RE: [PWE3] Comments on pwe3 framework document - Maintenance Architecture

[Tim.Frost@Zarlink.Com]

Hi Prayson, Stewart,

I am also interested in any work on the maintenance architecture, 
specifically with respect to the setup and negotiation of pseudo-wires. 
I've looked at Stewart's draft, and it is some help:

"During the set-up process, the PEs need to exchange some information 
(i.e., learn each others' capabilities). The tunneling control protocol 
may be extended to provide mechanisms to enable the PEs to exchange all 
necessary information on behalf of the PW."

However, I'm concerned by the statement "The tunneling control protocol 
may be extended..." I'm not an expert on tunneling protocols, but this 
implies that current tunneling protocols may not afford all facilities 
required to setup and negotiate a pseudo-wire. Can anyone comment on this? 
Does there need to be more work here to allow the construction of 
inter-operable pseudo-wires?

For instance, I am particularly interested in TDM pseudo-wires using Sasha 
Vainshtein's draft. There are a number of service-specific parameters that 
have to be negotiated here, e.g. payload format, timestamp resolution etc. 
What is the process for exchanging information on the capability of the 
two PE devices, and agreeing the service parameters to be used?

Regards,
Tim


Tel.:   +44 1752 693840
Email:  tim.frost@zarlink.com





"Prayson Pate" <prayson.pate@overturenetworks.com>
Sent by: pwe3-admin@ietf.org
17/05/2002 20:52

 
        To:     "Stewart Bryant" <stbryant@cisco.com>, "Danny McPherson" <danny@tcb.net>, 
"Vinay Bannai" <Vinay@luminous.com>, <pwe3@ietf.org>
        cc: 
        Subject:        RE: [PWE3] Comments on pwe3 framework document


Vinay,

> I have some comments on the framework document. 
> 
> - Section 3.3 (Maintenance Architecture) is very skeletal. Is any 
> work being done in this area?

Section 7 of the protocol layering draft (draft-bryant-pwe3-protocol-01)
has some work on maintenance.  Stewart may want to comment on this.

 
> - I would like to see some explanation or clarification on the 
> mailing list for the debug capability of the PW? 

So would I ;-)


> For instance, I have bundled multiple
> services in the same tunnel and one of the service instance is having
> problems. How do we go about tracing the particular service 
> instance through
> the network to figure where things are going bad? Or is it tied to the
> traceability of the tunnel LSP itself? I don't see much mention in the
> document regarding this.

Section 4.7 talks about the need for a traceroute mechanism, but 
there needs to be a more detailed discussion of this somewhere.


> -  Section 3.5.2.1.3 talks about packet fragmentation. Doesn't 
> this violate the charter of the PWE3 working group?

I am not touching this one.  Danny?

Prayson



_______________________________________________
pwe3 mailing list
pwe3@ietf.org
https://www1.ietf.org/mailman/listinfo/pwe3




_______________________________________________
pwe3 mailing list
pwe3@ietf.org
https://www1.ietf.org/mailman/listinfo/pwe3