RE: [PWE3] RE: IMPORTANT: Strategy for VPN work in IETF - vpls provisioning.

I would like to reinforce the importance of this provisioning aspect for the
INTRA-AS case.
In my expectation, membership of an L2 VPN / vpls instance would be a
relatively static thing. Although in a large network, there may be
significant churn. Depending on the service scale objectives, mechanized
provisioning support may be required - hopefully something a little better
than a hacked-up CLI can be found in order to minimize lifecycle OPEX for
this kind of service.
Steven Wright


> -----Original Message-----
> From: Juha Heinanen [mailto:[email protected]]
> Sent: Thursday, May 08, 2003 10:24 AM
> To: [email protected]
> Cc: Paul Knight; 'Alex Zinin'; [email protected];
> [email protected];
> 'Wijnen, Bert (Bert)'; 'Thomas Narten'; 'Peterson, Jon'; 'Allison
> Mankin'
> Subject: RE: [PWE3] RE: IMPORTANT: Strategy for VPN work in IETF
>
>
> Thomas D. Nadeau writes:
>
>  >     You raise a good point. Is there going to be any discussion of
>  > actual provisioning of VPNs?
>
> provisioning is indeed an important issue.  i have been told that
> commercial provisioning systems for bgp/mpls vpns are very expensive.
> may be that is because provisioning of them is difficult to do.
>
> a good solution is such for which writing automatic provisioning tools
> is very easy.  this is true for radius based solutions, where the
> provisioning system needs little interaction with the actual pes.
>
> -- juha
>
>