Re: [pcp] Will PCP do it?

It is not related to source-trace.

We keep our promises with one another - no matter what!

Best Regards,
Tina TSOU
http://tinatsou.weebly.com/contact.html

-----Original Message-----
From: Dan Wing [mailto:dwing@cisco.com] 
Sent: Wednesday, March 09, 2011 12:06 PM
To: 'Tina Tsou'; pcp@ietf.org
Subject: RE: [pcp] Will PCP do it?

> -----Original Message-----
> From: Tina Tsou [mailto:tena@huawei.com]
> Sent: Wednesday, March 09, 2011 11:52 AM
> To: 'Dan Wing'; pcp@ietf.org
> Subject: RE: [pcp] Will PCP do it?
> 
> My understanding about PCP is that it allows the PCP client to instruct
> the
> PCP server to allocate a port number with a transport protocol
> (pinhole) so
> that the PCP client can send the NAT's public IP address, transport
> protocol
> and port number to the application server outside the NAT to allow that
> application server to send information to the device with the PCP
> client. 

Yes, that covers two of the use-cases described in draft-ietf-pcp-base,
" Operating a Server" (Section 7.1) and "Operating a Symmetric
Client/Server" (Section 7.3).

> A PCP client already has the public IP address, transport port protocol
> and
> port number  What I described does not involve the device behind the
> NAT but
> another entity that is interested in knowing which private IP address
> is
> assigned with a specific port number for a transport protocol and NAT's
> public IP address.  Is it in your opinion that PCP can be enhanced to
> support this type of inquiry? Just ask. 

It could be.  I'm asked "what is the use case" to understand why
you're wanting this functionality.

> If not, I will go for another
> candidate protocol.
>
> In the mobile operator network, two interfaces are needed if mobile GW
> and
> NAT are separate:
> 1. with NAT: request contains the NAT's public IP address, transport
> protocol and port number; response contains the private IP address and
> the
> port number.
> 2. with mobile GW: request contains the private IP address and response
> contains the mobile identities (e.g. IMSI, MSISDN, etc.)
> 
> If the mobile GW is integrated with the NAT, one interface is needed.
> The
> request contains the mobile GW's/NAT's public IP address, transport
> protocol
> and port number, and the response contains the mobile identities.

Sorry, I still don't understand the use case.

Is the use case to allow an external entity to perform user identification,
akin to draft-chen-intarea-v4-uid-header-option or
draft-zhang-v6ops-cgn-source-trace or draft-wing-nat-reveal-option ??

-d

> We keep our promises with one another - no matter what!
> 
> Best Regards,
> Tina TSOU
> http://tinatsou.weebly.com/contact.html
> 
> 
> -----Original Message-----
> From: Dan Wing [mailto:dwing@cisco.com]
> Sent: Wednesday, March 09, 2011 11:35 AM
> To: 'Tina TSOU'; pcp@ietf.org
> Subject: RE: [pcp] Will PCP do it?
> 
> > -----Original Message-----
> > From: pcp-bounces@ietf.org [mailto:pcp-bounces@ietf.org] On Behalf Of
> > Tina TSOU
> > Sent: Tuesday, March 08, 2011 7:36 AM
> > To: pcp@ietf.org
> > Subject: [pcp] Will PCP do it?
> >
> > Hi,
> > NAT has the private IP address, transport protocol and port # on the
> > "internal" side and the public IP address, transport protocol and
> port
> > # on the "external" side.  I'm looking for to allow an entity (e.g.,
> > not the PCP client who uses PCP to request a port number assignment
> on
> > the "external" side) to know the private IP address on the "internal"
> > side for the given public IP address, transport protocol and port #
> on
> > the "external" side.
> > Will PCP do it?
> 
> No.
> 
> What is the use case?
> 
> -d
> 
> 
> >
> > B. R.
> > Tina
> > http://tinatsou.weebly.com
> >
> >
> >
> >
> > _______________________________________________
> > pcp mailing list
> > pcp@ietf.org
> > https://www.ietf.org/mailman/listinfo/pcp