Robert,

Due to historic reason, our product was initially only a strict-router, it was 
later added the functionality to be a loose-router,
But many our customer's deployment is still strict-router, I think it's 
necessary to clarify this inconsistence, 
though I do not see the immediate pain on this issue.

Thanks a lot

Regards,
-Rockson 

-----Original Message-----
From: Robert Sparks [mailto:[EMAIL PROTECTED] 
Sent: Monday, August 18, 2008 11:36 PM
To: Rockson Li (zhengyli)
Cc: [email protected]
Subject: Re: [Sip-implementors] Can proxy forward a req to a local determined 
strict-router?


On Aug 16, 2008, at 10:58 AM, Rockson Li (zhengyli) wrote:

> Hi folks,
>
> RFC3261
> sec 16.6 Request Forwarding
>
> 6. Postprocess routing information
>
>         If the proxy has a local policy that mandates that the request
>         visit one specific proxy, an alternative to pushing a Route
>         value into the Route header field is to bypass the forwarding
>         logic of item 10 below, and instead just send the request to
>         the address, port, and transport for that specific proxy.  If
>         the request has a Route header field, this alternative MUST 
> NOT
>         be used unless it is known that next hop proxy is a loose
>         router.  Otherwise, this approach MAY be used,
>
> so I think this means if the request contains no Router header, this 
> request might be forwarded to a next element - either a strict-router 
> or
> loose-router- directly without
> adding Route header.
>
> However, the next sub section
>
>    7. Determine Next-Hop Address, Port, and Transport
>
>         The proxy MAY have a local policy to send the request to a
>         specific IP address, port, and transport, independent of the
>         values of the Route and Request-URI.  Such a policy MUST NOT 
> be
>         used if the proxy is not certain that the IP address, port, 
> and
>         transport correspond to a server that is a loose router.
>
> It totally forbids this forwarding-to-strict-router, ONLY loose router 
> is allowed.
>
> Why there's such a difference?

Ok - I take what I sent to Iñaki back - I see the inconsistency and will log it 
for clarification.
(Which I suspect will come in the form of removing strict routers from the spec)

To directly answer the why - we knew the strict routing behavior was really 
broken when we were working on 3261 - the creation of loose-routing was one of 
the big bug-fixes that went into that spec. However, _at the time_ there were 
still several strict routers deployed and we worked to leave in some backwards 
compatibility. Allowing a blind forward to a strict router when no Route header 
was present was asked for fairly late in the editing process (if I remember
correctly) and we missed that place to touch when editing it in.

So, if this inconsistency is causing you pain in a real deployment situation 
(as opposed to this being something you just noticed as an inconsistency), I'd 
like to know. (And I strongly encourage you to do whatever it takes to change 
any proxies you have that are still strict routing.)

RjS

>
>
> thanks
>
> Regards,
> -Rockson
> _______________________________________________
> Sip-implementors mailing list
> [email protected]
> https://lists.cs.columbia.edu/cucslists/listinfo/sip-implementors


_______________________________________________
Sip-implementors mailing list
[email protected]
https://lists.cs.columbia.edu/cucslists/listinfo/sip-implementors

Reply via email to