Bruce,
Good point about the store state in the intermediate peers I think that we
need to do something about it. Having a flag in the forwarding header may be
a good option. 
 
I think that stored state can also happens in SRR if the response fails in
an intermediate peer that gets the response before the one that kept the
state so I think that some timeout on keeping the state is necessary anyhow
for the base draft.

Regards
Roni Even

-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of
Bruce Lowekamp
Sent: Wednesday, November 12, 2008 10:10 PM
To: jiangxingfeng 36340
Cc: [EMAIL PROTECTED]; [email protected]
Subject: Re: [P2PSIP] Fwd:New Version Notification for
draft-jiang-p2psip-relay-00

I think it's great to see this written up.  While I've argued against
incorporating DRR and RPR into the base spec, I think developing it as
an extension is an excellent idea.   A few high-level comments.

I think the right thing to do is to separate out an p2p-based
implementation of nat-behavior-discovery as a separate usage and put
it in its own draft.  I'm more than happy to contribute to such an
effort.  Note that because nat-behavior-discovery is experimental,
such a draft (and anything that depends on it, such as this) will need
to be experimental as well.

I agree with the decision to use a ForwardingOption for signalling the
desired response address.  But a couple details concern me.
- I'm not sure I agree DESTINATION_CRITICAL needs to be set here.  It
seems like if the responder doesn't support this routing mode, it will
just return via SRR like normal.
- The only complication I see is that reload's SRR allows intermediate
peers to keep state about transactions rather than store the state in
the Via list (5.1.2.2).  Not using SRR would cause such a peer to have
a possible state explosion waiting for transactions to time out.
Options I can think of offhand for solving that are: 1) ignore the
problem, 2) make the ForwardingOption FORWARD_CRITICAL, 3) add a new
flag to the forwarding header that suggests intermediate peers not
keep state.

Bruce


On Sun, Oct 26, 2008 at 8:20 PM, jiangxingfeng 36340
<[EMAIL PROTECTED]> wrote:
> Hi, folks:
>
> We've just submitted a draft which proposes an extension to RELOAD to
support direct response and relay peer routing mode. This topic has been
discussed during Dublin meeting and is based on the
draft-jiang-p2psip-sep-01.
>
> Any comments are appreciated.
>
> Regards
>
> Jiang XingFeng
>
>
> ---------- Forwarded message ----------
> From: IETF I-D Submission Tool <[EMAIL PROTECTED]>
> To: [EMAIL PROTECTED]
> Date: Sat, 25 Oct 2008 06:52:12 -0700 (PDT)
> Subject: New Version Notification for draft-jiang-p2psip-relay-00
>
> A new version of I-D, draft-jiang-p2psip-relay-00.txt has been successfuly
submitted by XingFeng Jiang and posted to the IETF repository.
>
> Filename:        draft-jiang-p2psip-relay
> Revision:        00
> Title:           An extension to RELOAD to support Direct Response and
Relay Peer routing
> Creation_date:   2008-10-25
> WG ID:           Independent Submission
> Number_of_pages: 19
>
> Abstract:
> This document proposes an extension to RELOAD to support direct
> response and relay peer routing modes.  The document starts with the
> problem statement and address concerns about these two routing modes
> mentioned in RELOAD.  Then methods about how to discover NAT behavior
> of the client in P2PSIP situations are discussed.  The last part
> proposes extensions to RELOAD for supporting these additional routing
> modes.
>
>
>
> The IETF Secretariat.
>
>
>
> _______________________________________________
> P2PSIP mailing list
> [email protected]
> https://www.ietf.org/mailman/listinfo/p2psip
>
>
_______________________________________________
P2PSIP mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/p2psip

_______________________________________________
P2PSIP mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/p2psip

Reply via email to