Return-Path: X-Original-To: mpls@ietfa.amsl.com Delivered-To: mpls@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 429A1130E6F for ; Thu, 27 Sep 2018 01:30:59 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.998 X-Spam-Level: X-Spam-Status: No, score=-1.998 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_NONE=-0.0001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com 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 wDEffDVPM_6v for ; Thu, 27 Sep 2018 01:30:55 -0700 (PDT) Received: from mail-pg1-x536.google.com (mail-pg1-x536.google.com [IPv6:2607:f8b0:4864:20::536]) (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 62033130E13 for ; Thu, 27 Sep 2018 01:30:55 -0700 (PDT) Received: by mail-pg1-x536.google.com with SMTP id v133-v6so1411749pgb.2 for ; Thu, 27 Sep 2018 01:30:55 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025; h=mime-version:references:in-reply-to:from:date:message-id:subject:to :cc; bh=6A9Q7qz7TWx4zr8ShYJ/msFvI6Th+FaGPqztIoDZyhQ=; b=SzKjor2dX6NKGyb8DJpSsruxSJv271oj5bNKy5M7h6kQ0wJLLur5vnjKEZ+lJudwW/ J/k0lTe59b+sIQYawAjUVLRFj9yfguR5e5WsbiRq8vBARDL11/x2PjtBIqbg3lnWkInl VtLOkgIFvh3jCzUdUdEQXpx76TPbtoiG3HyumR3GJtSkU+wt3V0yrvmEnPmZtyAVfTKT SJq65C/+bViLV02mwN5LVnz0dAr2qV1SPdSJIdShDtuLLVVB4o9COAnVv3XgtNQ2BE2e kLu9ibTOkFSykuBXLuEFSpDuS9Xvj0BgYzqchiD0QEiN2j1ZDZEGYHylht5dZNZM2ZVv XtZg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to:cc; bh=6A9Q7qz7TWx4zr8ShYJ/msFvI6Th+FaGPqztIoDZyhQ=; b=OwahjxQT/UE+9iNPHP9WsekiOWwOmFHjwS8vgvSnS5jlCa8x2PTaXl+wfbbinwOONH YZuosvqf7N0clLJ0pP4o/eQDMOagfVFsncIIAegYjFuRouUOoNFE0ygW2RYWconqWOvc S/l9Jqt8AqfRKJfA/dgYlj4DRPylgRflTpG8QBpzKPIfOpdPcHoFzzlGPvjWiu52CDt2 nv4rjNdewGUYBEZfEOEkQ++d38OHHzPSAhduLOzDEk+Ncbi5K+oAIMtydBhxRWMssDFy rIu1PsSeeBsINEfVw8TIC9hOzCqbDDBgmbnD808YaPWSa5aNG4LlJh7kKqyYipic0wag MKCg== X-Gm-Message-State: ABuFfogB/48wMvOikrvyRVDum76gsstvBjEbhxwg/j2HInKNp2sxKndZ RW6Efc3dD9fvnQZMkmgvJ6S3jDw0MH4WGN2+a+o= X-Google-Smtp-Source: ACcGV602HLHJwNjE11janxXYKtFMOEBs3znB0CENGZYWcAOjN+jCxun7pKnLU0hiSZ/XBUmYDgGqWOgFDTbzIq90Mt0= X-Received: by 2002:a62:56d9:: with SMTP id h86-v6mr10209490pfj.229.1538037054718; Thu, 27 Sep 2018 01:30:54 -0700 (PDT) MIME-Version: 1.0 References: <211770d4-8279-33e2-b6bf-289261b6f6ff@gmail.com> <5437736B-6E12-4595-A333-367AB7232692@gmail.com> <60C0E897-0D5D-4230-9094-4367524F91EE@gmail.com> In-Reply-To: From: Vishnu Pavan Beeram Date: Thu, 27 Sep 2018 04:30:42 -0400 Message-ID: To: mhartley.ietf@gmail.com Cc: Alexander Okonnikov , IETF MPLS List Content-Type: multipart/alternative; boundary="0000000000004c49640576d6285e" Archived-At: Subject: Re: [mpls] Requesting WG LC for draft-ietf-mpls-rsvp-shared-labels X-BeenThere: mpls@ietf.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Multi-Protocol Label Switching WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 27 Sep 2018 08:30:59 -0000 --0000000000004c49640576d6285e Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Matt, Hi! The purpose of this document is to provide a mechanism to bring up MPLS RSVP-TE LSPs on a shared forwarding plane. There is a set of key RSVP-TE features/functionalities specified in the Introduction Section that would continue to work with this forwarding plane (without needing any special extensions). The authors believe that a deployable solution can be implemented with this set of features/functionalities. If there are any other desirable features/functionalities (that are not specified as supported in this document) that can be supported on this forwarding plane by introducing a few protocol extensions, then those are expected to be discussed separately (draft-chandra-mpls-rsvp-shared-labels-np is an example of that). Regards, -Pavan On Wed, Sep 26, 2018 at 10:56 AM Matt Hartley wrote: > Pavan, > > The problem with lists like this (in general) is that it's not clear what > the status is for anything that isn't on the list, and most lists will > probably manage to omit *something*. I think there's three reasonable > options: > > 1. State unequivocally that everything works > 2. List the functionalities that definitely don't work, and make it clear > that everything else does > 3. List the functionalities that definitely work, and explicitly make no > guarantees about anything that isn't listed. > > It isn't really clear what you're trying to do here. > > Note that in this case option 1 isn't on the table because we've already > established that nnhop FRR requires further extensions. > > Cheers > > Matt > > On Wed, Sep 26, 2018 at 10:17 AM Vishnu Pavan Beeram < > vishnupavan@gmail.com> wrote: > >> Alexander Hi! >> >> >> >> The intent of the following statement in Section 1 is certainly not to b= e >> evasive (slightly or otherwise). >> >> Functionalities such as bandwidth admission control, LSP >> >> priorities, preemption, auto-bandwidth and Fast Reroute >> >> continue to work with this forwarding plane. >> >> >> >> We (the authors) still don=E2=80=99t see any problem with the above stat= ement. >> But we=E2=80=99ll go ahead and make a slight adjustment (see below) to a= ddress your >> concern. >> >> Key functionalities such as bandwidth admission control, LSP >> >> priorities, preemption, auto-bandwidth and Fast Reroute via >> >> facility backup protection continue to work with this >> >> forwarding plane. >> >> >> >> Regards, >> >> -Pavan >> >> >> On Tue, Sep 25, 2018 at 11:34 AM Alexander Okonnikov < >> alexander.okonnikov@gmail.com> wrote: >> >>> Hi Panav, >>> >>> Ok, 1:1 is not to be supported by this approach. In general, 1:1 has it= s >>> own benefits - for example, it is more attractive versus N:1 in ring >>> topologies. After all, it is FRR too, as N:1 one. My point was that >>> description in section 1 is slightly evasive regarding FRR and other >>> RSVP-TE properties (PMTUD). >>> >>> Thank you. >>> >>> >>> 16 =D1=81=D0=B5=D0=BD=D1=82. 2018 =D0=B3., =D0=B2 4:38, Vishnu Pavan Be= eram >>> =D0=BD=D0=B0=D0=BF=D0=B8=D1=81=D0=B0=D0=BB(=D0=B0): >>> >>> Alexander, Hi! >>> >>> >>> Please see inline for responses (prefixed VPB). >>> >>> >>> Regards, >>> -Pavan >>> >>> On Thu, Sep 13, 2018 at 1:37 PM Alexander Okonnikov < >>> alexander.okonnikov@gmail.com> wrote: >>> >>>> Hi Panav, >>>> >>>> From section 7: >>>> >>>> >>>> "If the label type is a delegation label, then the stacking procedure >>>> stops at that delegation hop." >>>> It is OK for "Stack to Reach Delegation Hop" approach, but it doesn't >>>> work for "Stack to Reach Egress", isn't it? >>>> >>> >>> [VPB] The logic specified in Section 7 could be used by any node >>> constructing the label stack (this could be the ingress or a delegation >>> hop). The sentence immediately following the above quoted sentence in >>> Section 7 is important. It currently reads =E2=80=93 >>> Approaches in Section 5.1 >>> >>> SHOULD be used to determine how the delegation labels are pushed in the >>> label stack. >>> >>> >>> The intent here is to say that if you encounter a delegation label, use >>> the procedures outlined in Section 5.1 to determine how the delegation >>> labels are pushed in the label stack. The following change to the text >>> should address this comment: >>> >>> OLD: >>> >>> If the label type is a delegation label, then the stacking procedure st= ops at that delegation hop. >>> Approaches in Section 5.1 SHOULD be used to determine how the dele= gation labels are pushed in the label stack. >>> >>> >>> NEW: >>> If the label type is a delegation label, then the type of stacking >>> approach chosen by the ingress for this LSP (Section 5.1) MUST be used = to >>> determine how the delegation labels are pushed in the label stack. >>> >>> >>> >>>> >>>> Also, regarding FRR support. Section 1 says: >>>> >>>> "Functionalities such as bandwidth admission control, LSP priorities, >>>> preemption, >>>> auto-bandwidth and Fast Reroute continue to work with this forwarding >>>> plane." >>>> It seems that shared labels approach supports only facility bypass >>>> link-protection. It doesn't support one-to-one link- and node-protecti= on, >>>> per my understanding. Facility bypass node-protection is not supported= as >>>> well (as mentioned in Section 8). Hence, FRR support is very limited, = and >>>> section 1 needs correction. >>>> >>> >>> [VPB] I don=E2=80=99t see anything wrong with the quoted text. Fast Rer= oute for >>> MPLS-TE LSPs can be realized by either the 1-to-1 protection mechanism >>> (detours) or the facility bypass mechanism. The authors don=E2=80=99t i= ntend to add >>> procedures for 1-to-1 link/node protection (who needs it?). The >>> facility bypass link-protection procedure is discussed in this draft. T= he >>> facility bypass node-protection procedure is discussed in >>> https://tools.ietf.org/html/draft-chandra-mpls-rsvp-shared-labels-np-00 >>> (this was presented at the last IETF). >>> >>> >>>> >>>> Thank you. >>>> >>>> 13 =D1=81=D0=B5=D0=BD=D1=82. 2018 =D0=B3., =D0=B2 20:28, Alexander Oko= nnikov < >>>> alexander.okonnikov@gmail.com> =D0=BD=D0=B0=D0=BF=D0=B8=D1=81=D0=B0=D0= =BB(=D0=B0): >>>> >>>> Hi Panav, >>>> >>>> Questions regarding ETLD: >>>> >>>> The draft is not clear about signaling of ETLD attribute. It says that >>>> ETLD is conveyed as per-hop attribute. Is my understanding correct tha= t it >>>> is conveyed as RRO Hop attribute? Probably it could be cleaned to avoi= d >>>> confusion whether ERO or RRO Hop attribute mechanism is used. >>>> >>>> Next, the draft says that: >>>> >>>> "... If a node is reached where the ETLD set from the previous hop is >>>> 1, then that >>>> node MUST select itself as the delegation hop. If a node is reached >>>> and it is >>>> determined that this hop cannot receive more than one transport label, >>>> then that node >>>> MUST select itself as the delegation hop. ..." >>>> >>>> What is purpose of the second sentence/rule? >>>> >>>> Next: >>>> >>>> "If there is a node or a sequence of nodes along the path of the LSP >>>> that do not >>>> support ETLD, then the immediate hop that supports ETLD MUST select >>>> itself as the >>>> delegation hop." >>>> >>>> If some node (consecutive nodes) doesn't support ETLD then it doesn't >>>> support TE labels. Hence, that node (regular RSVP-TE LSR) will do SWAP= and >>>> not POP. As a result non-decremented ETLD is OK and immediate hop that >>>> supports ETLD not necessary should become delegation hop? >>>> >>>> Also, from Section 9.7: >>>> >>>> "The ETLD field specifies the maximum number of transport labels that >>>> this hop can potentially send to its downstream hop. It MUST be set t= o a >>>> non-zero value." >>>> >>>> Strictly speaking it is not correct. ETLD reflects decrementing counte= r >>>> and not capability of some transit node. I.e. if we consider LSP R1-R2= -R3, >>>> R1 puts value 5 in ETLD,and R2 supports imposing of 2 labels, it doesn= 't >>>> mean that R2 should rewrite ETLD with value 2. It just should decremen= t >>>> value 5. Correct? >>>> >>>> Also, per my understanding it is supposed that in fact ETLD value will >>>> not be just decremented, but it will be copied from previous RRO Hop >>>> attributes subobject into being inserted RRO Hop attributes subobject = with >>>> decrementing. May be it would be better to signal ETLD value in LSP >>>> Attributes object (and each capable node decrements ETLD value there), >>>> while signaling support of ETLD itself in RRO Hop attributes subobject= ? >>>> >>>> Thank you. >>>> >>>> 13 =D1=81=D0=B5=D0=BD=D1=82. 2018 =D0=B3., =D0=B2 20:22, Alexander Oko= nnikov < >>>> alexander.okonnikov@gmail.com> =D0=BD=D0=B0=D0=BF=D0=B8=D1=81=D0=B0=D0= =BB(=D0=B0): >>>> >>>> Hi Pavan, >>>> >>>> I'm sorry for delay with answer. >>>> >>>> If ingress uses regular Path MTU discovery mechanism, it could produce >>>> value of MTU lower than actual one. This is because ingress doesn't kn= ow >>>> MTU per each hop. Let's consider case with four routers: R1 - R2 - R3 = - R4. >>>> MTU for R1-R2 link is 2000, MTU for R2-R3 is 1600 and MTU for R3-R4 is >>>> 2000. By virtue of regular Path MTU discovery mechanism R1 will derive= from >>>> FLOWSPEC that path MTU is 1600. As soon as R1 doesn't know how many la= bels >>>> in the stack will be on the lowest MTU hop, it can only set LSP MTU to= most >>>> conservative value (1600 - 4 - 4 =3D 1592, provided that R4 has advert= ised >>>> implicit null label). In fact actual path MTU is 1596 (1600 - 4 on R2-= R3 >>>> hop). Of course, it could be acceptable, but calculated LSP MTU as mor= e >>>> lower than actual as longer LSP path. For correct path MTU discovery >>>> ingress needs to know MTU per each hop. >>>> >>>> Thank you. >>>> >>>> 6 =D1=81=D0=B5=D0=BD=D1=82. 2018 =D0=B3., =D0=B2 18:27, Vishnu Pavan B= eeram >>>> =D0=BD=D0=B0=D0=BF=D0=B8=D1=81=D0=B0=D0=BB(=D0=B0): >>>> >>>> Alexander, Hi! >>>> >>>> I apologize for the delayed response. >>>> >>>> This draft does not propose any changes to the standard RSVP MTU >>>> signaling procedures (Int Serv object specific signaling procedures). = After >>>> the initial signaling sequence is complete, an ingress implementation >>>> (RFC3209) would typically take the path MTU learnt via signaling, run = it >>>> through some local logic and then arrive at an MTU value that can be >>>> assigned to the LSP. This local logic typically involves deducting the >>>> number of bytes in the label stack used for the LSP from the path MTU >>>> learnt via signaling. The ingress implementation supporting this draft= will >>>> rely on the Resv RRO to accurately determine the max-number of labels >>>> pushed along the path of the LSP (note that with delegation, downstrea= m >>>> hops can impose label stacks) and account for it in the local logic us= ed to >>>> arrive at the MTU value assigned to the LSP. >>>> >>>> I hope this addresses your question.. >>>> >>>> Regards, >>>> -Pavan >>>> >>>> On Tue, Aug 7, 2018 at 12:22 PM Alexander Okonnikov < >>>> alexander.okonnikov@gmail.com> wrote: >>>> >>>>> Hi Pavan, >>>>> >>>>> >>>>> Regular RSVP-TE LSPs use standard RSVP path MTU discovery mechanism. >>>>> That one cannot be used "as is" for approach described in the draft, = and >>>>> the draft doesn't address path MTU identification. Is it to be consid= ered? >>>>> >>>>> >>>>> Thank you. >>>>> >>>>> 26.07.2018 06:07, Vishnu Pavan Beeram =D0=BF=D0=B8=D1=88=D0=B5=D1=82: >>>>> >>>>> Chairs, Hi! >>>>> >>>>> As mentioned (in our presentation) in last week's WG session, we >>>>> believe that the draft is sufficiently baked and ready to progress to= the >>>>> next stage. We would like to formally request this draft to be consid= ered >>>>> for WG LC. >>>>> >>>>> Regards, >>>>> - Pavan (on behalf of the authors) >>>>> >>>>> >>>>> _______________________________________________ >>>>> mpls mailing listmpls@ietf.orghttps://www.ietf.org/mailman/listinfo/m= pls >>>>> >>>>> >>>>> _______________________________________________ >>>>> mpls mailing list >>>>> mpls@ietf.org >>>>> https://www.ietf.org/mailman/listinfo/mpls >>>>> >>>> >>>> >>>> >>>> >>> _______________________________________________ >> mpls mailing list >> mpls@ietf.org >> https://www.ietf.org/mailman/listinfo/mpls >> > --0000000000004c49640576d6285e Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable

Matt, Hi!

=C2=A0

The purpose of this document is to provide a mechanism to bring up MPLS RSVP-TE= LSPs on a shared forwarding plane. There is a set of key RSVP-TE features/= functionalities specified in the Introduction Section that would continue to w= ork with this forwarding plane (without needing any special extensions). The authors believe that a deploy= able solution can be implemented with this set of features/functionalities.

=C2=A0

If there are any other desirable features/functionalities (that are not specified as supported in this document) that can be supported on this forwarding plane = by introducing a few protocol extensions, then those are expected to be dis= cussed separately (draft-chandra-mpls-rsvp-shared-labels-np is an example of that).

=C2=A0

Regards,<= /p>

-Pavan


On Wed, Sep 26, 2018 = at 10:56 AM Matt Hartley <mha= rtley.ietf@gmail.com> wrote:
Pavan,

The problem with lis= ts like this (in general) is that it's not clear what the status is for= anything that isn't on the list, and most lists will probably manage t= o omit *something*. I think there's three reasonable options:

1. State unequivocally that everything works
2. L= ist the functionalities that definitely don't work, and make it clear t= hat everything else does
3. List the functionalities that def= initely work, and explicitly make no guarantees about anything that isn'= ;t listed.

It isn't really clear what you'= re trying to do here.

Note that in this case optio= n 1 isn't on the table because we've already established that nnhop= FRR requires further extensions.

Cheers
=

Matt

On Wed, Sep 26, 2018 at 10:17 AM Vishnu Pavan Beeram <vishnupavan@gmail.com<= /a>> wrote:

Alexander Hi!

=C2=A0

The intent of the following statement in Section 1 is certainly not to be evasi= ve (slightly or otherwise).

=C2=A0=C2=A0 Functionalities s= uch as bandwidth admission control, LSP

=C2=A0=C2=A0 priorities, preem= ption, auto-bandwidth and Fast Reroute

=C2=A0=C2=A0 continue to work = with this forwarding plane.

=C2=A0

We (the authors) still don=E2= =80=99t see any problem with the above statement. But we=E2=80=99ll go ahead and make a= slight adjustment (see below) to address your concern.

=C2=A0=C2=A0 Key functionaliti= es such as bandwidth admission control, LSP

=C2=A0=C2=A0 priorities, preem= ption, auto-bandwidth and Fast Reroute via

=C2=A0=C2=A0 facility backup p= rotection continue to work with this

=C2=A0=C2=A0 forwarding plane.=

=C2=A0

Regards,<= /p>

-Pavan

On Tue, Sep 25, 2018 at= 11:34 AM Alexander Okonnikov <alexander.okonnikov@gmail.com> wrote:
<= /div>
Hi Panav,

Ok, 1:1 is not to be= supported by this approach. In general, 1:1 has its own benefits =C2=A0- f= or example, it is more attractive versus N:1 in ring topologies. After all,= it is FRR too, as N:1 one. My point was that description in section 1 is s= lightly evasive regarding FRR and other RSVP-TE properties (PMTUD).

Thank you.


16 =D1=81=D0=B5=D0=BD=D1=82. 2018 =D0=B3., =D0=B2 4:38, Vishnu Pava= n Beeram <vis= hnupavan@gmail.com> =D0=BD=D0=B0=D0=BF=D0=B8=D1=81=D0=B0=D0=BB(=D0= =B0):

Alexander, Hi!

= =C2=A0

Please see inline for responses (prefixed VPB).

=C2=A0

Regards,
-Pavan

On Thu, Sep 13, 2018 at 1:3= 7 PM Alexander Okonnikov <alexander.okonnikov@gmail.com> wrote:
=
Hi Panav,

From section 7:


"If the label type is a delega= tion label, then the stacking procedure stops at that delegation hop."=
It is OK for "Stack to Reach Delegation Hop&= quot; approach, but it doesn't work for "Stack to Reach Egress&quo= t;, isn't it?

[VPB] The logic specified in Section 7 could be used by any node constructing the lab= el stack (this could be the ingress or a delegation hop). The sentence immedia= tely following the above quoted sentence in Section 7 is important. It currently= reads =E2=80=93
Approaches in Section 5.1<= /span> SHOULD be used to determine how the delegation labels are pushed in the label stack.
<= p class=3D"MsoNormal" style=3D"margin:0in 0in 0.0001pt;font-size:11pt;font-= family:"Calibri",sans-serif">=C2=A0

=
The intent here is to say that if you encounter a delegation label, use the procedures outlined in Section 5.1 to determine how the delegation labels a= re pushed in the label stack. The following change to the text should address this comment:

OLD:
If the label type is a delegation label, then the stac=
king procedure stops at that delegation hop.=C2=A0 
Approac=
hes in Section =
5.1 SHOULD be used to determine how the delegation labels are pu=
shed in the label stack.

=C2=A0

NEW:
If the label type is a delegation label, then the type of stacking approach chosen by the ingress for this LS= P (Section 5.1) MUST be used to determine how the delegation labels are pushed in the = label stack.

=C2=A0

Also, regardi= ng FRR support. Section 1 says:

"Function= alities such as bandwidth admission control, LSP priorities, preemption,=C2= =A0
auto-bandwidth and Fast Reroute continue to work with thi= s forwarding plane."
It seems that shared lab= els approach supports only facility bypass link-protection. It doesn't = support one-to-one link- and node-protection, per my understanding. Facilit= y bypass node-protection is not supported as well (as mentioned in Section = 8). Hence, FRR support is very limited, and section 1 needs correction.

[VPB] I don= =E2=80=99t see anything wrong with the quoted text. Fast Reroute for MPLS-TE LSPs can = be realized by either the 1-to-1 protection mechanism (detours) or the facility bypass = mechanism. The authors don=E2=80=99t intend to add procedures for 1-to-1 link/node pro= tection (who needs it?). =C2=A0The facility bypass link-protection procedure is discussed in this draft. The facility bypass node-protection procedure is discussed in https://tools.ietf.org/html/draft-= chandra-mpls-rsvp-shared-labels-np-00 (this was presented at the last IETF).
=C2=A0

Thank you.

13 =D1=81=D0=B5=D0=BD=D1=82. 2018 =D0= =B3., =D0=B2 20:28, Alexander Okonnikov <alexander.okonnikov@gmail.com> = =D0=BD=D0=B0=D0=BF=D0=B8=D1=81=D0=B0=D0=BB(=D0=B0):

Hi Panav,

Questions regar= ding ETLD:

The draft is not clear about signaling = of ETLD attribute. It says that ETLD is conveyed as per-hop attribute. Is m= y understanding correct that it is conveyed as RRO Hop attribute? Probably = it could be cleaned to avoid confusion whether ERO or RRO Hop attribute mec= hanism is used.

Next, the draft says that:

"... If a node is reached where the ETLD set fro= m the previous hop is 1, then that
node MUST select itself as= the delegation hop.=C2=A0 If a node is reached and it is=C2=A0
<= span class=3D"m_-1830687956985248383m_1343084799946502028m_2764878390082138= 059m_879809512118735882Apple-tab-span" style=3D"white-space:pre-wrap"> determined that this hop cannot receive more than one transport label, t= hen that node=C2=A0
MUST select itself as the delegation hop.= ..."

What is purpose of the second sentence/= rule?

Next:

"I= f there is a node or a sequence of nodes along the path of the LSP that do = not=C2=A0
support ETLD, then the immediate hop that supports = ETLD MUST select itself as the=C2=A0
delegation hop."

If some node (consecutive nodes) doesn't support= ETLD then it doesn't support TE labels. Hence, that node (regular RSVP= -TE LSR) will do SWAP and not POP. As a result non-decremented ETLD is OK a= nd immediate hop that supports ETLD not necessary should become delegation = hop?

Also, from Section 9.7:

<= div>"The ETLD field specifies the maximum number of transport labels t= hat this hop can potentially send to its downstream hop.=C2=A0 It MUST be s= et to a non-zero value."

Strictly speaking it= is not correct. ETLD reflects decrementing counter and not capability of s= ome transit node. I.e. if we consider LSP R1-R2-R3, R1 puts value 5 in ETLD= ,and R2 supports imposing of 2 labels, it doesn't mean that R2 should r= ewrite ETLD with value 2. It just should decrement value 5. Correct?
<= div>
Also, per my understanding it is supposed that in fact E= TLD value will not be just decremented, but it will be copied from previous= RRO Hop attributes subobject into being inserted RRO Hop attributes subobj= ect with decrementing. May be it would be better to signal ETLD value in LS= P Attributes object (and each capable node decrements ETLD value there), wh= ile signaling support of ETLD itself in RRO Hop attributes subobject?
=

Thank you.

13 =D1=81=D0=B5=D0=BD=D1=82. 2018 =D0=B3., =D0=B2 20:22, Alexander Okonnik= ov <a= lexander.okonnikov@gmail.com> =D0=BD=D0=B0=D0=BF=D0=B8=D1=81=D0=B0= =D0=BB(=D0=B0):

=
Hi Pa= van,

I'm sorry for delay with answer.

If ingress uses regular Path MTU discovery mechanism, it c= ould produce value of MTU lower than actual one. This is because ingress do= esn't know MTU per each hop. Let's consider case with four routers:= R1 - R2 - R3 - R4. MTU for R1-R2 link is 2000, MTU for R2-R3 is 1600 and M= TU for R3-R4 is 2000. By virtue of regular Path MTU discovery mechanism R1 = will derive from FLOWSPEC that path MTU is 1600. As soon as R1 doesn't = know how many labels in the stack will be on the lowest MTU hop, it can onl= y set LSP MTU to most conservative value (1600 - 4 - 4 =3D 1592, provided t= hat R4 has advertised implicit null label). In fact actual path MTU is 1596= (1600 - 4 on R2-R3 hop). Of course, it could be acceptable, but calculated= LSP MTU as more lower than actual as longer LSP path. For correct path MTU= discovery ingress needs to know MTU per each hop.

Thank you.

6 =D1=81=D0=B5=D0= =BD=D1=82. 2018 =D0=B3., =D0=B2 18:27, Vishnu Pavan Beeram <vishnupavan@gmail.com>= ; =D0=BD=D0=B0=D0=BF=D0=B8=D1=81=D0=B0=D0=BB(=D0=B0):

Ale= xander, Hi!
=C2=A0
I apologize for the delayed response.
=C2=A0This draft does not propose any changes to the standard RSVP MTU signaling= procedures (Int Serv object specific signaling procedures). After the init= ial signaling sequence is complete, an ingress implementation (RFC3209) wou= ld typically take the path MTU learnt via signaling, run it through some lo= cal logic and then arrive at an MTU value that can be assigned to the LSP. = This local logic typically involves deducting the number of bytes in the la= bel stack used for the LSP from the path MTU learnt via signaling. The ingr= ess implementation supporting this draft will rely on the Resv RRO to accur= ately determine the max-number of labels pushed along the path of the LSP (= note that with delegation, downstream hops can impose label stacks) and acc= ount for it in the local logic used to arrive at the MTU value assigned to = the LSP.
=C2=A0
I hope this addresses your question..
=C2=A0
Re= gards,
-Pavan

On Tue, Aug 7, 2018 at 12:22 PM Alexander Okonnikov <alexander.okonnikov@gm= ail.com> wrote:
=20 =20 =20

Hi Pavan,


Regular RSVP-TE LSPs use standard RSVP path MTU discovery mechanism. That one cannot be used "as is" for approach des= cribed in the draft, and the draft doesn't address path MTU identification. Is it to be considered?


Thank you.


26.07.201= 8 06:07, Vishnu Pavan Beeram =D0=BF=D0=B8=D1=88=D0=B5=D1=82:
=20
Chairs, Hi!

As mentioned (in our presentation) in last week's WG session, we believe that the draft is sufficiently baked and ready to progress to the next stage. We would like to formally request this draft to be considered for WG LC.

Regards,
- Pavan (on behalf of the authors)


_______________________________________________
mpls mailing list
mpls@ietf.org
https://www.ie=
tf.org/mailman/listinfo/mpls

_______________________________________________
mpls mailing list
mpls@ietf.org
https://www.ietf.org/mailman/listinfo/mpls




_______________________________________________
mpls mailing list
mpls@ietf.org
https://www.ietf.org/mailman/listinfo/mpls
--0000000000004c49640576d6285e--