MIME-Version: 1.0
From: "Andrew G. Malis" <agmalis@gmail.com>
Date: Wed, 23 May 2018 15:52:13 -0400
Message-ID: <CAA=duU1o2LRsrphdRXh3re2ZMkusyD8rQQKmUS5_E527uajsqQ@mail.gmail.com>
To: draft-ietf-mpls-sfc@ietf.org
Cc: mpls@ietf.org, sfc@ietf.org
Content-Type: multipart/alternative; boundary="00000000000041e558056ce4e005"
Archived-At: <https://mailarchive.ietf.org/arch/msg/mpls/y4YMMFKC72nLNJI9UMaOf-Wmua8>
Subject: [mpls] Comments on draft-ietf-mpls-sfc-01
Precedence: list

--00000000000041e558056ce4e005
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

Adrian et al,

I just had a chance to review draft-ietf-mpls-sfc-01 in detail, and have a
few comments.

First of all, thanks for including the section 4.5 text.

That said, I=E2=80=99m having a hard time distinguishing between using MPLS=
-SR for
SFC and the =E2=80=9Cmicro chains=E2=80=9D discussed in section 4.4. I agre=
e that the
=E2=80=9Cmicro chains" are a logical extreme of section 4.2, but that said,=
 section
4.2 does include swapping as well as popping, while section 4.4 is pure
popping. It would be greatly appreciated if you could add text that
differentiates =E2=80=9Cmicro chains=E2=80=9D from MPLS-SR, otherwise it=E2=
=80=99s just a different
name for the same functionality. If there is actually is no difference,
then section 4.4 should be removed.

I also have a comment in section 12. The text currently says "It should be
noted that the MPLS encoding is slightly less functional than the direct
use of the NSH.=E2=80=9D In this case, =E2=80=9Cslightly=E2=80=9D is a judg=
ement for the reader to
make, and I see a huge loss of functionality because there=E2=80=99s no sup=
port for
per-packet metadata. Please remove the word =E2=80=9Cslightly=E2=80=9D in t=
his text.
Without the word =E2=80=9Cslightly=E2=80=9D, the text changes from an opini=
on to a fact.

Thanks,
Andy

--00000000000041e558056ce4e005
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">Adrian et al,<div><br></div><div>I just had a chance to re=
view draft-ietf-mpls-sfc-01 in detail, and have a few comments.</div><div><=
br></div><div>First of all, thanks for including the section 4.5 text.</div=
><div><br></div><div>That said, I=E2=80=99m having a hard time distinguishi=
ng between using MPLS-SR for SFC and the =E2=80=9Cmicro chains=E2=80=9D dis=
cussed in section 4.4. I agree that the =E2=80=9Cmicro chains&quot; are a l=
ogical extreme of section 4.2, but that said, section 4.2 does include swap=
ping as well as popping, while section 4.4 is pure popping. It would be gre=
atly appreciated if you could add text that differentiates =E2=80=9Cmicro c=
hains=E2=80=9D from MPLS-SR, otherwise it=E2=80=99s just a different name f=
or the same functionality. If there is actually is no difference, then sect=
ion 4.4 should be removed.</div><div><br></div><div>I also have a comment i=
n section 12. The text currently says &quot;It should be noted that the MPL=
S encoding is slightly less functional than the direct use of the NSH.=E2=
=80=9D In this case, =E2=80=9Cslightly=E2=80=9D is a judgement for the read=
er to make, and I see a huge loss of functionality because there=E2=80=99s =
no support for per-packet metadata. Please remove the word =E2=80=9Cslightl=
y=E2=80=9D in this text. Without the word =E2=80=9Cslightly=E2=80=9D, the t=
ext changes from an opinion to a fact.</div><div><br></div><div>Thanks,</di=
v><div>Andy</div><div><br></div></div>

--00000000000041e558056ce4e005--