Re: [pcp] IWF & PCP Port Number (was RE: draft-wing-pcp-base-01 posted)

["Dan Wing" <[e-mail]>]

> -----Original Message-----
> From: [e-mail]
> [mailto:[e-mail]]
> Sent: Tuesday, October 26, 2010 1:57 AM
> To: Dan Wing; [e-mail]
> Subject: IWF & PCP Port Number (was RE: [pcp] draft-wing-pcp-base-01
> posted)
> 
> Re-,
> 
>    I thought the base PCP document focuses on basic functions; but I'm
>    not sure IGD and NAT-PMP IWFs are part of that set.  These functions
>    are to be discussed IMHO in dedicated document(s) whenever required.
>    I know my position may be seen as biased because I'm author of
>    https://datatracker.ietf.org/doc/draft-bpw-pcp-upnp-igd-
> interworking/
>    which includes more information than what is in the pcp-base
>    document.

There is more detail necessary to describe UPnP IGD interworking,
such as how UPnP IGD's add/delete actions are mapped to PCP.  

>    I'm not sure there is a justification for a PCP Server to send a
> NAT-
>    PMP error message instead of returning a PCP Error message with a
>    code error set to Unsupported version only for the backward
>    compatibility with NAT-PMP.  What a NAT-PMP Client will do when it
>    receives that error message, it will stop to send subsequent
> messages
>    to the PCP Server? 

Dunno.  that's why the NAT-PMP Compatibility section (Section 14)
ends with:

     "[Ed.  Note: More analysis is necessary on NAT-PMP backward
      compatibility, including checking if NAT-PMP clients are compliant
      with [I-D.cheshire-nat-pmp]] regarding error handling."

>  Isn't it easier to use a distinct port number
>  for  PCP?

Sure, we can do that, too.  A device that supports NAT-PMP today
can continue listening on the NAT-PMP port, and PCP could get its
own new port from IANA.

-d

>    With a distinct port number for PCP, there is not conflict between
>    NAT-PMP and PCP.
> 
>    A NAT-PMP speaker which is upgraded to be PCP-speaker will
>    know the PCP dedicated port.
> 
>    A NAT-PMP speaker which is not PCP-speaker won't never sent a NAT-
> PMP
>    message to a PCP Server.
> 
>    In conclusion:
> 
>    o  I suggest IWFs are specified in dedicated documents whenever
>       required.
> 
>    o  To avoid any interference with NAT-PMP, a dedicated port number
> to
>       be assigned to PCP.
> 
>    Cheers,
> 
>    Med
> *********************************
> This message and any attachments (the "message") are confidential and
> intended solely for the addressees.
> Any unauthorised use or dissemination is prohibited.
> Messages are susceptible to alteration.
> France Telecom Group shall not be liable for the message if altered,
> changed or falsified.
> If you are not the intended addressee of this message, please cancel it
> immediately and inform the sender.
> ********************************