RE: [Idr] Inter-operable BGP implementations - Call for information

["Susan Hares" <[e-mail]>]

Curtis: 

Thanks for sending out this note.

My motivation was to show that there are
a lot of inter-operable implementations --
but the survey was too long.

I am trying to take an informal survey
(no RFC 2119 binding) to find out of
those who have a BGP4 implementation
that matches draft-ietf-idr-bgp4-23.txt 
(shortly to be -24.txt)

1) Why did you not respond to the survey:
	a) length of survey
	b) other (please comment)
2) Do you interoperate with other BGP 
   implementations?

  If so, which bgp implementations?
 
For those that respond, I will need:
1) contact (email, name) if different than
   the person sending in the response.

2) implementation: product/revision
   and date of revision.

So...anyone else who wants to send
me this information, please send it 
within the next 2 weeks. 

Sue

-----Original Message-----
From: Curtis Villamizar [mailto:[e-mail]]
Sent: Thursday, May 20, 2004 12:49 PM
To: Pekka Savola
Cc: Susan Hares; [e-mail]
Subject: Re: [Idr] Inter-operable BGP implementations - Call for
information 



In message <[e-mail]>, Pekka Savola 
writes:
> On Thu, 20 May 2004, Susan Hares wrote:
> > I am revising the BGP implementation survey, and
> > need an updated on inter-operable implementations.
> > I will accept input for 2 weeks ending 6/2/04.  
> > At that time, I will release version -01 of the BGP
> > implementation survey. 
> 
> FWIW, if you look at RFC2026, it's not sufficient that you have
> interoperable implementations (in general), there must be two
> interoperable implementations of _every feature_ of the spec.
> 
> >From 4.1.2,
>                                                             For the
>    purposes of this section, "interoperable" means to be functionally
>    equivalent or interchangeable components of the system or process in
>    which they are used.
> 
> The interpretation of what this means has been quite blur, but it 
> would be better that if interop tests explicitly tested each feature 
> of the spec, not the spec as a whole.
> 
> -- 
> Pekka Savola                 "You each name yourselves king, yet the
> Netcore Oy                    kingdom bleeds."
> Systems. Networks. Security. -- George R.R. Martin: A Clash of Kings



Field experience has already proven many BGP implementations to be
interoperable by the definition above.  You can pull out a brand X
router and put in a brand Y router it all works fine, for a large set
of X and Y.

The spec has a lot of MUST and SHOULD and the first survey asked for
an answer for each one of them.  A lot of providers just didn't make
it through verifying which of the SHOULDs they needed to check off
before the cutoff for accepting these (which was just a few weeks).

I don't really know Sue's motivation for this latest request.

What might me useful would be to include the responses to the original
survey, add any that come in, and include the interoperability
experience (without full surveys) that Sue has asked for just as
further evidence that there are a lot of interoperable BGP
implementations.

We should be able to update the protocol experiences doc (if we want
to) after the whole set goes through six months or a year from now if
additional vendors come forward with completed surveys.

Curtis


_______________________________________________
Idr mailing list
[e-mail]
https://www1.ietf.org/mailman/listinfo/idr