Re: [mpls-linux-general] DaveM Linux MPLS stack

[James R. L. <jl...@mi...>]

On Wed, Apr 30, 2003 at 08:47:00PM -0400, Fran=E7ois Desloges wrote:
<snip>
> James, did Dave ever give you hint about what he finds crappy ?
> Dave normally sets very high standard performance-wise. I think you are=
 more=20
> concern with flexibility. I have not seen his code yet but I can tell f=
rom=20
> experience that he could certainly get good hints from you, so that eve=
rybody=20
> wins.=20

We've talked.  Many of his issues revolve around some of code I wrote whi=
le
being lazy and some of the hacks I put in place because of my lack of
understanding of areas of the linux net stack.  He neglected to use the
word 'crappy' in direct conversation with me, but I guess that is 'par fo=
r
the course'.

> I mean, it would be stupid to not involve the guy in charge of MPLS int=
erop=20
> testing at the Networld Interop show for so many years.

Funny you mention that, I just got back from Las Vegas where I was workin=
g in
the Advanced Internetworking Initiative iLab at Networld + Interop.  Bill
Jensen, myself, and the rest of the team were quite successfully.  Most o=
f the
focus was on MPLS services (VPLS, Layer 2 Transports 'ala Martini', and
RFC2547bis) running over resilient technologies (SONET, RSTP, MPLS FRR).
We had hoped to work with some of the 'graceful restart' mechanisms for
the routing protocols, but were unable to convince enough vendors to ship
us code.  This year we started what will hopefully be a long term
relationship with ISOCORE (a MPLS verification and interoperability lab).

Good to hear from you Fran?ois
--=20
James R. Leu