Re: [bess] [sfc] [mpls] Progress with draft-farrel-mpls-sfc

[mohamed.boucadair]

Dear Avinash,

But these various encapsulation options (VXLAN, IP-in-IP,..) are already there 
thanks to the transport-agnostic approach endorsed by the SFC WG.

An alternative approach would be to mimic NSH for each of the mechanisms you 
listed. That approach is suboptimal, IMO.

FWIW, we used to have this text in the SFC Control Plane Requirements I-D:

   Various transport encapsulation schemes and/or versions of SFC header
   implementations may be supported by one or several nodes of an SFC-
   enabled domain.  For the sake of coherent configuration, the SFC
   control plane is responsible for instructing all the involved SFC
   data plane functional elements about the behavior to adopt to select
   the transport encapsulation scheme(s), the version of the SFC header
   to enable, etc.

Cheers,
Med

De : LINGALA, AVINASH [mailto:ar9...@att.com]
Envoyé : mardi 20 mars 2018 10:06
À : BOUCADAIR Mohamed IMT/OLN; UTTARO, JAMES; Henderickx, Wim (Nokia - 
BE/Antwerp); Robert Raszuk; Adrian Farrel
Cc : mpls; SPRING WG List; bess@ietf.org; s...@ietf.org
Objet : RE: [sfc] [mpls] Progress with draft-farrel-mpls-sfc


I support this effort and I agree with Jim.  As an operator we would like to 
see various encapsulation options for SFC. This would help an operator to pick 
the suitable encapsulation option for their networks.

Same approach that IETF took for EVPN  with various encap options like MPLS, 
VXLAN, GENEVE,..

​
Thanks,
Avinash Lingala


From: BESS [mailto:bess-boun...@ietf.org] On Behalf Of 
mohamed.boucad...@orange.com<mailto:mohamed.boucad...@orange.com>
Sent: Monday, March 19, 2018 10:25 AM
To: UTTARO, JAMES <ju1...@att.com<mailto:ju1...@att.com>>; Henderickx, Wim 
(Nokia - BE/Antwerp) 
<wim.henderi...@nokia.com<mailto:wim.henderi...@nokia.com>>; Robert Raszuk 
<rob...@raszuk.net<mailto:rob...@raszuk.net>>; Adrian Farrel 
<adr...@olddog.co.uk<mailto:adr...@olddog.co.uk>>
Cc: mpls <m...@ietf.org<mailto:m...@ietf.org>>; SPRING WG List 
<spr...@ietf.org<mailto:spr...@ietf.org>>; bess@ietf.org<mailto:bess@ietf.org>; 
s...@ietf.org<mailto:s...@ietf.org>
Subject: Re: [bess] [sfc] [mpls] Progress with draft-farrel-mpls-sfc

Re-,

This is really a matter of taste.

Jim, whatever scheme we use for identifying service chains, there are 
requirements/constraints/new practices/new OAM procedures that need to be 
supported/honored for service chaining purposes.

Those are not simple nor complex in MPLS vs. NSH over MPLS. I’m wrong?

Cheers,
Med

De : UTTARO, JAMES [mailto:ju1...@att.com]
Envoyé : lundi 19 mars 2018 14:10
À : BOUCADAIR Mohamed IMT/OLN; Henderickx, Wim (Nokia - BE/Antwerp); Robert 
Raszuk; Adrian Farrel
Cc : mpls; SPRING WG List; s...@ietf.org<mailto:s...@ietf.org>; 
bess@ietf.org<mailto:bess@ietf.org>
Objet : RE: [sfc] [mpls] Progress with draft-farrel-mpls-sfc

Med,

When I say simply, I am speaking as an operator. The 
operations, systems, tools, institutional knowledge etc… in this space is 
around MPLS. There is a simpler path to creating simple chains by using MPLS 
instead of introducing a new encap.

Thanks,
Jim Uttaro

From: mohamed.boucad...@orange.com<mailto:mohamed.boucad...@orange.com> 
[mailto:mohamed.boucad...@orange.com]
Sent: Monday, March 19, 2018 10:03 AM
To: UTTARO, JAMES <ju1...@att.com<mailto:ju1...@att.com>>; Henderickx, Wim 
(Nokia - BE/Antwerp) 
<wim.henderi...@nokia.com<mailto:wim.henderi...@nokia.com>>; Robert Raszuk 
<rob...@raszuk.net<mailto:rob...@raszuk.net>>; Adrian Farrel 
<adr...@olddog.co.uk<mailto:adr...@olddog.co.uk>>
Cc: mpls <m...@ietf.org<mailto:m...@ietf.org>>; SPRING WG List 
<spr...@ietf.org<mailto:spr...@ietf.org>>; s...@ietf.org<mailto:s...@ietf.org>; 
bess@ietf.org<mailto:bess@ietf.org>
Subject: RE: [sfc] [mpls] Progress with draft-farrel-mpls-sfc

Re-,

I’m afraid that you cannot ‘simply’ re-use MPLS for chaining purposes without 
any code upgrade.


NSH does provide the simple functionality you need; that is the information to 
identify a chain + avoid infinite loops. This is known as: MD Type 0x2 with 
length is 0x2.

Of course you can encode that information using another channel, but still code 
change is needed.

Please note that NSH is not at the same level as “GENEVE, VXLAN”.

Cheers,
Med

De : UTTARO, JAMES [mailto:ju1...@att.com]
Envoyé : lundi 19 mars 2018 13:48
À : BOUCADAIR Mohamed IMT/OLN; Henderickx, Wim (Nokia - BE/Antwerp); Robert 
Raszuk; Adrian Farrel
Cc : mpls; SPRING WG List; s...@ietf.org<mailto:s...@ietf.org>; 
bess@ietf.org<mailto:bess@ietf.org>
Objet : RE: [sfc] [mpls] Progress with draft-farrel-mpls-sfc

Med,

We run an MPLS network so there is no NSH deployed anywhere. I 
want to create simple chains that we can make available to our WAN customers 
and I want to keep it

Re: [bess] [sfc] [mpls] Progress with draft-farrel-mpls-sfc

[LINGALA, AVINASH]

I support this effort and I agree with Jim.  As an operator we would like to 
see various encapsulation options for SFC. This would help an operator to pick 
the suitable encapsulation option for their networks.

Same approach that IETF took for EVPN  with various encap options like MPLS, 
VXLAN, GENEVE,..

​
Thanks,
Avinash Lingala


From: BESS [mailto:bess-boun...@ietf.org] On Behalf Of 
mohamed.boucad...@orange.com
Sent: Monday, March 19, 2018 10:25 AM
To: UTTARO, JAMES <ju1...@att.com>; Henderickx, Wim (Nokia - BE/Antwerp) 
<wim.henderi...@nokia.com>; Robert Raszuk <rob...@raszuk.net>; Adrian Farrel 
<adr...@olddog.co.uk>
Cc: mpls <m...@ietf.org>; SPRING WG List <spr...@ietf.org>; bess@ietf.org; 
s...@ietf.org
Subject: Re: [bess] [sfc] [mpls] Progress with draft-farrel-mpls-sfc

Re-,

This is really a matter of taste.

Jim, whatever scheme we use for identifying service chains, there are 
requirements/constraints/new practices/new OAM procedures that need to be 
supported/honored for service chaining purposes.

Those are not simple nor complex in MPLS vs. NSH over MPLS. I’m wrong?

Cheers,
Med

De : UTTARO, JAMES [mailto:ju1...@att.com]
Envoyé : lundi 19 mars 2018 14:10
À : BOUCADAIR Mohamed IMT/OLN; Henderickx, Wim (Nokia - BE/Antwerp); Robert 
Raszuk; Adrian Farrel
Cc : mpls; SPRING WG List; s...@ietf.org<mailto:s...@ietf.org>; 
bess@ietf.org<mailto:bess@ietf.org>
Objet : RE: [sfc] [mpls] Progress with draft-farrel-mpls-sfc

Med,

When I say simply, I am speaking as an operator. The 
operations, systems, tools, institutional knowledge etc… in this space is 
around MPLS. There is a simpler path to creating simple chains by using MPLS 
instead of introducing a new encap.

Thanks,
Jim Uttaro

From: mohamed.boucad...@orange.com<mailto:mohamed.boucad...@orange.com> 
[mailto:mohamed.boucad...@orange.com]
Sent: Monday, March 19, 2018 10:03 AM
To: UTTARO, JAMES <ju1...@att.com<mailto:ju1...@att.com>>; Henderickx, Wim 
(Nokia - BE/Antwerp) 
<wim.henderi...@nokia.com<mailto:wim.henderi...@nokia.com>>; Robert Raszuk 
<rob...@raszuk.net<mailto:rob...@raszuk.net>>; Adrian Farrel 
<adr...@olddog.co.uk<mailto:adr...@olddog.co.uk>>
Cc: mpls <m...@ietf.org<mailto:m...@ietf.org>>; SPRING WG List 
<spr...@ietf.org<mailto:spr...@ietf.org>>; s...@ietf.org<mailto:s...@ietf.org>; 
bess@ietf.org<mailto:bess@ietf.org>
Subject: RE: [sfc] [mpls] Progress with draft-farrel-mpls-sfc

Re-,

I’m afraid that you cannot ‘simply’ re-use MPLS for chaining purposes without 
any code upgrade.


NSH does provide the simple functionality you need; that is the information to 
identify a chain + avoid infinite loops. This is known as: MD Type 0x2 with 
length is 0x2.

Of course you can encode that information using another channel, but still code 
change is needed.

Please note that NSH is not at the same level as “GENEVE, VXLAN”.

Cheers,
Med

De : UTTARO, JAMES [mailto:ju1...@att.com]
Envoyé : lundi 19 mars 2018 13:48
À : BOUCADAIR Mohamed IMT/OLN; Henderickx, Wim (Nokia - BE/Antwerp); Robert 
Raszuk; Adrian Farrel
Cc : mpls; SPRING WG List; s...@ietf.org<mailto:s...@ietf.org>; 
bess@ietf.org<mailto:bess@ietf.org>
Objet : RE: [sfc] [mpls] Progress with draft-farrel-mpls-sfc

Med,

We run an MPLS network so there is no NSH deployed anywhere. I 
want to create simple chains that we can make available to our WAN customers 
and I want to keep it simple from a technology and operations POV.. At this 
point I do not see the need to introduce NSH for what we need to do. I can 
simply re-use MPLS.

Not sure why NSH is the winner here there are folks who advocate for GENEVE, 
NSH, VXLAN etc… If IETF is pushing for one encap than it would be helpful to 
define the set of requirements/criteria and compare the encaps.

Thanks,
Jim Uttaro

From: mohamed.boucad...@orange.com<mailto:mohamed.boucad...@orange.com> 
[mailto:mohamed.boucad...@orange.com]
Sent: Monday, March 19, 2018 9:22 AM
To: UTTARO, JAMES <ju1...@att.com<mailto:ju1...@att.com>>; Henderickx, Wim 
(Nokia - BE/Antwerp) 
<wim.henderi...@nokia.com<mailto:wim.henderi...@nokia.com>>; Robert Raszuk 
<rob...@raszuk.net<mailto:rob...@raszuk.net>>; Adrian Farrel 
<adr...@olddog.co.uk<mailto:adr...@olddog.co.uk>>
Cc: mpls <m...@ietf.org<mailto:m...@ietf.org>>; SPRING WG List 
<spr...@ietf.org<mailto:spr...@ietf.org>>; s...@ietf.org<mailto:s...@ietf.org>; 
bess@ietf.org<mailto:bess@ietf.org>
Subject: RE: [sfc] [mpls] Progress with draft-farrel-mpls-sfc

Hi Jim,

Perhaps I missed your point, but I’m not asking to disallow whatever transport 
encapsulation scheme deployed in a network for SFC purposes.

What I’m saying is:
* the IETF has defined a generic SFC architecture and went with a 
tra

Re: [bess] [sfc] [mpls] Progress with draft-farrel-mpls-sfc

[NAPIERALA, MARIA H]

I also would like to point out that the service chain forwarding defined by 
draft-ietf-bess-service-chaining is based on original IP packet header. This 
allows to attract the traffic into the chain based on existence of a route to 
the traffic destination. In other words, it allows to insert service chains 
from a routed network.

Maria


From: BESS [mailto:bess-boun...@ietf.org] On Behalf Of Henderickx, Wim (Nokia - 
BE/Antwerp)
Sent: Sunday, March 18, 2018 4:02 PM
To: Robert Raszuk <rob...@raszuk.net>; Adrian Farrel <adr...@olddog.co.uk>
Cc: mpls <m...@ietf.org>; bess@ietf.org; s...@ietf.org
Subject: Re: [bess] [sfc] [mpls] Progress with draft-farrel-mpls-sfc

I echo robert’s comment, while I agree that draft-ietf-bess-service-chaining is 
useful work, we should not define yet another data-plane for it.


From: <rras...@gmail.com<mailto:rras...@gmail.com>> on behalf of Robert Raszuk 
<rob...@raszuk.net<mailto:rob...@raszuk.net>>
Date: Sunday, 18 March 2018 at 14:27
To: Adrian Farrel <adr...@olddog.co.uk<mailto:adr...@olddog.co.uk>>
Cc: "Henderickx, Wim (Nokia - BE/Antwerp)" 
<wim.henderi...@nokia.com<mailto:wim.henderi...@nokia.com>>, mpls 
<m...@ietf.org<mailto:m...@ietf.org>>, "s...@ietf.org<mailto:s...@ietf.org>" 
<s...@ietf.org<mailto:s...@ietf.org>>, "bess@ietf.org<mailto:bess@ietf.org>" 
<bess@ietf.org<mailto:bess@ietf.org>>
Subject: Re: [sfc] [mpls] Progress with draft-farrel-mpls-sfc

Adrian,

> draft-farrel-mpls-sfc provides another transition tool on the migration to 
> RFC 8300.

Very honestly to me it looks like a road block to faster adoption of NSH not as 
help to migrate to RFC 8300 in any way.

> It allows SFFs to be built as a minor mod to existing routers before there is 
> forwarding plane support for the NSH.

I don't agree with that. "MPLS support" in today's equipment is basic three 
label operations IMPOSE, POP & SWAP. I don't see how hardware which can support 
those three mechanisms can effectively play any role in what one would expect 
from NSH alternative.

And as I mentioned before there is no such a thing like MPLS *only* networks. 
One company tried to build MPLS only (without IP forwarding) router but they 
dropped that plan. So basic IP can carry NSH front ended packet to whichever 
device can process NSH headers.

Adding equivalent processing of now both NSH and MPLS headers (far from basic 
pop,swap, impose) operations does not seems like a useful standards and 
technology investment at this point.

Best,
Robert.







On Sun, Mar 18, 2018 at 11:10 AM, Adrian Farrel 
<adr...@olddog.co.uk<mailto:adr...@olddog.co.uk>> wrote:
Wim and Robert,

[Dropping SPRING at this point as (as previously discussed) we have taken / are 
taking SR out of this document]

I think that draft-ietf-bess-service-chaining is really important work: it 
expresses a technique that is implemented and shipping.

On the other hand, this approach is not fully consistent with RFC 7665.

But it does describe an actual SFC technology. Whether it remains in the field 
or is a migration technology only time (and operators) will tell.

Now, if we want to support RFC 7665 and RFC 8300 and use a control plane to 
discover the SFFs and to which SFs they provide access and to install 
"forwarding state" for SFPs, then we also have 
draft-ietf-bess-nsh-bgp-control-plane.

That draft was originally written with RFC 8300 in mind, but with the addition 
of one sub-TLV to indicate the encoding, it also supports 
draft-farrel-mpls-sfc. That should not be a surprise as draft-farrel-mpls-sfc 
attempts to model RFC 8300 as much as possible.

And that brings us back to "Where do we end up, what transition tools should we 
have, and how many steps to transition are there?"

draft-farrel-mpls-sfc provides another transition tool on the migration to RFC 
8300. It allows SFFs to be built as a minor mod to existing routers before 
there is forwarding plane support for the NSH.

But I want to reiterate that the discussion of wat encoding the SF supports is 
a red herring (certainly in the context of RFC 7665). An SF is either 
"SFC-aware" or not [RFC 7665 fig. 3], that is, it either can support the SFC 
encoding (such as NSH) or it can't. But also, an SF is either locally attached 
to the SFF or not. A local attachment is (of course) easier to operate and 
allows "bump in the wire" proxies very easily. A remote attached SF is (IMHO) 
attached via a tunnel.

The question of "remotely attached SFs" is one that should concern all 
implementations of RFC 7665 because no one (as yet) has worked on a protocol to 
bind SFs to SFFs. Robert is right that providing bump in the wire proxy for 
remotely attached SFs means that it is hard to know/control what goes where. 
But that problem exists to so

Re: [bess] [sfc] [mpls] Progress with draft-farrel-mpls-sfc

[mohamed.boucadair]

Re-,

This is really a matter of taste.

Jim, whatever scheme we use for identifying service chains, there are 
requirements/constraints/new practices/new OAM procedures that need to be 
supported/honored for service chaining purposes.

Those are not simple nor complex in MPLS vs. NSH over MPLS. I’m wrong?

Cheers,
Med

De : UTTARO, JAMES [mailto:ju1...@att.com]
Envoyé : lundi 19 mars 2018 14:10
À : BOUCADAIR Mohamed IMT/OLN; Henderickx, Wim (Nokia - BE/Antwerp); Robert 
Raszuk; Adrian Farrel
Cc : mpls; SPRING WG List; s...@ietf.org; bess@ietf.org
Objet : RE: [sfc] [mpls] Progress with draft-farrel-mpls-sfc

Med,

When I say simply, I am speaking as an operator. The 
operations, systems, tools, institutional knowledge etc… in this space is 
around MPLS. There is a simpler path to creating simple chains by using MPLS 
instead of introducing a new encap.

Thanks,
Jim Uttaro

From: mohamed.boucad...@orange.com<mailto:mohamed.boucad...@orange.com> 
[mailto:mohamed.boucad...@orange.com]
Sent: Monday, March 19, 2018 10:03 AM
To: UTTARO, JAMES <ju1...@att.com<mailto:ju1...@att.com>>; Henderickx, Wim 
(Nokia - BE/Antwerp) 
<wim.henderi...@nokia.com<mailto:wim.henderi...@nokia.com>>; Robert Raszuk 
<rob...@raszuk.net<mailto:rob...@raszuk.net>>; Adrian Farrel 
<adr...@olddog.co.uk<mailto:adr...@olddog.co.uk>>
Cc: mpls <m...@ietf.org<mailto:m...@ietf.org>>; SPRING WG List 
<spr...@ietf.org<mailto:spr...@ietf.org>>; s...@ietf.org<mailto:s...@ietf.org>; 
bess@ietf.org<mailto:bess@ietf.org>
Subject: RE: [sfc] [mpls] Progress with draft-farrel-mpls-sfc

Re-,

I’m afraid that you cannot ‘simply’ re-use MPLS for chaining purposes without 
any code upgrade.


NSH does provide the simple functionality you need; that is the information to 
identify a chain + avoid infinite loops. This is known as: MD Type 0x2 with 
length is 0x2.

Of course you can encode that information using another channel, but still code 
change is needed.

Please note that NSH is not at the same level as “GENEVE, VXLAN”.

Cheers,
Med

De : UTTARO, JAMES [mailto:ju1...@att.com]
Envoyé : lundi 19 mars 2018 13:48
À : BOUCADAIR Mohamed IMT/OLN; Henderickx, Wim (Nokia - BE/Antwerp); Robert 
Raszuk; Adrian Farrel
Cc : mpls; SPRING WG List; s...@ietf.org<mailto:s...@ietf.org>; 
bess@ietf.org<mailto:bess@ietf.org>
Objet : RE: [sfc] [mpls] Progress with draft-farrel-mpls-sfc

Med,

We run an MPLS network so there is no NSH deployed anywhere. I 
want to create simple chains that we can make available to our WAN customers 
and I want to keep it simple from a technology and operations POV.. At this 
point I do not see the need to introduce NSH for what we need to do. I can 
simply re-use MPLS.

Not sure why NSH is the winner here there are folks who advocate for GENEVE, 
NSH, VXLAN etc… If IETF is pushing for one encap than it would be helpful to 
define the set of requirements/criteria and compare the encaps.

Thanks,
Jim Uttaro

From: mohamed.boucad...@orange.com<mailto:mohamed.boucad...@orange.com> 
[mailto:mohamed.boucad...@orange.com]
Sent: Monday, March 19, 2018 9:22 AM
To: UTTARO, JAMES <ju1...@att.com<mailto:ju1...@att.com>>; Henderickx, Wim 
(Nokia - BE/Antwerp) 
<wim.henderi...@nokia.com<mailto:wim.henderi...@nokia.com>>; Robert Raszuk 
<rob...@raszuk.net<mailto:rob...@raszuk.net>>; Adrian Farrel 
<adr...@olddog.co.uk<mailto:adr...@olddog.co.uk>>
Cc: mpls <m...@ietf.org<mailto:m...@ietf.org>>; SPRING WG List 
<spr...@ietf.org<mailto:spr...@ietf.org>>; s...@ietf.org<mailto:s...@ietf.org>; 
bess@ietf.org<mailto:bess@ietf.org>
Subject: RE: [sfc] [mpls] Progress with draft-farrel-mpls-sfc

Hi Jim,

Perhaps I missed your point, but I’m not asking to disallow whatever transport 
encapsulation scheme deployed in a network for SFC purposes.

What I’m saying is:
* the IETF has defined a generic SFC architecture and went with a 
transport-agnostic approach that can be deployed in conjunction with one’s 
favorite transport encapsulation protocol.
* Having a transport-agnostic approach get us away from redundant solutions to 
solve the same problem, redundant codes, etc.
* If we allow to mimic NSH in MPLS, there is no reason to do this for MPLS only.
* Instead of mimic NSH, I would personally favor re-using NSH.

Cheers,
Med

De : UTTARO, JAMES [mailto:ju1...@att.com]
Envoyé : lundi 19 mars 2018 12:33
À : BOUCADAIR Mohamed IMT/OLN; Henderickx, Wim (Nokia - BE/Antwerp); Robert 
Raszuk; Adrian Farrel
Cc : mpls; SPRING WG List; s...@ietf.org<mailto:s...@ietf.org>; 
bess@ietf.org<mailto:bess@ietf.org>
Objet : RE: [sfc] [mpls] Progress with draft-farrel-mpls-sfc

Where I get a little lost in this discussion is assuming that there must be one 
encap for SFC chains.. IMO SFC should define encap agnostic

Re: [bess] [sfc] [mpls] Progress with draft-farrel-mpls-sfc

[UTTARO, JAMES]

Med,

When I say simply, I am speaking as an operator. The 
operations, systems, tools, institutional knowledge etc… in this space is 
around MPLS. There is a simpler path to creating simple chains by using MPLS 
instead of introducing a new encap.

Thanks,
Jim Uttaro

From: mohamed.boucad...@orange.com [mailto:mohamed.boucad...@orange.com]
Sent: Monday, March 19, 2018 10:03 AM
To: UTTARO, JAMES <ju1...@att.com>; Henderickx, Wim (Nokia - BE/Antwerp) 
<wim.henderi...@nokia.com>; Robert Raszuk <rob...@raszuk.net>; Adrian Farrel 
<adr...@olddog.co.uk>
Cc: mpls <m...@ietf.org>; SPRING WG List <spr...@ietf.org>; s...@ietf.org; 
bess@ietf.org
Subject: RE: [sfc] [mpls] Progress with draft-farrel-mpls-sfc

Re-,

I’m afraid that you cannot ‘simply’ re-use MPLS for chaining purposes without 
any code upgrade.


NSH does provide the simple functionality you need; that is the information to 
identify a chain + avoid infinite loops. This is known as: MD Type 0x2 with 
length is 0x2.

Of course you can encode that information using another channel, but still code 
change is needed.

Please note that NSH is not at the same level as “GENEVE, VXLAN”.

Cheers,
Med

De : UTTARO, JAMES [mailto:ju1...@att.com]
Envoyé : lundi 19 mars 2018 13:48
À : BOUCADAIR Mohamed IMT/OLN; Henderickx, Wim (Nokia - BE/Antwerp); Robert 
Raszuk; Adrian Farrel
Cc : mpls; SPRING WG List; s...@ietf.org<mailto:s...@ietf.org>; 
bess@ietf.org<mailto:bess@ietf.org>
Objet : RE: [sfc] [mpls] Progress with draft-farrel-mpls-sfc

Med,

We run an MPLS network so there is no NSH deployed anywhere. I 
want to create simple chains that we can make available to our WAN customers 
and I want to keep it simple from a technology and operations POV.. At this 
point I do not see the need to introduce NSH for what we need to do. I can 
simply re-use MPLS.

Not sure why NSH is the winner here there are folks who advocate for GENEVE, 
NSH, VXLAN etc… If IETF is pushing for one encap than it would be helpful to 
define the set of requirements/criteria and compare the encaps.

Thanks,
Jim Uttaro

From: mohamed.boucad...@orange.com<mailto:mohamed.boucad...@orange.com> 
[mailto:mohamed.boucad...@orange.com]
Sent: Monday, March 19, 2018 9:22 AM
To: UTTARO, JAMES <ju1...@att.com<mailto:ju1...@att.com>>; Henderickx, Wim 
(Nokia - BE/Antwerp) 
<wim.henderi...@nokia.com<mailto:wim.henderi...@nokia.com>>; Robert Raszuk 
<rob...@raszuk.net<mailto:rob...@raszuk.net>>; Adrian Farrel 
<adr...@olddog.co.uk<mailto:adr...@olddog.co.uk>>
Cc: mpls <m...@ietf.org<mailto:m...@ietf.org>>; SPRING WG List 
<spr...@ietf.org<mailto:spr...@ietf.org>>; s...@ietf.org<mailto:s...@ietf.org>; 
bess@ietf.org<mailto:bess@ietf.org>
Subject: RE: [sfc] [mpls] Progress with draft-farrel-mpls-sfc

Hi Jim,

Perhaps I missed your point, but I’m not asking to disallow whatever transport 
encapsulation scheme deployed in a network for SFC purposes.

What I’m saying is:
* the IETF has defined a generic SFC architecture and went with a 
transport-agnostic approach that can be deployed in conjunction with one’s 
favorite transport encapsulation protocol.
* Having a transport-agnostic approach get us away from redundant solutions to 
solve the same problem, redundant codes, etc.
* If we allow to mimic NSH in MPLS, there is no reason to do this for MPLS only.
* Instead of mimic NSH, I would personally favor re-using NSH.

Cheers,
Med

De : UTTARO, JAMES [mailto:ju1...@att.com]
Envoyé : lundi 19 mars 2018 12:33
À : BOUCADAIR Mohamed IMT/OLN; Henderickx, Wim (Nokia - BE/Antwerp); Robert 
Raszuk; Adrian Farrel
Cc : mpls; SPRING WG List; s...@ietf.org<mailto:s...@ietf.org>; 
bess@ietf.org<mailto:bess@ietf.org>
Objet : RE: [sfc] [mpls] Progress with draft-farrel-mpls-sfc

Where I get a little lost in this discussion is assuming that there must be one 
encap for SFC chains.. IMO SFC should define encap agnostic behaviors, NSH is 
an encap that has tons of functionality but if a simple chain is needed why is 
it that an existing encap should be disallowed by the IETF?? I want to simplify 
the network, when I say network it is all of the plumbing to realize a service 
for a customer including, WAN, MAN, DC etc.… From an OpS POV having a single 
encap across an integrated solution is quite attractive.

Thanks,
Jim Uttaro

From: sfc [mailto:sfc-boun...@ietf.org] On Behalf Of 
mohamed.boucad...@orange.com<mailto:mohamed.boucad...@orange.com>
Sent: Monday, March 19, 2018 5:52 AM
To: Henderickx, Wim (Nokia - BE/Antwerp) 
<wim.henderi...@nokia.com<mailto:wim.henderi...@nokia.com>>; Robert Raszuk 
<rob...@raszuk.net<mailto:rob...@raszuk.net>>; Adrian Farrel 
<adr...@olddog.co.uk<mailto:adr...@olddog.co.uk>>
Cc: mpls <m...@ietf.org<mailto:m...@ietf.org>

Re: [bess] [sfc] [mpls] Progress with draft-farrel-mpls-sfc

[mohamed.boucadair]

Re-,

I’m afraid that you cannot ‘simply’ re-use MPLS for chaining purposes without 
any code upgrade.


NSH does provide the simple functionality you need; that is the information to 
identify a chain + avoid infinite loops. This is known as: MD Type 0x2 with 
length is 0x2.

Of course you can encode that information using another channel, but still code 
change is needed.

Please note that NSH is not at the same level as “GENEVE, VXLAN”.

Cheers,
Med

De : UTTARO, JAMES [mailto:ju1...@att.com]
Envoyé : lundi 19 mars 2018 13:48
À : BOUCADAIR Mohamed IMT/OLN; Henderickx, Wim (Nokia - BE/Antwerp); Robert 
Raszuk; Adrian Farrel
Cc : mpls; SPRING WG List; s...@ietf.org; bess@ietf.org
Objet : RE: [sfc] [mpls] Progress with draft-farrel-mpls-sfc

Med,

We run an MPLS network so there is no NSH deployed anywhere. I 
want to create simple chains that we can make available to our WAN customers 
and I want to keep it simple from a technology and operations POV.. At this 
point I do not see the need to introduce NSH for what we need to do. I can 
simply re-use MPLS.

Not sure why NSH is the winner here there are folks who advocate for GENEVE, 
NSH, VXLAN etc… If IETF is pushing for one encap than it would be helpful to 
define the set of requirements/criteria and compare the encaps.

Thanks,
Jim Uttaro

From: mohamed.boucad...@orange.com<mailto:mohamed.boucad...@orange.com> 
[mailto:mohamed.boucad...@orange.com]
Sent: Monday, March 19, 2018 9:22 AM
To: UTTARO, JAMES <ju1...@att.com<mailto:ju1...@att.com>>; Henderickx, Wim 
(Nokia - BE/Antwerp) 
<wim.henderi...@nokia.com<mailto:wim.henderi...@nokia.com>>; Robert Raszuk 
<rob...@raszuk.net<mailto:rob...@raszuk.net>>; Adrian Farrel 
<adr...@olddog.co.uk<mailto:adr...@olddog.co.uk>>
Cc: mpls <m...@ietf.org<mailto:m...@ietf.org>>; SPRING WG List 
<spr...@ietf.org<mailto:spr...@ietf.org>>; s...@ietf.org<mailto:s...@ietf.org>; 
bess@ietf.org<mailto:bess@ietf.org>
Subject: RE: [sfc] [mpls] Progress with draft-farrel-mpls-sfc

Hi Jim,

Perhaps I missed your point, but I’m not asking to disallow whatever transport 
encapsulation scheme deployed in a network for SFC purposes.

What I’m saying is:
* the IETF has defined a generic SFC architecture and went with a 
transport-agnostic approach that can be deployed in conjunction with one’s 
favorite transport encapsulation protocol.
* Having a transport-agnostic approach get us away from redundant solutions to 
solve the same problem, redundant codes, etc.
* If we allow to mimic NSH in MPLS, there is no reason to do this for MPLS only.
* Instead of mimic NSH, I would personally favor re-using NSH.

Cheers,
Med

De : UTTARO, JAMES [mailto:ju1...@att.com]
Envoyé : lundi 19 mars 2018 12:33
À : BOUCADAIR Mohamed IMT/OLN; Henderickx, Wim (Nokia - BE/Antwerp); Robert 
Raszuk; Adrian Farrel
Cc : mpls; SPRING WG List; s...@ietf.org<mailto:s...@ietf.org>; 
bess@ietf.org<mailto:bess@ietf.org>
Objet : RE: [sfc] [mpls] Progress with draft-farrel-mpls-sfc

Where I get a little lost in this discussion is assuming that there must be one 
encap for SFC chains.. IMO SFC should define encap agnostic behaviors, NSH is 
an encap that has tons of functionality but if a simple chain is needed why is 
it that an existing encap should be disallowed by the IETF?? I want to simplify 
the network, when I say network it is all of the plumbing to realize a service 
for a customer including, WAN, MAN, DC etc.… From an OpS POV having a single 
encap across an integrated solution is quite attractive.

Thanks,
Jim Uttaro

From: sfc [mailto:sfc-boun...@ietf.org] On Behalf Of 
mohamed.boucad...@orange.com<mailto:mohamed.boucad...@orange.com>
Sent: Monday, March 19, 2018 5:52 AM
To: Henderickx, Wim (Nokia - BE/Antwerp) 
<wim.henderi...@nokia.com<mailto:wim.henderi...@nokia.com>>; Robert Raszuk 
<rob...@raszuk.net<mailto:rob...@raszuk.net>>; Adrian Farrel 
<adr...@olddog.co.uk<mailto:adr...@olddog.co.uk>>
Cc: mpls <m...@ietf.org<mailto:m...@ietf.org>>; SPRING WG List 
<spr...@ietf.org<mailto:spr...@ietf.org>>; s...@ietf.org<mailto:s...@ietf.org>; 
bess@ietf.org<mailto:bess@ietf.org>
Subject: Re: [sfc] [mpls] Progress with draft-farrel-mpls-sfc

Hi all,

Wim has a point here.


For all these proposals, a new behavior is needed to be followed by SFC-aware 
nodes. What differs is the channel used to signal a chain and to supply 
additional data for SFC purposes.



Leveraging on existing code/capabilities is good for a vendor/implementer, but 
the risk is that a given solution will need to support all/many of these 
flavors. Which is not optimal.



The IETF has already a consensus on a transport-agnostic solution. If we open 
the door for MPLS, we need to open it also for IPv6 EH and so on. Are we OK to 
go that way? If yes, what is the NE

Re: [bess] [sfc] [mpls] Progress with draft-farrel-mpls-sfc

[UTTARO, JAMES]

Med,

We run an MPLS network so there is no NSH deployed anywhere. I 
want to create simple chains that we can make available to our WAN customers 
and I want to keep it simple from a technology and operations POV.. At this 
point I do not see the need to introduce NSH for what we need to do. I can 
simply re-use MPLS.

Not sure why NSH is the winner here there are folks who advocate for GENEVE, 
NSH, VXLAN etc… If IETF is pushing for one encap than it would be helpful to 
define the set of requirements/criteria and compare the encaps.

Thanks,
Jim Uttaro

From: mohamed.boucad...@orange.com [mailto:mohamed.boucad...@orange.com]
Sent: Monday, March 19, 2018 9:22 AM
To: UTTARO, JAMES <ju1...@att.com>; Henderickx, Wim (Nokia - BE/Antwerp) 
<wim.henderi...@nokia.com>; Robert Raszuk <rob...@raszuk.net>; Adrian Farrel 
<adr...@olddog.co.uk>
Cc: mpls <m...@ietf.org>; SPRING WG List <spr...@ietf.org>; s...@ietf.org; 
bess@ietf.org
Subject: RE: [sfc] [mpls] Progress with draft-farrel-mpls-sfc

Hi Jim,

Perhaps I missed your point, but I’m not asking to disallow whatever transport 
encapsulation scheme deployed in a network for SFC purposes.

What I’m saying is:
* the IETF has defined a generic SFC architecture and went with a 
transport-agnostic approach that can be deployed in conjunction with one’s 
favorite transport encapsulation protocol.
* Having a transport-agnostic approach get us away from redundant solutions to 
solve the same problem, redundant codes, etc.
* If we allow to mimic NSH in MPLS, there is no reason to do this for MPLS only.
* Instead of mimic NSH, I would personally favor re-using NSH.

Cheers,
Med

De : UTTARO, JAMES [mailto:ju1...@att.com]
Envoyé : lundi 19 mars 2018 12:33
À : BOUCADAIR Mohamed IMT/OLN; Henderickx, Wim (Nokia - BE/Antwerp); Robert 
Raszuk; Adrian Farrel
Cc : mpls; SPRING WG List; s...@ietf.org<mailto:s...@ietf.org>; 
bess@ietf.org<mailto:bess@ietf.org>
Objet : RE: [sfc] [mpls] Progress with draft-farrel-mpls-sfc

Where I get a little lost in this discussion is assuming that there must be one 
encap for SFC chains.. IMO SFC should define encap agnostic behaviors, NSH is 
an encap that has tons of functionality but if a simple chain is needed why is 
it that an existing encap should be disallowed by the IETF?? I want to simplify 
the network, when I say network it is all of the plumbing to realize a service 
for a customer including, WAN, MAN, DC etc.… From an OpS POV having a single 
encap across an integrated solution is quite attractive.

Thanks,
Jim Uttaro

From: sfc [mailto:sfc-boun...@ietf.org] On Behalf Of 
mohamed.boucad...@orange.com<mailto:mohamed.boucad...@orange.com>
Sent: Monday, March 19, 2018 5:52 AM
To: Henderickx, Wim (Nokia - BE/Antwerp) 
<wim.henderi...@nokia.com<mailto:wim.henderi...@nokia.com>>; Robert Raszuk 
<rob...@raszuk.net<mailto:rob...@raszuk.net>>; Adrian Farrel 
<adr...@olddog.co.uk<mailto:adr...@olddog.co.uk>>
Cc: mpls <m...@ietf.org<mailto:m...@ietf.org>>; SPRING WG List 
<spr...@ietf.org<mailto:spr...@ietf.org>>; s...@ietf.org<mailto:s...@ietf.org>; 
bess@ietf.org<mailto:bess@ietf.org>
Subject: Re: [sfc] [mpls] Progress with draft-farrel-mpls-sfc

Hi all,

Wim has a point here.


For all these proposals, a new behavior is needed to be followed by SFC-aware 
nodes. What differs is the channel used to signal a chain and to supply 
additional data for SFC purposes.



Leveraging on existing code/capabilities is good for a vendor/implementer, but 
the risk is that a given solution will need to support all/many of these 
flavors. Which is not optimal.



The IETF has already a consensus on a transport-agnostic solution. If we open 
the door for MPLS, we need to open it also for IPv6 EH and so on. Are we OK to 
go that way? If yes, what is the NEW problem are we trying to solve?



Cheers,

Med

De : sfc [mailto:sfc-boun...@ietf.org] De la part de Henderickx, Wim (Nokia - 
BE/Antwerp)
Envoyé : dimanche 18 mars 2018 07:26
À : Robert Raszuk; Adrian Farrel
Cc : mpls; SPRING WG List; bess@ietf.org<mailto:bess@ietf.org>; 
s...@ietf.org<mailto:s...@ietf.org>
Objet : Re: [sfc] [mpls] Progress with draft-farrel-mpls-sfc

Indeed, this is exactly my point. If you want an interim solution you want to 
use what we have and draft-ietf-bess-service-chaining-04 is an example of how 
you can use the existing data-plane for service chaining. draft-farrel-mpls-sfc 
requires an implementation change in the data-plane, whether we like it or not 
and an upgrade is required even in brownfield deployments. So, you better go 
directly to the final solution defined in IETF SFC WG. If we standardize 
draft-farrel-mpls-sfc we end up supporting both forever.

From: <rras...@gmail.com<mailto:rras...@gmail.com>> on behalf of Robert Raszuk 
<rob...@raszuk.net<mailto:rob...@rasz

Re: [bess] [sfc] [mpls] Progress with draft-farrel-mpls-sfc

[mohamed.boucadair]

Hi Jim,

Perhaps I missed your point, but I’m not asking to disallow whatever transport 
encapsulation scheme deployed in a network for SFC purposes.

What I’m saying is:
* the IETF has defined a generic SFC architecture and went with a 
transport-agnostic approach that can be deployed in conjunction with one’s 
favorite transport encapsulation protocol.
* Having a transport-agnostic approach get us away from redundant solutions to 
solve the same problem, redundant codes, etc.
* If we allow to mimic NSH in MPLS, there is no reason to do this for MPLS only.
* Instead of mimic NSH, I would personally favor re-using NSH.

Cheers,
Med

De : UTTARO, JAMES [mailto:ju1...@att.com]
Envoyé : lundi 19 mars 2018 12:33
À : BOUCADAIR Mohamed IMT/OLN; Henderickx, Wim (Nokia - BE/Antwerp); Robert 
Raszuk; Adrian Farrel
Cc : mpls; SPRING WG List; s...@ietf.org; bess@ietf.org
Objet : RE: [sfc] [mpls] Progress with draft-farrel-mpls-sfc

Where I get a little lost in this discussion is assuming that there must be one 
encap for SFC chains.. IMO SFC should define encap agnostic behaviors, NSH is 
an encap that has tons of functionality but if a simple chain is needed why is 
it that an existing encap should be disallowed by the IETF?? I want to simplify 
the network, when I say network it is all of the plumbing to realize a service 
for a customer including, WAN, MAN, DC etc.… From an OpS POV having a single 
encap across an integrated solution is quite attractive.

Thanks,
Jim Uttaro

From: sfc [mailto:sfc-boun...@ietf.org] On Behalf Of 
mohamed.boucad...@orange.com<mailto:mohamed.boucad...@orange.com>
Sent: Monday, March 19, 2018 5:52 AM
To: Henderickx, Wim (Nokia - BE/Antwerp) 
<wim.henderi...@nokia.com<mailto:wim.henderi...@nokia.com>>; Robert Raszuk 
<rob...@raszuk.net<mailto:rob...@raszuk.net>>; Adrian Farrel 
<adr...@olddog.co.uk<mailto:adr...@olddog.co.uk>>
Cc: mpls <m...@ietf.org<mailto:m...@ietf.org>>; SPRING WG List 
<spr...@ietf.org<mailto:spr...@ietf.org>>; s...@ietf.org<mailto:s...@ietf.org>; 
bess@ietf.org<mailto:bess@ietf.org>
Subject: Re: [sfc] [mpls] Progress with draft-farrel-mpls-sfc

Hi all,

Wim has a point here.


For all these proposals, a new behavior is needed to be followed by SFC-aware 
nodes. What differs is the channel used to signal a chain and to supply 
additional data for SFC purposes.



Leveraging on existing code/capabilities is good for a vendor/implementer, but 
the risk is that a given solution will need to support all/many of these 
flavors. Which is not optimal.



The IETF has already a consensus on a transport-agnostic solution. If we open 
the door for MPLS, we need to open it also for IPv6 EH and so on. Are we OK to 
go that way? If yes, what is the NEW problem are we trying to solve?



Cheers,

Med

De : sfc [mailto:sfc-boun...@ietf.org] De la part de Henderickx, Wim (Nokia - 
BE/Antwerp)
Envoyé : dimanche 18 mars 2018 07:26
À : Robert Raszuk; Adrian Farrel
Cc : mpls; SPRING WG List; bess@ietf.org<mailto:bess@ietf.org>; 
s...@ietf.org<mailto:s...@ietf.org>
Objet : Re: [sfc] [mpls] Progress with draft-farrel-mpls-sfc

Indeed, this is exactly my point. If you want an interim solution you want to 
use what we have and draft-ietf-bess-service-chaining-04 is an example of how 
you can use the existing data-plane for service chaining. draft-farrel-mpls-sfc 
requires an implementation change in the data-plane, whether we like it or not 
and an upgrade is required even in brownfield deployments. So, you better go 
directly to the final solution defined in IETF SFC WG. If we standardize 
draft-farrel-mpls-sfc we end up supporting both forever.

From: <rras...@gmail.com<mailto:rras...@gmail.com>> on behalf of Robert Raszuk 
<rob...@raszuk.net<mailto:rob...@raszuk.net>>
Date: Saturday, 17 March 2018 at 19:13
To: Adrian Farrel <adr...@olddog.co.uk<mailto:adr...@olddog.co.uk>>
Cc: "Henderickx, Wim (Nokia - BE/Antwerp)" 
<wim.henderi...@nokia.com<mailto:wim.henderi...@nokia.com>>, mpls 
<m...@ietf.org<mailto:m...@ietf.org>>, SPRING WG List 
<spr...@ietf.org<mailto:spr...@ietf.org>>, 
"s...@ietf.org<mailto:s...@ietf.org>" <s...@ietf.org<mailto:s...@ietf.org>>, 
"bess@ietf.org<mailto:bess@ietf.org>" <bess@ietf.org<mailto:bess@ietf.org>>
Subject: Re: [sfc] [mpls] Progress with draft-farrel-mpls-sfc

Hi Adrian,

> That proxy may be a bump in the wire between the SFF and SF

I am not so sure about that ... If this would be just "bump in the wire" you 
would have zero guarantees that all packets which need to go via given function 
will actually hit that bump - so this is far from a reliable network service.

There must be associated control plane component attracting traffic to such 
bump.

That mechanism with basic MPLS (where labels by base

Re: [bess] [sfc] [mpls] Progress with draft-farrel-mpls-sfc

[UTTARO, JAMES]

Where I get a little lost in this discussion is assuming that there must be one 
encap for SFC chains.. IMO SFC should define encap agnostic behaviors, NSH is 
an encap that has tons of functionality but if a simple chain is needed why is 
it that an existing encap should be disallowed by the IETF?? I want to simplify 
the network, when I say network it is all of the plumbing to realize a service 
for a customer including, WAN, MAN, DC etc.… From an OpS POV having a single 
encap across an integrated solution is quite attractive.

Thanks,
Jim Uttaro

From: sfc [mailto:sfc-boun...@ietf.org] On Behalf Of 
mohamed.boucad...@orange.com
Sent: Monday, March 19, 2018 5:52 AM
To: Henderickx, Wim (Nokia - BE/Antwerp) <wim.henderi...@nokia.com>; Robert 
Raszuk <rob...@raszuk.net>; Adrian Farrel <adr...@olddog.co.uk>
Cc: mpls <m...@ietf.org>; SPRING WG List <spr...@ietf.org>; s...@ietf.org; 
bess@ietf.org
Subject: Re: [sfc] [mpls] Progress with draft-farrel-mpls-sfc

Hi all,

Wim has a point here.


For all these proposals, a new behavior is needed to be followed by SFC-aware 
nodes. What differs is the channel used to signal a chain and to supply 
additional data for SFC purposes.



Leveraging on existing code/capabilities is good for a vendor/implementer, but 
the risk is that a given solution will need to support all/many of these 
flavors. Which is not optimal.



The IETF has already a consensus on a transport-agnostic solution. If we open 
the door for MPLS, we need to open it also for IPv6 EH and so on. Are we OK to 
go that way? If yes, what is the NEW problem are we trying to solve?



Cheers,

Med

De : sfc [mailto:sfc-boun...@ietf.org] De la part de Henderickx, Wim (Nokia - 
BE/Antwerp)
Envoyé : dimanche 18 mars 2018 07:26
À : Robert Raszuk; Adrian Farrel
Cc : mpls; SPRING WG List; bess@ietf.org<mailto:bess@ietf.org>; 
s...@ietf.org<mailto:s...@ietf.org>
Objet : Re: [sfc] [mpls] Progress with draft-farrel-mpls-sfc

Indeed, this is exactly my point. If you want an interim solution you want to 
use what we have and draft-ietf-bess-service-chaining-04 is an example of how 
you can use the existing data-plane for service chaining. draft-farrel-mpls-sfc 
requires an implementation change in the data-plane, whether we like it or not 
and an upgrade is required even in brownfield deployments. So, you better go 
directly to the final solution defined in IETF SFC WG. If we standardize 
draft-farrel-mpls-sfc we end up supporting both forever.

From: <rras...@gmail.com<mailto:rras...@gmail.com>> on behalf of Robert Raszuk 
<rob...@raszuk.net<mailto:rob...@raszuk.net>>
Date: Saturday, 17 March 2018 at 19:13
To: Adrian Farrel <adr...@olddog.co.uk<mailto:adr...@olddog.co.uk>>
Cc: "Henderickx, Wim (Nokia - BE/Antwerp)" 
<wim.henderi...@nokia.com<mailto:wim.henderi...@nokia.com>>, mpls 
<m...@ietf.org<mailto:m...@ietf.org>>, SPRING WG List 
<spr...@ietf.org<mailto:spr...@ietf.org>>, 
"s...@ietf.org<mailto:s...@ietf.org>" <s...@ietf.org<mailto:s...@ietf.org>>, 
"bess@ietf.org<mailto:bess@ietf.org>" <bess@ietf.org<mailto:bess@ietf.org>>
Subject: Re: [sfc] [mpls] Progress with draft-farrel-mpls-sfc

Hi Adrian,

> That proxy may be a bump in the wire between the SFF and SF

I am not so sure about that ... If this would be just "bump in the wire" you 
would have zero guarantees that all packets which need to go via given function 
will actually hit that bump - so this is far from a reliable network service.

There must be associated control plane component attracting traffic to such 
bump.

That mechanism with basic MPLS (where labels by based MPLS architecture are of 
local significance) is available with L3VPN extensions as already progressing 
in BESS (draft-ietf-bess-service-chaining-04) so why not use this for as you 
state "interim" ?

No one really addressed that question yet and I think it is a critical one to 
make any further judgement  as to the future of this individual submission.

Cheers,
R.



On Sat, Mar 17, 2018 at 6:46 PM, Adrian Farrel 
<adr...@olddog.co.uk<mailto:adr...@olddog.co.uk>> wrote:
Hi Wim,

Thanks for reading the draft so carefully.

> Adrian, on replacement of NSH. You will have to change the SF with this 
> proposal
> in Non proxy case so this proposal does not solve a brownfield case. Which 
> SF(s)
> support MPLS?

This is not about "replacing" the NSH. As you'll see from point 2, below, this 
is about providing an interim / migration technology.

Clearly (and I think you agree) in the case where an SF is not SFC-aware, a 
proxy must be used. That proxy may be a bump in the wire between the SFF and 
SF, a module of the SFF, or a module of the SF. In the case of PNFs, only the 
first two options are available. In the case of a VNF, all three

Re: [bess] [sfc] [mpls] Progress with draft-farrel-mpls-sfc

[mohamed.boucadair]

Hi all,

Wim has a point here.


For all these proposals, a new behavior is needed to be followed by SFC-aware 
nodes. What differs is the channel used to signal a chain and to supply 
additional data for SFC purposes.



Leveraging on existing code/capabilities is good for a vendor/implementer, but 
the risk is that a given solution will need to support all/many of these 
flavors. Which is not optimal.



The IETF has already a consensus on a transport-agnostic solution. If we open 
the door for MPLS, we need to open it also for IPv6 EH and so on. Are we OK to 
go that way? If yes, what is the NEW problem are we trying to solve?



Cheers,

Med

De : sfc [mailto:sfc-boun...@ietf.org] De la part de Henderickx, Wim (Nokia - 
BE/Antwerp)
Envoyé : dimanche 18 mars 2018 07:26
À : Robert Raszuk; Adrian Farrel
Cc : mpls; SPRING WG List; bess@ietf.org; s...@ietf.org
Objet : Re: [sfc] [mpls] Progress with draft-farrel-mpls-sfc

Indeed, this is exactly my point. If you want an interim solution you want to 
use what we have and draft-ietf-bess-service-chaining-04 is an example of how 
you can use the existing data-plane for service chaining. draft-farrel-mpls-sfc 
requires an implementation change in the data-plane, whether we like it or not 
and an upgrade is required even in brownfield deployments. So, you better go 
directly to the final solution defined in IETF SFC WG. If we standardize 
draft-farrel-mpls-sfc we end up supporting both forever.

From: > on behalf of Robert Raszuk 
>
Date: Saturday, 17 March 2018 at 19:13
To: Adrian Farrel >
Cc: "Henderickx, Wim (Nokia - BE/Antwerp)" 
>, mpls 
>, SPRING WG List 
>, 
"s...@ietf.org" >, 
"bess@ietf.org" >
Subject: Re: [sfc] [mpls] Progress with draft-farrel-mpls-sfc

Hi Adrian,

> That proxy may be a bump in the wire between the SFF and SF

I am not so sure about that ... If this would be just "bump in the wire" you 
would have zero guarantees that all packets which need to go via given function 
will actually hit that bump - so this is far from a reliable network service.

There must be associated control plane component attracting traffic to such 
bump.

That mechanism with basic MPLS (where labels by based MPLS architecture are of 
local significance) is available with L3VPN extensions as already progressing 
in BESS (draft-ietf-bess-service-chaining-04) so why not use this for as you 
state "interim" ?

No one really addressed that question yet and I think it is a critical one to 
make any further judgement  as to the future of this individual submission.

Cheers,
R.



On Sat, Mar 17, 2018 at 6:46 PM, Adrian Farrel 
> wrote:
Hi Wim,

Thanks for reading the draft so carefully.

> Adrian, on replacement of NSH. You will have to change the SF with this 
> proposal
> in Non proxy case so this proposal does not solve a brownfield case. Which 
> SF(s)
> support MPLS?

This is not about "replacing" the NSH. As you'll see from point 2, below, this 
is about providing an interim / migration technology.

Clearly (and I think you agree) in the case where an SF is not SFC-aware, a 
proxy must be used. That proxy may be a bump in the wire between the SFF and 
SF, a module of the SFF, or a module of the SF. In the case of PNFs, only the 
first two options are available. In the case of a VNF, all three options exist.

Now, let us recall where we are starting from. There are PNFs and there are 
VNFs built to look like PNFs. These SFs do not support MPLS or NSH.

Similarly, there are routers that do not support the NSH.

Now, of course, we would all love to sell major upgrades so that every 
component of the network is SFC-aware. But we would also like to start 
deploying SFC into existing network infrastructure.

So your question misses the point. The question to ask is which brownfield 
routers and SFs support NSH?

Cheers,
Adrian
___
BESS mailing list
BESS@ietf.org
https://www.ietf.org/mailman/listinfo/bess

Re: [bess] [sfc] [mpls] Progress with draft-farrel-mpls-sfc

[Robert Raszuk]

Adrian,

> draft-farrel-mpls-sfc provides another transition tool on the migration
to RFC 8300.

Very honestly to me it looks like a road block to faster adoption of NSH
not as help to migrate to RFC 8300 in any way.

> It allows SFFs to be built as a minor mod to existing routers before
there is forwarding plane support for the NSH.

I don't agree with that. "MPLS support" in today's equipment is basic three
label operations IMPOSE, POP & SWAP. I don't see how hardware which can
support those three mechanisms can effectively play any role in what one
would expect from NSH alternative.

And as I mentioned before there is no such a thing like MPLS *only*
networks. One company tried to build MPLS only (without IP forwarding)
router but they dropped that plan. So basic IP can carry NSH front ended
packet to whichever device can process NSH headers.

Adding equivalent processing of now both NSH and MPLS headers (far from
basic pop,swap, impose) operations does not seems like a useful standards
and technology investment at this point.

Best,
Robert.







On Sun, Mar 18, 2018 at 11:10 AM, Adrian Farrel <adr...@olddog.co.uk> wrote:

> Wim and Robert,
>
>
>
> [Dropping SPRING at this point as (as previously discussed) we have taken
> / are taking SR out of this document]
>
>
>
> I think that draft-ietf-bess-service-chaining is really important work:
> it expresses a technique that is implemented and shipping.
>
>
>
> On the other hand, this approach is not fully consistent with RFC 7665.
>
>
>
> But it does describe an actual SFC technology. Whether it remains in the
> field or is a migration technology only time (and operators) will tell.
>
>
>
> Now, if we want to support RFC 7665 and RFC 8300 and use a control plane
> to discover the SFFs and to which SFs they provide access and to install
> "forwarding state" for SFPs, then we also have draft-ietf-bess-nsh-bgp-
> control-plane.
>
>
>
> That draft was originally written with RFC 8300 in mind, but with the
> addition of one sub-TLV to indicate the encoding, it also supports draft-
> farrel-mpls-sfc. That should not be a surprise as draft-farrel-mpls-sfc
> attempts to model RFC 8300 as much as possible.
>
>
>
> And that brings us back to "Where do we end up, what transition tools
> should we have, and how many steps to transition are there?"
>
>
>
> draft-farrel-mpls-sfc provides another transition tool on the migration
> to RFC 8300. It allows SFFs to be built as a minor mod to existing routers
> before there is forwarding plane support for the NSH.
>
>
>
> But I want to reiterate that the discussion of wat encoding the SF
> supports is a red herring (certainly in the context of RFC 7665). An SF is
> either "SFC-aware" or not [RFC 7665 fig. 3], that is, it either can support
> the SFC encoding (such as NSH) or it can't. But also, an SF is either
> locally attached to the SFF or not. A local attachment is (of course)
> easier to operate and allows "bump in the wire" proxies very easily. A
> remote attached SF is (IMHO) attached via a tunnel.
>
>
>
> The question of "remotely attached SFs" is one that should concern all
> implementations of RFC 7665 because no one (as yet) has worked on a
> protocol to bind SFs to SFFs. Robert is right that providing bump in the
> wire proxy for remotely attached SFs means that it is hard to know/control
> what goes where. But that problem exists to some extent for any remotely
> attached SF. For that reason (among others) I would implement the proxy as
> part of the SFF.
>
>
>
> Cheers,
>
> Adrian
>
>
>
> *From:* Henderickx, Wim (Nokia - BE/Antwerp) [mailto:wim.henderickx@nokia.
> com]
> *Sent:* 18 March 2018 07:26
> *To:* Robert Raszuk; Adrian Farrel
> *Cc:* mpls; SPRING WG List; s...@ietf.org; bess@ietf.org
>
> *Subject:* Re: [sfc] [mpls] Progress with draft-farrel-mpls-sfc
>
>
>
> Indeed, this is exactly my point. If you want an interim solution you want
> to use what we have and draft-ietf-bess-service-chaining-04 is an example
> of how you can use the existing data-plane for service chaining.
> draft-farrel-mpls-sfc requires an implementation change in the data-plane,
> whether we like it or not and an upgrade is required even in brownfield
> deployments. So, you better go directly to the final solution defined in
> IETF SFC WG. If we standardize draft-farrel-mpls-sfc we end up supporting
> both forever.
>
>
>
> *From: *<rras...@gmail.com> on behalf of Robert Raszuk <rob...@raszuk.net>
> *Date: *Saturday, 17 March 2018 at 19:13
> *To: *Adrian Farrel <adr...@olddog.co.uk>
> *Cc: *"Henderickx, Wim (Nokia - BE/Antwerp)" <wim.he

Re: [bess] [sfc] [mpls] Progress with draft-farrel-mpls-sfc

[Adrian Farrel]

Wim and Robert,
 
[Dropping SPRING at this point as (as previously discussed) we have taken / are 
taking SR out of this document]
 
I think that draft-ietf-bess-service-chaining is really important work: it 
expresses a technique that is implemented and shipping.
 
On the other hand, this approach is not fully consistent with RFC 7665. 
 
But it does describe an actual SFC technology. Whether it remains in the field 
or is a migration technology only time (and operators) will tell.
 
Now, if we want to support RFC 7665 and RFC 8300 and use a control plane to 
discover the SFFs and to which SFs they provide access and to install 
"forwarding state" for SFPs, then we also have 
draft-ietf-bess-nsh-bgp-control-plane.
 
That draft was originally written with RFC 8300 in mind, but with the addition 
of one sub-TLV to indicate the encoding, it also supports 
draft-farrel-mpls-sfc. That should not be a surprise as draft-farrel-mpls-sfc 
attempts to model RFC 8300 as much as possible.
 
And that brings us back to "Where do we end up, what transition tools should we 
have, and how many steps to transition are there?"
 
draft-farrel-mpls-sfc provides another transition tool on the migration to RFC 
8300. It allows SFFs to be built as a minor mod to existing routers before 
there is forwarding plane support for the NSH.
 
But I want to reiterate that the discussion of wat encoding the SF supports is 
a red herring (certainly in the context of RFC 7665). An SF is either 
"SFC-aware" or not [RFC 7665 fig. 3], that is, it either can support the SFC 
encoding (such as NSH) or it can't. But also, an SF is either locally attached 
to the SFF or not. A local attachment is (of course) easier to operate and 
allows "bump in the wire" proxies very easily. A remote attached SF is (IMHO) 
attached via a tunnel.
 
The question of "remotely attached SFs" is one that should concern all 
implementations of RFC 7665 because no one (as yet) has worked on a protocol to 
bind SFs to SFFs. Robert is right that providing bump in the wire proxy for 
remotely attached SFs means that it is hard to know/control what goes where. 
But that problem exists to some extent for any remotely attached SF. For that 
reason (among others) I would implement the proxy as part of the SFF.
 
Cheers,
Adrian
 
From: Henderickx, Wim (Nokia - BE/Antwerp) [mailto:wim.henderi...@nokia.com] 
Sent: 18 March 2018 07:26
To: Robert Raszuk; Adrian Farrel
Cc: mpls; SPRING WG List; s...@ietf.org; bess@ietf.org
Subject: Re: [sfc] [mpls] Progress with draft-farrel-mpls-sfc
 
Indeed, this is exactly my point. If you want an interim solution you want to 
use what we have and draft-ietf-bess-service-chaining-04 is an example of how 
you can use the existing data-plane for service chaining. draft-farrel-mpls-sfc 
requires an implementation change in the data-plane, whether we like it or not 
and an upgrade is required even in brownfield deployments. So, you better go 
directly to the final solution defined in IETF SFC WG. If we standardize 
draft-farrel-mpls-sfc we end up supporting both forever.
 
From: <rras...@gmail.com> on behalf of Robert Raszuk <rob...@raszuk.net>
Date: Saturday, 17 March 2018 at 19:13
To: Adrian Farrel <adr...@olddog.co.uk>
Cc: "Henderickx, Wim (Nokia - BE/Antwerp)" <wim.henderi...@nokia.com>, mpls 
<m...@ietf.org>, SPRING WG List <spr...@ietf.org>, "s...@ietf.org" 
<s...@ietf.org>, "bess@ietf.org" <bess@ietf.org>
Subject: Re: [sfc] [mpls] Progress with draft-farrel-mpls-sfc
 
Hi Adrian,
 
> That proxy may be a bump in the wire between the SFF and SF
 
I am not so sure about that ... If this would be just "bump in the wire" you 
would have zero guarantees that all packets which need to go via given function 
will actually hit that bump - so this is far from a reliable network service. 
 
There must be associated control plane component attracting traffic to such 
bump. 
 
That mechanism with basic MPLS (where labels by based MPLS architecture are of 
local significance) is available with L3VPN extensions as already progressing 
in BESS (draft-ietf-bess-service-chaining-04) so why not use this for as you 
state "interim" ? 
 
No one really addressed that question yet and I think it is a critical one to 
make any further judgement  as to the future of this individual submission. 
 
Cheers,
R. 
 
 
 
On Sat, Mar 17, 2018 at 6:46 PM, Adrian Farrel < <mailto:adr...@olddog.co.uk> 
adr...@olddog.co.uk> wrote:
Hi Wim,

Thanks for reading the draft so carefully.

> Adrian, on replacement of NSH. You will have to change the SF with this 
> proposal
> in Non proxy case so this proposal does not solve a brownfield case. Which 
> SF(s)
> support MPLS?

This is not about "replacing" the NSH. As you'll see from point 2, below, this 
is about providing an interim / migration tec

Re: [bess] [sfc] [mpls] Progress with draft-farrel-mpls-sfc

[Henderickx, Wim (Nokia - BE/Antwerp)]

Indeed, this is exactly my point. If you want an interim solution you want to 
use what we have and draft-ietf-bess-service-chaining-04 is an example of how 
you can use the existing data-plane for service chaining. draft-farrel-mpls-sfc 
requires an implementation change in the data-plane, whether we like it or not 
and an upgrade is required even in brownfield deployments. So, you better go 
directly to the final solution defined in IETF SFC WG. If we standardize 
draft-farrel-mpls-sfc we end up supporting both forever.

From: <rras...@gmail.com> on behalf of Robert Raszuk <rob...@raszuk.net>
Date: Saturday, 17 March 2018 at 19:13
To: Adrian Farrel <adr...@olddog.co.uk>
Cc: "Henderickx, Wim (Nokia - BE/Antwerp)" <wim.henderi...@nokia.com>, mpls 
<m...@ietf.org>, SPRING WG List <spr...@ietf.org>, "s...@ietf.org" 
<s...@ietf.org>, "bess@ietf.org" <bess@ietf.org>
Subject: Re: [sfc] [mpls] Progress with draft-farrel-mpls-sfc

Hi Adrian,

> That proxy may be a bump in the wire between the SFF and SF


I am not so sure about that ... If this would be just "bump in the wire" you 
would have zero guarantees that all packets which need to go via given function 
will actually hit that bump - so this is far from a reliable network service.

There must be associated control plane component attracting traffic to such 
bump.

That mechanism with basic MPLS (where labels by based MPLS architecture are of 
local significance) is available with L3VPN extensions as already progressing 
in BESS (draft-ietf-bess-service-chaining-04) so why not use this for as you 
state "interim" ?

No one really addressed that question yet and I think it is a critical one to 
make any further judgement  as to the future of this individual submission.

Cheers,
R.



On Sat, Mar 17, 2018 at 6:46 PM, Adrian Farrel 
<adr...@olddog.co.uk<mailto:adr...@olddog.co.uk>> wrote:
Hi Wim,

Thanks for reading the draft so carefully.

> Adrian, on replacement of NSH. You will have to change the SF with this 
> proposal
> in Non proxy case so this proposal does not solve a brownfield case. Which 
> SF(s)
> support MPLS?

This is not about "replacing" the NSH. As you'll see from point 2, below, this 
is about providing an interim / migration technology.

Clearly (and I think you agree) in the case where an SF is not SFC-aware, a 
proxy must be used. That proxy may be a bump in the wire between the SFF and 
SF, a module of the SFF, or a module of the SF. In the case of PNFs, only the 
first two options are available. In the case of a VNF, all three options exist.

Now, let us recall where we are starting from. There are PNFs and there are 
VNFs built to look like PNFs. These SFs do not support MPLS or NSH.

Similarly, there are routers that do not support the NSH.

Now, of course, we would all love to sell major upgrades so that every 
component of the network is SFC-aware. But we would also like to start 
deploying SFC into existing network infrastructure.

So your question misses the point. The question to ask is which brownfield 
routers and SFs support NSH?

Cheers,
Adrian
___
BESS mailing list
BESS@ietf.org
https://www.ietf.org/mailman/listinfo/bess

Re: [bess] [sfc] [mpls] Progress with draft-farrel-mpls-sfc

[Robert Raszuk]

Hi Adrian,

> That proxy may be a bump in the wire between the SFF and SF

I am not so sure about that ... If this would be just "bump in the wire"
you would have zero guarantees that all packets which need to go via given
function will actually hit that bump - so this is far from a reliable
network service.

There must be associated control plane component attracting traffic to such
bump.

That mechanism with basic MPLS (where labels by based MPLS architecture are
of local significance) is available with L3VPN extensions as already
progressing in BESS (draft-ietf-bess-service-chaining-04) so why not use
this for as you state "interim" ?

No one really addressed that question yet and I think it is a critical one
to make any further judgement  as to the future of this individual
submission.

Cheers,
R.



On Sat, Mar 17, 2018 at 6:46 PM, Adrian Farrel  wrote:

> Hi Wim,
>
> Thanks for reading the draft so carefully.
>
> > Adrian, on replacement of NSH. You will have to change the SF with this
> proposal
> > in Non proxy case so this proposal does not solve a brownfield case.
> Which SF(s)
> > support MPLS?
>
> This is not about "replacing" the NSH. As you'll see from point 2, below,
> this is about providing an interim / migration technology.
>
> Clearly (and I think you agree) in the case where an SF is not SFC-aware,
> a proxy must be used. That proxy may be a bump in the wire between the SFF
> and SF, a module of the SFF, or a module of the SF. In the case of PNFs,
> only the first two options are available. In the case of a VNF, all three
> options exist.
>
> Now, let us recall where we are starting from. There are PNFs and there
> are VNFs built to look like PNFs. These SFs do not support MPLS or NSH.
>
> Similarly, there are routers that do not support the NSH.
>
> Now, of course, we would all love to sell major upgrades so that every
> component of the network is SFC-aware. But we would also like to start
> deploying SFC into existing network infrastructure.
>
> So your question misses the point. The question to ask is which brownfield
> routers and SFs support NSH?
>
> Cheers,
> Adrian
___
BESS mailing list
BESS@ietf.org
https://www.ietf.org/mailman/listinfo/bess