Re: [sfc] draft-ietf-sfc-nsh-04 NSH-MD-type 1 use-cases

Actually I think the draft you are looking for is:

https://datatracker.ietf.org/doc/draft-napper-sfc-nsh-broadband-allocation<https://datatracker.ietf.org/doc/draft-napper-sfc-nsh-broadband-allocation/>

Jim

Sent from my iPhone

On May 6, 2016, at 10:48 AM, Dave Dolson <ddolson@sandvine.com<mailto:ddolson@sandvine.com>> wrote:

This expired draft proposed using CH2 and CH3 for sub/endpoint ID.
https://tools.ietf.org/html/draft-napper-sfc-nsh-mobility-allocation-02

I think it is worth reviving it.

-----Original Message-----
From: sfc [mailto:sfc-bounces@ietf.org] On Behalf Of Christian Koenning
Sent: Friday, May 06, 2016 9:30 AM
To: sfc@ietf.org<mailto:sfc@ietf.org>
Subject: [sfc] draft-ietf-sfc-nsh-04 NSH-MD-type 1 use-cases

Dear WG:

We are currently looking into SFC use-cases for mobile networks and are thinking of using the NSH Metadata to carry subscriber identity.

The ip-address to subscriber mapping in todays network is usually propagated via external control-planes, e.g. RADIUS accounting.

Looking at the possibilities  of NSH, the subscriber identity could easily be carried as metadata per packet.

However, the 2 most common identifiers in mobile networks are

- Mobile Station International Subscriber Directory Number (MSISDN)
 E.164 defines this to a maximum of 15 digits

- International Mobile Subscriber Identity (IMSI), also defined as a 15 digit number

Now looking at draft-ietf-sfc-nsh-04.txt p.12 3.4 NSH-MD-type 1 each Mandatory Context Header is defined as 4 Bytes.

    0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   |Ver|O|C|R|R|R|R|R|R|   Length  |  MD-type=0x1  | Next Protocol |
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   |          Service Path ID                      | Service Index |
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   |                Mandatory Context Header                       |
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   |                Mandatory Context Header                       |
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   |                Mandatory Context Header                       |
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   |                Mandatory Context Header                       |
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

As you can see, a 15 digit number cannot be represented with 4 bytes.
Doing the math it shows that  ceiling [ log2 ( ( 10^16 ) -1 ) ]  = 54 bits would be needed
for simple binary representation.

Question: Are there any plans to address this in the MD type 1, for example by defining
concatenation of two Mandatory Context Headers, which then with 64 bits or 8 byte would fit the subscriber identity ?

Many Thanks

Christian Koenning

_______________________________________________
sfc mailing list
sfc@ietf.org<mailto:sfc@ietf.org>
https://www.ietf.org/mailman/listinfo/sfc

_______________________________________________
sfc mailing list
sfc@ietf.org<mailto:sfc@ietf.org>
https://www.ietf.org/mailman/listinfo/sfc