X-Trac-Ticket-URL: http://trac.tools.ietf.org/wg/6man/trac/ticket/8#comment:1
Message-ID: <082.779ea0cba3a3fda448c718ef6764041e@trac.tools.ietf.org>
References: <067.39ff52585c7babe2e3fa02e2043cbad6@trac.tools.ietf.org>
X-Trac-Ticket-ID: 8
In-Reply-To: <067.39ff52585c7babe2e3fa02e2043cbad6@trac.tools.ietf.org>
X-SA-Exim-Connect-IP: ::1
X-SA-Exim-Rcpt-To: pthubert@cisco.com, mariainesrobles@gmail.com, roll@ietf.org, 6man@ietf.org
X-SA-Exim-Mail-From: trac+6man@trac.tools.ietf.org
X-SA-Exim-Scanned: No (on zinfandel.tools.ietf.org); SAEximRunCond expanded to false
Archived-At: http://mailarchive.ietf.org/arch/msg/roll/mq-J8GItU2IGTDlsFiPCSrhq_5Y
X-Mailman-Approved-At: Thu, 14 Aug 2014 00:26:06 -0700
Cc: roll@ietf.org, 6man@ietf.org
Subject: Re: [Roll] [6man] #8 (draft-thubert-6man-flow-label-for-rpl): draft-thubert-6man-flow-label-for-rpl-03 relation with draft-bormann-6lo-rpl-mesh-00
X-BeenThere: roll@ietf.org
X-Mailman-Version: 2.1.15
Reply-To: 6man@ietf.org, Routing Over Low power and Lossy networks <roll@ietf.org>
List-Id: Routing Over Low power and Lossy networks <roll.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/roll>, <mailto:roll-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/roll/>
List-Post: <mailto:roll@ietf.org>
List-Help: <mailto:roll-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/roll>, <mailto:roll-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 14 Aug 2014 07:23:51 -0000
#8: draft-thubert-6man-flow-label-for-rpl-03 relation with draft-bormann-6lo- rpl-mesh-00 Comment (by mariainesrobles@gmail.com): Source: http://www.ietf.org/mail-archive/web/roll/current/msg08808.html From: "Pascal Thubert (pthubert)" <pthubert at cisco.com> Date: Sun, 10 Aug 2014 17:05:34 +0000 Hi Carsten I think you jumped too quickly on my suggestion at 6lo. It makes a lot of sense to work on a 6lo routing dispatch and in particular enable the compression of all RPL artifacts. It will take some time to figure out what goes in there but my suggestion at 6lo includes routing header compression, RPL option, but also source destination if we decide to route fragments. I hope this happens and I'll see with Laurent if we use a respin of his draft or I start a new one. This work will be beneficial in the long run, but doubly fails to address the issues at stake here: - the draft proposes to update the rules in 6437 with stateful and with domain scoped operations. This is used by the draft itself but also by published and deployed standards that came up before 6437 (eg ISA100.11a) and we are now acknowledging and legalizing a de facto situation. - it will take a year or 2 at best to agree on the new routing dispatch and its operation. We need an improvement now. And btw we cannot necessarily assume 6lo together with RPL, I know of at least one RPL implementation that does not use that compression. Finally, Using a dispatch for the RPL option alone seems like a waste, but that is a discussion for 6lo. I'll publish there when I'm back from vacations. Cheers, Pascal ------------------------------- Source: http://www.ietf.org/mail-archive/web/roll/current/msg08811.html From: Carsten Bormann <cabo at tzi.org> Date: Sun, 10 Aug 2014 21:07:04 +0200 On 10 Aug 2014, at 19:05, Pascal Thubert (pthubert) <pthubert at cisco.com> wrote: > - it will take a year or 2 at best to agree on the new routing dispatch and its operation. We need an improvement now. And btw we cannot necessarily assume 6lo together with RPL, I know of at least one RPL implementation that does not use that compression. I don’t understand why we can’t do the saner equivalent of the flow label hack now and come up with a source routing header (which is not covered by the flow label hack) later. Please tell us more about how ISA100.11a uses the flow label; that is information that has been lacking from the discussion so far. Grüße, Carsten ------------------------------------ Source: http://www.ietf.org/mail-archive/web/roll/current/msg08817.html From: Carsten Bormann <cabo at tzi.org> Date: Tue, 12 Aug 2014 10:40:19 +0200 On 12 Aug 2014, at 09:15, Ines Robles <mariainesrobles at googlemail.com> wrote: > This technique saves bytes Again, the straw man at http://tools.ietf.org/html/draft-bormann-6lo-rpl- mesh-00 saves the same number of bytes. Actually, for common cases, it enables one additional optimization (saving one more byte) that is not available with the flow label hack. (On the other hand, due to the way RFC 6282 works, the flow label hack saves one byte where both it and ECN are in use and the common case does not apply.) The disadvantage of the RPL mesh header is that this approach is not available outside 6lo networks (i.e., networks employing adaptation layers that use RFC 6282). I’m not aware yet of a use case where that is a problem. In either encoding (RPL mesh header or flow label), it is probably a good idea anyway to unpack the compressed information back into the RPL IPv6 option when leaving the 6lo networks. Grüße, Carsten ------------------------------- Source: http://www.ietf.org/mail-archive/web/roll/current/msg08824.html From: "Pascal Thubert (pthubert)" <pthubert at cisco.com> Date: Wed, 13 Aug 2014 16:51:21 +0000 I disagree, Carsten. To save all the bytes, we need to update the 6MAN rules and allow to change a non-zero flow label, be it only to reset and elide it when coming from the Internet. More in my other answer to you today... Pascal ----------------------------- Source: http://www.ietf.org/mail-archive/web/roll/current/msg08818.html From: Brian E Carpenter <brian.e.carpenter at gmail.com> Date: Wed, 13 Aug 2014 11:38:53 +1200 Ines, Can you perhaps clarify the exact question you are asking 6man? (I'm thinking that 6man isn't being asked whether the number of bits saved is a useful energy saving, or whether Carsten's proposal is better.) Regards Brian On 12/08/2014 19:15, Ines Robles wrote: > Hi, > > On April, it was discussed and got consensus in ROLL [ > http://www.ietf.org/mail-archive/web/roll/current/msg08634.html]. > > This technique saves bytes, as was demonstrated by Xavier in his > implementation [ start on Slide 41 - > https://bytebucket.org/6tisch/meetings/raw/712902bb451d113494a7045e9730f1bb50335a79/140720_ietf90_plugfest_toronto/ietf90_toronto_plugfest_slides.pdf]. > Maybe we should enumerate the advantages and disadvantages of using it. > (Some of these are tracked in ticket #5, #6 and #7). > > > Anyway, demostratedIt needs the 6man approval to go forward. > > Cheers, > > Ines ------------------------- Source: http://www.ietf.org/mail-archive/web/roll/current/msg08820.html From: Ines Robles <mariainesrobles at googlemail.com> Date: Wed, 13 Aug 2014 11:00:12 +0300 Hi Brian, Thanks for your comment. Actually, I did not ask a question, I just wanted to say that the Pascal's draft got consensus at ROLL in april, and the WGLC was done between 6man and ROLL to know the 6man opinion about the use of flow-label (tickets #5, #6 and #7). With the new Carsten proposal (6lo draft), we (constrained environment WGs) should analyze this approach and decide the further steps on Pascal draft related with that. Thanks, Ines --------------------------- Source: http://www.ietf.org/mail-archive/web/roll/current/msg08822.html From: "Pascal Thubert (pthubert)" <pthubert at cisco.com> Date: Wed, 13 Aug 2014 16:37:27 +0000 Hi Brian: The question to 6MAN is really whether the use that the draft makes of the flow label within the RPL domain is acceptable or not. In particular Section 1.3 does not force a particular usage of the flow label inside the LLN, it simply updates the rules of what is allowed and what is not within that domain, knowing that the root will set the label properly towards the Internet and reset it from the Internet. It is up to ROLL and other external standards adopting the work (ISA100, WCI) to decide of its usefulness. Cheers, Pascal -------------------------------- Source: http://www.ietf.org/mail-archive/web/roll/current/msg08819.html From: Carsten Bormann <cabo at tzi.org> Date: Wed, 13 Aug 2014 03:19:12 +0200 On 13 Aug 2014, at 01:38, Brian E Carpenter <brian.e.carpenter at gmail.com> wrote: > Can you perhaps clarify the exact question you are asking 6man? Well, I’m not Ines, but I’m trying to find out why we are in this peculiar position of discussing this draft right now. ROLL has a routing protocol that requires some data to be shipped around together with IP packets. RFC 6553 and 6554 define ways to do this that are consistent with the IP architecture. Unfortunately, the overhead (signal-to-fluff ratio) is relatively high, and in this area of application, that matters. Normally, the next step would have been looking at ways to do header compression. We did talk about compressing ROLL, but with a view to compressing the (control plane) ROLL messages, not so much about the (data plane) IP packets themselves. GHC is trying to be a reasonably useful, but also reasonably general way to do the control plane messages. For the data packets, the flow label (and its now predominant non-use) provides an attractive hole in the IPv6 packets to ship around the RFC 6533 information, but not the RFC 6554 information. In 6lo networks, normally RFC 6282 compresses away empty flow labels, but it is cheap to put them in, so a flow label really only costs 3 bytes (instead of 8 bytes RFC 6553 costs). The most useful information from RFC 6553 can be stuffed into 19 bits, as demonstrated by the draft we are discussing here. RFC 6282 has extension points (GHC uses one of them), but not really useful ones for the ROLL data plane. So it appears it never occurred to us that the best way to handle these 19 bits is to actually sidestep RFC 6282, and use the existing extension points of RFC 4944. Until Laurent showed one way of doing this. I just went from there and did a strawman using Laurent’s idea for compressing the RFC 6553 option, in a way that is as efficient as (or, in most cases, actually more efficient than) using the flow label hole. So to me it seems there no longer is a good technical argument to ask 6man to liberate the flow label for carrying RFC 6553. But that is maybe something that should now be discussed between 6lo and ROLL; I don’t think that this is a 6man issue (unless the answer is a simple “sure, go ahead, take it, we don’t need it any more anyway”, which is not what I have perceived RFC 6437 to say). When 6lo and ROLL have generated consensus on the need to do this, we may or may not want to go back to 6man and actually pry that flow label hole out of your fingers. Grüße, Carsten ------------------------- Source: http://www.ietf.org/mail-archive/web/roll/current/msg08823.html From: "Pascal Thubert (pthubert)" <pthubert at cisco.com> Date: Wed, 13 Aug 2014 16:48:06 +0000 Hello Carsten: In my view, this work is not competing with 6lo: for the specific value of keeping the draft whatever 6lo decides to do: - If this draft is not adopted, the flow label from LLN will probably stay all zeroes as it is today and the goal of 6437 will not be achieved. There is no way a device will waste 20 bits for a random that has no value inside the LLN. The text tells the RPL root to set the flow label. - If this draft is not adopted, the root of the LLN will have to forward the flow label coming from the Internet down to the end destination device, deep in the mesh, wasting 20 bits in each frame. These is a waste that 6lo compression will not be allowed to legally elide if today we decide that we do not allow it. Note: even if the correspondent sets the FL to zeroes to help the LLN, there is no way to guarantee that the flow label will not be set on the way since that much is lawful. - If the draft is not adopted, existing usages in shipping standards that incorporate IPv6 will stay borderline/unlawful. E.g. the draft legalizes the way ISA100.11a uses the flow label for a stateful flow identification within the ISA100 domain. - If the draft is not adopted, we have no leverage over WCI that is defining the backbone router for ISA100.11a devices. The draft indicates that the root should ensure that the flow label is correctly set when going outside, which is a minor increment to WCI, if pushed in due time with a proper RFC number that demands it. The need for this work was agreed at ROLL. We need a better solution than available, for application in RPL domains, that are often but not necessarily congruent with 6LowPAN domains. I'm fully supportive to have a 6lo compression that will improve the situation for a 6lowpan network, and at this point we have 3 proposals on the table, Laurent's, yours, and mine, which is to have a routing dispatch for route-over in parallel to the mesh dispatch that is mesh-under. I do hope we create a momentum there and converge. But there is no guarantee. All in all, I will not drop something that works now, with running code and all, and reached last call status, against the sirens of a compression that will not solve the issues above. Cheers, Pascal ------------------ Source: http://www.ietf.org/mail-archive/web/roll/current/msg08826.html From: Philip Levis <pal at cs.stanford.edu> Date: Wed, 13 Aug 2014 12:07:04 -0700 On Aug 13, 2014, at 9:48 AM, Pascal Thubert (pthubert) <pthubert at cisco.com> wrote: > If this draft is not adopted, the flow label from LLN will probably stay all zeroes as it is today and the goal of 6437 will not be achieved. Pascal, I'm trying to reconcile your claim that the goal of 6437 is to allow enclosed networks to use the flow label with Brian's statement > Actually that's why I don't want to see a formal > update to 6437, because the only rational update would be to allow > any closed domain to invent its own usage. We had that argument at > length during the development of 6437, and decided against it. Phil -------------- Source: http://www.ietf.org/mail-archive/web/roll/current/msg08830.html From: Brian E Carpenter <brian.e.carpenter at gmail.com> Date: Thu, 14 Aug 2014 08:52:56 +1200 Right. I'm drawing a very subtle line between (a) stating an exception to 6437 for this particular usage and (b) opening the door to other usages. Since 6man clearly didn't want (b) during the development of 6437 I think we do need to limit ourselves to (a). Brian -------------------------- Source: http://www.ietf.org/mail-archive/web/roll/current/msg08828.html From: Philip Levis <pal at cs.stanford.edu> Date: Wed, 13 Aug 2014 12:11:48 -0700 On Aug 13, 2014, at 9:48 AM, Pascal Thubert (pthubert) <pthubert at cisco.com> wrote: > The need for this work was agreed at ROLL. We need a better solution than available, for application in RPL domains, that are often but not necessarily congruent with 6LowPAN domains. This seems to me to be the strong technical argument in favor of the draft. While Carsten's approach seems better technically, it only works for 6lowpan, and RPL should be independent of the underlying link layer. Phil ------------------------- Source: http://www.ietf.org/mail-archive/web/roll/current/msg08829.html From: Ralph Droms <rdroms.ietf at gmail.com> Date: Wed, 13 Aug 2014 15:49:45 -0400 On Aug 13, 2014, at 3:11 PM 8/13/14, Philip Levis <pal at cs.stanford.edu> wrote: > On Aug 13, 2014, at 9:48 AM, Pascal Thubert (pthubert) <pthubert at cisco.com> wrote: > >> The need for this work was agreed at ROLL. We need a better solution than available, for application in RPL domains, that are often but not necessarily congruent with 6LowPAN domains. > > This seems to me to be the strong technical argument in favor of the draft. While Carsten's approach seems better technically, it only works for 6lowpan, and RPL should be independent of the underlying link layer. Could we consider Carsten's approach a compression mechanism for the RPL information options, rather than a replacement, thereby allowing a one-to- one translation between compressed mode on 6lowpan networks and uncompressed mode in other networks? - Ralph ------------------------------------------------ Source: http://www.ietf.org/mail-archive/web/roll/current/msg08831.html From: Carsten Bormann <cabo at tzi.org> Date: Wed, 13 Aug 2014 23:02:30 +0200 >> This seems to me to be the strong technical argument in favor of the draft. While Carsten's approach seems better technically, it only works for 6lowpan, and RPL should be independent of the underlying link layer. The transport of RPL specific information for data plane packets already *is* independent of the underlying link layer, as specified in RFC 6553 and RFC 6554. All we are talking about here is an optimization for those (for 6553, specifically; actually 6554 is probably the one even more in need). > Could we consider Carsten's approach a compression mechanism for the RPL information options, rather than a replacement, thereby allowing a one-to- one translation between compressed mode on 6lowpan networks and uncompressed mode in other networks? Indeed, that is probably the best way to view this approach. (In the next version of the draft, I’ll emphasize this.) My assumption here is: — Any subnet that could not live with transporting RFC 6553 as is, i.e. 8 bytes extra compared to the flow label approach (for uncompressed packets*), already needs (and, at this point, has) a header compression solution for the expansive IPv6 header anyway. So adding RFC 6553 compression to that solution is the natural approach. — All subnets that can live with the IPv6 header as is, also can live with RFC 6553 as is. This argument is on a technical level. I’m not talking about cognitive dissonance**) here, or political issues. If there are important political issues (I have no idea what the ISA100/WCI talk is about), they probably need to be brought out in the open here. Grüße, Carsten *) less difference for header-compressed packets, e.g., 5 bytes for 6lo. **) http://www.iab.org/wp-content/IAB-uploads/2011/04/Bormann.pdf, section 2.6 -- -------------------------------------------------+------------------------- Reporter: mariainesrobles@gmail.com | Owner: Type: defect | pthubert@cisco.com Priority: major | Status: new Component: draft-thubert-6man-flow-label-for- | Milestone: rpl | Version: Severity: In WG Last Call | Resolution: Keywords: | -------------------------------------------------+------------------------- Ticket URL: <http://trac.tools.ietf.org/wg/6man/trac/ticket/8#comment:1> 6man <http://tools.ietf.org/wg/6man/>
- [Roll] [6man] #8 (draft-thubert-6man-flow-label-f… 6man issue tracker
- Re: [Roll] [6man] #8 (draft-thubert-6man-flow-lab… 6man issue tracker
- Re: [Roll] [6man] #8 (draft-thubert-6man-flow-lab… 6man issue tracker