RE: Compressed Routing Header idea

Ron, in my application I will be using IPv6-in-IPv6-in-IP encapsulation. So, we have
an inner IPv6 layer, an IPv6 mid-layer and an outer IP layer (could be IPv4 or IPv6).
I want to apply Segment Routing at both the inner IPv6 layer and the IPv6 mid-layer.

It is the IPv6 mid-layer that I have been referring to in my past few messages, but
I am now of the opinion that what I need for that would be something more special
purpose and limited in scope in the spirit of Routing Header Type 2. I will define that
in my document in the same fashion as was done in RFC6275. For the inner IPv6
layer, I would apply a general-purpose Routing Header and I can make do with
just 16 bits so I believe your 16-bit CRH would be a good fit for that.

In terms of removing the Routing Header, it is the mid-layer RH and not the inner
CRH that I am proposing to remove. And since I control the insertion and deletion
of the mid-layer header (plus there will not be any AH) I don't see why I can't
remove the RH at the penultimate hop *at that layer*. The inner layer RH
could still be left alone and, as you say, it should be comparatively small.
What do you think?

Thanks - Fred

From: Ron Bonica [mailto:rbonica@juniper.net]
Sent: Thursday, May 21, 2020 7:22 PM
To: Templin (US), Fred L <Fred.L.Templin@boeing.com>; Gyan Mishra <hayabusagsm@gmail.com>; 6MAN <6man@ietf.org>
Cc: IPv6 List <ipv6@ietf.org>; Pascal Thubert (pthubert) <pthubert@cisco.com>
Subject: RE: Compressed Routing Header idea

Fred,

Typically, the CRH is very short. An 8-byte CRH can carry 2 SIDs. A 16-byte CRH can carry 6 SIDs.

It doesn't really save that much bandwidth.

                                                      Ron

Juniper Business Use Only
From: Templin (US), Fred L <Fred.L.Templin@boeing.com<mailto:Fred.L.Templin@boeing.com>>
Sent: Thursday, May 21, 2020 9:44 PM
To: Ron Bonica <rbonica@juniper.net<mailto:rbonica@juniper.net>>; Gyan Mishra <hayabusagsm@gmail.com<mailto:hayabusagsm@gmail.com>>; 6MAN <6man@ietf.org<mailto:6man@ietf.org>>
Cc: IPv6 List <ipv6@ietf.org<mailto:ipv6@ietf.org>>; Pascal Thubert (pthubert) <pthubert@cisco.com<mailto:pthubert@cisco.com>>
Subject: RE: Compressed Routing Header idea

[External Email. Be cautious of content]

Hi Ron,

What I am concerned with  is the case where the ultimate hop is over a very slow link
(e.g., 100Kbps or less). Any form of compression can help, and compressing away a
Routing Header could result in a small savings - but a savings nonetheless.

So, why not just truncate it? The final destination won't miss it and will never be
aware that it was ever there in the first place.

Thanks - Fred

From: Ron Bonica [mailto:rbonica@juniper.net]
Sent: Thursday, May 21, 2020 3:33 PM
To: Templin (US), Fred L <Fred.L.Templin@boeing.com<mailto:Fred.L.Templin@boeing.com>>; Gyan Mishra <hayabusagsm@gmail.com<mailto:hayabusagsm@gmail.com>>; 6MAN <6man@ietf.org<mailto:6man@ietf.org>>
Cc: IPv6 List <ipv6@ietf.org<mailto:ipv6@ietf.org>>; Pascal Thubert (pthubert) <pthubert@cisco.com<mailto:pthubert@cisco.com>>
Subject: RE: Compressed Routing Header idea

Fred,

When a node decrements Segments Left to 0, it causes all subsequent nodes to ignore the Routing header. So, I don't see a good reason to remove the Routing header. It will be ignored by all downstream nodes anyway.

                                                          Ron

Juniper Business Use Only
From: Templin (US), Fred L <Fred.L.Templin@boeing.com<mailto:Fred.L.Templin@boeing.com>>
Sent: Thursday, May 21, 2020 4:00 PM
To: Gyan Mishra <hayabusagsm@gmail.com<mailto:hayabusagsm@gmail.com>>; 6MAN <6man@ietf.org<mailto:6man@ietf.org>>; Ron Bonica <rbonica@juniper.net<mailto:rbonica@juniper.net>>
Cc: IPv6 List <ipv6@ietf.org<mailto:ipv6@ietf.org>>; Pascal Thubert (pthubert) <pthubert@cisco.com<mailto:pthubert@cisco.com>>
Subject: RE: Compressed Routing Header idea

[External Email. Be cautious of content]

Hi, just seeing this question now (below). The idea is that the router that processes
the penultimate SID would "pair" it with the ultimate SID so that the penultimate SID
is written as the final IPv6 destination while the ultimate SID (which may include an
address and port number) is used as a destination for IPv6-in-IP encapsulation. So,
the final hop router before the final destination would be the one to extract it.

I had a somewhat related question - can the final hop router before the final
destination delete the Routing Header before forwarding?

Thanks - Fred

Fred

Curious what would be the particular application use case for variable compressed routing header to add ancillary info like port or other miscellaneous info.

 I am guessing the final destination would have to extract the ancillary.  In a connection the tcp or udp source is the same unless using RPC or an app using dynamic port allocation and want to save the port info somewhere else.

https://datatracker.ietf.org/doc/draft-templin-6man-crh-variable/<https://urldefense.com/v3/__https:/datatracker.ietf.org/doc/draft-templin-6man-crh-variable/__;!!NEt6yMaO-gk!ViLHxbEJqhubTR_jBFZ9qv59Xn3jPyPGQcTN33jDxZuo2nzpBinDu4TXl9lvoN3y$>

Thanks

Gyan

On Mon, May 18, 2020 at 11:12 AM Templin (US), Fred L <Fred.L.Templin@boeing.com<mailto:Fred.L.Templin@boeing.com>> wrote:
Pascal, thanks I did not know about this but at first glance I do not believe RFC8138 fully
satisfies what I need. First, I want to be able to support both left-side (most significant
bits) and right-side (least significant bits) compression. Second, I want to be able to
compress to the byte granularity for any length from 0 to 16 bytes. And, third, I want
to be able to include an ancillary piece of information (e.g., an application port number)
with each IPv6 address. So, I submitted a short draft showing the format that I would
see as being flexible to support my use case and I think perhaps many other:

https://datatracker.ietf.org/doc/draft-templin-6man-crh-variable/<https://urldefense.com/v3/__https:/datatracker.ietf.org/doc/draft-templin-6man-crh-variable/__;!!NEt6yMaO-gk!ViLHxbEJqhubTR_jBFZ9qv59Xn3jPyPGQcTN33jDxZuo2nzpBinDu4TXl9lvoN3y$>

I did include a reference to RFC8138 - let me know your thoughts.

Fred

> -----Original Message-----
> From: Pascal Thubert (pthubert) [mailto:pthubert@cisco.com<mailto:pthubert@cisco.com>]
> Sent: Monday, May 18, 2020 7:55 AM
> To: Templin (US), Fred L <Fred.L.Templin@boeing.com<mailto:Fred.L.Templin@boeing.com>>; IPv6 List <ipv6@ietf.org<mailto:ipv6@ietf.org>>
> Subject: RE: Compressed Routing Header idea
>
> Hello Fred:
>
> Are you aware of RFC 8138? See https://tools.ietf.org/html/rfc8138#section-5.1<https://urldefense.com/v3/__https:/tools.ietf.org/html/rfc8138*section-5.1__;Iw!!NEt6yMaO-gk!ViLHxbEJqhubTR_jBFZ9qv59Xn3jPyPGQcTN33jDxZuo2nzpBinDu4TXl0b3-Gbz$>
> The addresses in the source route header can be compressed as follows:
>
> "
>
>      +-----------+----------------------+
>      |   6LoRH   | Length of compressed |
>      |   Type    | IPv6 address (bytes) |
>      +-----------+----------------------+
>      |    0      |       1              |
>      |    1      |       2              |
>      |    2      |       4              |
>      |    3      |       8              |
>      |    4      |      16              |
>      +-----------+----------------------+
>
>                        Figure 7: The SRH-6LoRH Types
>
> "
> You need multiple SRH-6loRH if you have different sizes to accommodate.
>
> Keep safe
>
> Pascal
>
> > -----Original Message-----
> > From: ipv6 <ipv6-bounces@ietf.org<mailto:ipv6-bounces@ietf.org>> On Behalf Of Templin (US), Fred L
> > Sent: lundi 18 mai 2020 16:04
> > To: IPv6 List <ipv6@ietf.org<mailto:ipv6@ietf.org>>
> > Subject: Compressed Routing Header idea
> >
> > Hi, I have a use case where some IPv6 addresses that would go into a routing
> > header are more compressible than others and so I am wondering if some kind
> > of "hybrid" compressed routing header would be possible. For example, if one
> > address can be compressed down to
> > 16 bits, then include only those 16 bits; if a different address can only be
> > compressed down to 32 bits, then include the 32 bits; if yet a different address
> > cannot be compressed at all, then include all 128 bits. And, there may be many
> > more sizes in between.
> >
> > RFC4191 Section 2.3 shows an example of how an IPv6 prefix/address can be
> > compressed to a variable length. Essentially, a length byte followed by a
> > variable-length prefix. That way there would still be "pretty good compression"
> > albeit with an extra byte per prefix. And, it would be a generalized form that
> > would only require a single routing header type value.
> > How would it be if we did something like that?
> >
> > Fred
> >
> >
> >
> > --------------------------------------------------------------------
> > IETF IPv6 working group mailing list
> > ipv6@ietf.org<mailto:ipv6@ietf.org>
> > Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6<https://urldefense.com/v3/__https:/www.ietf.org/mailman/listinfo/ipv6__;!!NEt6yMaO-gk!ViLHxbEJqhubTR_jBFZ9qv59Xn3jPyPGQcTN33jDxZuo2nzpBinDu4TXl0Q2OkFT$>
> > --------------------------------------------------------------------

--------------------------------------------------------------------
IETF IPv6 working group mailing list
ipv6@ietf.org<mailto:ipv6@ietf.org>
Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6<https://urldefense.com/v3/__https:/www.ietf.org/mailman/listinfo/ipv6__;!!NEt6yMaO-gk!ViLHxbEJqhubTR_jBFZ9qv59Xn3jPyPGQcTN33jDxZuo2nzpBinDu4TXl0Q2OkFT$>
--------------------------------------------------------------------
--
Gyan  Mishra
Network Engineering & Technology
Verizon
Silver Spring, MD 20904
Phone: 301 502-1347
Email: gyan.s.mishra@verizon.com<mailto:gyan.s.mishra@verizon.com>