Re: [Bier] I-D Action: draft-ietf-bier-te-isis-01.txt

Toerless Eckert <> Wed, 03 August 2022 04:04 UTC

Return-Path: <>
Received: from localhost (localhost []) by (Postfix) with ESMTP id E10B3C16ECA0 for <>; Tue, 2 Aug 2022 21:04:49 -0700 (PDT)
X-Virus-Scanned: amavisd-new at
X-Spam-Flag: NO
X-Spam-Score: -1.657
X-Spam-Status: No, score=-1.657 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HEADER_FROM_DIFFERENT_DOMAINS=0.248, RCVD_IN_DNSWL_BLOCKED=0.001, RCVD_IN_ZEN_BLOCKED_OPENDNS=0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, T_SCC_BODY_TEXT_LINE=-0.01, URIBL_BLOCKED=0.001, URIBL_DBL_BLOCKED_OPENDNS=0.001, URIBL_ZEN_BLOCKED_OPENDNS=0.001] autolearn=no autolearn_force=no
Received: from ([]) by localhost ( []) (amavisd-new, port 10024) with ESMTP id EQalGWe3rA9K for <>; Tue, 2 Aug 2022 21:04:46 -0700 (PDT)
Received: from ( []) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) by (Postfix) with ESMTPS id 3E026C14F740 for <>; Tue, 2 Aug 2022 21:04:45 -0700 (PDT)
Received: from ( [IPv6:2001:638:a000:4134::ffff:51]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (4096 bits) server-digest SHA256) (No client certificate requested) by (Postfix) with ESMTPS id 0BDDB549C4C; Wed, 3 Aug 2022 06:04:41 +0200 (CEST)
Received: by (Postfix, from userid 10463) id EA21D4EB643; Wed, 3 Aug 2022 06:04:40 +0200 (CEST)
Date: Wed, 03 Aug 2022 06:04:40 +0200
From: Toerless Eckert <>
To: Huaimo Chen <>
Cc: "Les Ginsberg (ginsberg)" <>, "" <>
Message-ID: <>
References: <> <> <>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Disposition: inline
In-Reply-To: <>
Archived-At: <>
Subject: Re: [Bier] I-D Action: draft-ietf-bier-te-isis-01.txt
X-Mailman-Version: 2.1.39
Precedence: list
List-Id: "\"Bit Indexed Explicit Replication discussion list\"" <>
List-Unsubscribe: <>, <>
List-Archive: <>
List-Post: <>
List-Help: <>
List-Subscribe: <>, <>
X-List-Received-Date: Wed, 03 Aug 2022 04:04:50 -0000

Hey folks

I am trying to wrap my head around how one would use these IS-IS extensions for BIER-TE,
and what if any of such semantic information might help the document...

The way i assume this extension to be used is that some poor system/operator assigns BIER-TE
bits and configures the actual BIER-TE BIFT entries with forward_connected entries to
the neighbors. Similarily to how the same poor system/operator would configure interface
IP addresses on routers. This IS-IS work does not help to eliminate any of this initial provisioning. [Yes/No]

The way the IS-IS kicks in is to disseminate the information from those forward_connected BIER-TE
BIFT entries across the domain, so that for example ingres-PE could calculate BIER-TE paths through the
topology. [Yes/No].

If i am on the right track, then let me continue:

The first detail is that these IS-IS entries will seemingly have some elements from rfc5303
that identify the interface and/or neighbor. So those need to match what would be found in
the BIFT. I am not sure if it is obvious to every implementer how to achieve that without
additional guidance. For example i guess we ALSO need to have the IS-IS signaling for the
label ranges for the SD/SI in BIER. Have we checked thart we simply should/can use them also
for the SD that we use for BIER-TE ? I hope so, but i can not remember the discussion.
Likewise for BIER-TE over L2 (not MPLS), we have the dependency against picking one particular
bift-id encoding, and the "i don't know which" signaling we may have to support it (could
of course be all manual configured).

Yes, this is just dependencies to make a deployment work, not necessarily something an
IGP specialist wants to see in an IGP draft/RFC, but i hope we can have this more
holistic discussion here ;-)

So, then RFC5305 has all those wonderful adjacency parameters to help for example ingres
PE CSPF calculations. And i remember those from  RSVP-TE days. But now the ISIS draft
proposal introduces the BAR/IPA parameters, and i have never seen those being mentioned
or used in conjunction with TE in general or CSPF specifically. So it would be great
to have at least one example of how to use PAR/IPA be explained here. If we can not
construct such examples, then BIER/IPA may be leftovers from the IS-IS encodings for
BIER, where of course there is no CSPF and the like...

Next question: Would it be possible to express with these IS-IS option the forward_routed()
I think it would be great if we would build the IS-IS extensions such that that is possible,
because forward_routed entries can be a great help to get along with the available bits.

I also would need to get my head around those IS-IS pseudonodes... if anyone has an example,
that would be great.

Thanks so much!

On Wed, Aug 03, 2022 at 02:58:53AM +0000, Huaimo Chen wrote:
> Hi Les,
>     Thank you very much for your valuable comments.
>     I have addressed them in the updated draft (uploaded).
>     My responses are also inline below with [HC].
> Best Regards,
> Huaimo
> -----Original Message-----
> From: BIER <> On Behalf Of Les Ginsberg (ginsberg)
> Sent: Friday, July 29, 2022 4:59 PM
> To:
> Subject: Re: [Bier] I-D Action: draft-ietf-bier-te-isis-01.txt
> Some pedantic comments on this draft:
> For the new BIER-TE sub-TLV:
> 1)The relevant registry has been renamed to be "IS-IS Sub-TLVs for TLVs Advertising Neighbor Information". Please use this new name.
> [HC]: I have renamed it as you suggested.
> 2)Please do NOT insert padding to keep content aligned on a 4 byte boundary. This may be useful for OSPF, but it is not for IS-IS as TLVs are NOT aligned inside an LSP. All you are doing is wasting space - which is precious in IS-IS given the limited LSP space.
> [HC]: Removed padding.
> 3)You say in Section 2:
> "Note that if each of BitPosition
>    and DrEndBitPosition uses more than 2 octets, we use 4 or more octets
>    for each of them."
> I do not see how the receiver of the sub-TLV could tell whether the 2 octet or 4 octet encoding was sent as you have nothing that specifies the length of these fields.
> You can't use the total length of the sub-TLV as it could include optional sub-sub-TLVs.
> ??
> [HC]: Deleted the quoted text.
> 4)In IANA section, I do not know why you specified "N" for TLVs 23 and 223.
> While RFC 5311 (which defines these TLVs) is not popular, if there was an implementation, I see no reason why it should be invalid to send the BIER-TE info in these TLVs.
> [HC]: Changed "n" to "y" for TLVs 23 and 223.
> Thanx.
>     Les
> > -----Original Message-----
> > From: BIER <> On Behalf Of
> > Sent: Friday, July 29, 2022 12:13 PM
> > To:
> > Cc:
> > Subject: [Bier] I-D Action: draft-ietf-bier-te-isis-01.txt
> > 
> > 
> > A New Internet-Draft is available from the on-line Internet-Drafts directories.
> > This draft is a work item of the Bit Indexed Explicit Replication WG of the IETF.
> > 
> >         Title           : IS-IS Extensions for BIER-TE
> >         Authors         : Huaimo Chen
> >                           Mike McBride
> >                           Aijun Wang
> >                           Gyan S. Mishra
> >                           Yanhe Fan
> >                           Lei Liu
> >                           Xufeng Liu
> >   Filename        : draft-ietf-bier-te-isis-01.txt
> >   Pages           : 7
> >   Date            : 2022-07-29
> > 
> > Abstract:
> >    This document describes IS-IS extensions for distributing
> >    BitPositions configured on the links in "Bit Index Explicit
> >    Replication Traffic Engineering" (BIER-TE) domain.
> > 
> > 
> > 
> > The IETF datatracker status page for this draft is:
> >;;sdata=99ojCa2m%2BTeMk0OfxeifsmVrhvNaK7jQJp0fu0syRN8%3D&amp;reserved=0
> > 
> > There is also an htmlized version available at:
> >;;sdata=%2F8qaH0ZvQBAfGM4WIre0jpcI6As7wjXjdqbxOfdGHxo%3D&amp;reserved=0
> > 
> > A diff from the previous version is available at:
> >;;sdata=L77o0dGWGuSwL9vSm1Durc56Fl5zONoDq%2F8ij12v9lc%3D&amp;reserved=0
> > 
> > 
> > Internet-Drafts are also available by rsync at
> > 
> > 
> > _______________________________________________
> > BIER mailing list
> >
> >;;sdata=em5K2lvVvXwkSlye1egSX%2BgtdfeqDArSsAycQJTPArI%3D&amp;reserved=0
> _______________________________________________
> BIER mailing list