Re: BGP autodiscovery design team

--===============7020081599764597954==
Content-Type: multipart/alternative; boundary="000000000000afaf790598e703a6"

--000000000000afaf790598e703a6
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

Hi John,

When this was announced I asked if this design team scope is to work on
discovering your BGP peers in DC space or also in WAN and IX environments.

As you know I have been focusing on the latter for a number of years and I
can share that discovering p2p eBGP peers requires a very different
solutions as compared with discovering your iBGP peers or IX members.

So please kindly clarify the scope of this effort here.

Just stating BGP autodiscovery IMHO is too broad at this point. Likewise
working on the DC space and declaring we are done with BGP autodiscovery
all together would not be in the best interest of the protocol.

I suggest that this effort be called BGP DC or p2p autodiscovery.

Many thx,
Robert.


On Wed, Dec 4, 2019 at 8:33 PM John Scudder <jgs=3D
[email protected]> wrote:

> Hi All,
>
> I realized this morning that although we announced in the last meeting
> that we=E2=80=99re forming this design team, we never announced it on the=
 list.
> This is to correct that mistake. Here=E2=80=99s the text of the slide we =
presented
> at the meeting (
> https://datatracker.ietf.org/meeting/106/materials/slides-106-idr-sessa-a=
uto-discovery-design-team-chairs-00
> ):
>
> ---
> Autodiscovery
>
> =E2=80=A2 Clear interest in WG to work on this topic.
> =E2=80=A2 No clear consensus on a specific approach.
> =E2=80=A2 Proposals (four of them) progressively closer semantically, see=
ms likely
> convergence can happen.
> =E2=80=A2 But, important differences remain.
>         =E2=80=A2 At least: transport (L2, L3), liveness, security, maybe=
 multihop.
> =E2=80=A2 Chairs propose to charter a design team to close on a single pr=
oposal,
> by next meeting.
>         =E2=80=A2 Might build on one of the existing drafts, might be a n=
ew draft,
> up to the design team.
>         =E2=80=A2 Emphasis on pragmatism, OK to limit applicability (for =
example,
> to a single administrative domain).
> =E2=80=A2 Questions? Comments? Volunteers?
> =E2=80=94
>
> We are in the process of forming the team and hope to announce it soon. W=
e
> have many good volunteers already but if you didn=E2=80=99t know and do w=
ant to
> volunteer, please contact Sue and me.
>
> Note that the design team will not make any binding decisions: as with al=
l
> WG work, the WG has the final word, so any output of the DT will serve as
> input to the WG as a whole. (So even if you aren=E2=80=99t part of the DT=
, that
> doesn=E2=80=99t mean you won=E2=80=99t have a chance to be part of the pr=
ocess.)
>
> Thanks,
>
> =E2=80=94John
> _______________________________________________
> Idr mailing list
> [email protected]
> https://www.ietf.org/mailman/listinfo/idr
>

--000000000000afaf790598e703a6
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">Hi John,<div><br></div><div>When this was announced=C2=A0I=
 asked if this design team scope is to work on discovering your BGP peers i=
n DC space or also in WAN and IX environments.=C2=A0</div><div><br></div><d=
iv>As you know I have been focusing on the latter for a number of=C2=A0year=
s and I can share that discovering p2p eBGP=C2=A0peers requires a very diff=
erent solutions as compared with discovering your iBGP peers or IX members.=
=C2=A0</div><div><br></div><div>So please kindly clarify the scope of this =
effort here.=C2=A0</div><div><br></div><div>Just stating BGP autodiscovery =
IMHO is too broad at this=C2=A0point. Likewise working on the DC space and =
declaring we are done with BGP autodiscovery all together would not be in t=
he best interest of the protocol.=C2=A0</div><div><br></div><div>I suggest =
that this effort be called BGP DC or p2p autodiscovery.=C2=A0</div><div><br=
></div><div>Many thx,</div><div>Robert.</div><div><br></div></div><br><div =
class=3D"gmail_quote"><div dir=3D"ltr" class=3D"gmail_attr">On Wed, Dec 4, =
2019 at 8:33 PM John Scudder &lt;jgs=3D<a href=3D"mailto:40juniper.net@dmar=
c.ietf.org">[email protected]</a>&gt; wrote:<br></div><blockquot=
e class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-left:1px s=
olid rgb(204,204,204);padding-left:1ex">Hi All,<br>
<br>
I realized this morning that although we announced in the last meeting that=
 we=E2=80=99re forming this design team, we never announced it on the list.=
 This is to correct that mistake. Here=E2=80=99s the text of the slide we p=
resented at the meeting (<a href=3D"https://datatracker.ietf.org/meeting/10=
6/materials/slides-106-idr-sessa-auto-discovery-design-team-chairs-00" rel=
=3D"noreferrer" target=3D"_blank">https://datatracker.ietf.org/meeting/106/=
materials/slides-106-idr-sessa-auto-discovery-design-team-chairs-00</a>):<b=
r>
<br>
---<br>
Autodiscovery<br>
<br>
=E2=80=A2 Clear interest in WG to work on this topic.<br>
=E2=80=A2 No clear consensus on a specific approach.<br>
=E2=80=A2 Proposals (four of them) progressively closer semantically, seems=
 likely convergence can happen. <br>
=E2=80=A2 But, important differences remain.<br>
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =E2=80=A2 At least: transport (L2, L3), livenes=
s, security, maybe multihop.<br>
=E2=80=A2 Chairs propose to charter a design team to close on a single prop=
osal, by next meeting.<br>
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =E2=80=A2 Might build on one of the existing dr=
afts, might be a new draft, up to the design team.<br>
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =E2=80=A2 Emphasis on pragmatism, OK to limit a=
pplicability (for example, to a single administrative domain). <br>
=E2=80=A2 Questions? Comments? Volunteers?<br>
=E2=80=94<br>
<br>
We are in the process of forming the team and hope to announce it soon. We =
have many good volunteers already but if you didn=E2=80=99t know and do wan=
t to volunteer, please contact Sue and me. <br>
<br>
Note that the design team will not make any binding decisions: as with all =
WG work, the WG has the final word, so any output of the DT will serve as i=
nput to the WG as a whole. (So even if you aren=E2=80=99t part of the DT, t=
hat doesn=E2=80=99t mean you won=E2=80=99t have a chance to be part of the =
process.)<br>
<br>
Thanks,<br>
<br>
=E2=80=94John<br>
_______________________________________________<br>
Idr mailing list<br>
<a href=3D"mailto:[email protected]" target=3D"_blank">[email protected]</a><br>
<a href=3D"https://www.ietf.org/mailman/listinfo/idr" rel=3D"noreferrer" ta=
rget=3D"_blank">https://www.ietf.org/mailman/listinfo/idr</a><br>
</blockquote></div>

--000000000000afaf790598e703a6--


--===============7020081599764597954==
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

_______________________________________________
Idr mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/idr

--===============7020081599764597954==--