RE: [Geopriv] Geopriv L7 LCP: New Requirement

From: Winterbottom, James ^lt;James.Winterbottom@andrew.com>
Date: Wed Feb 14 2007 - 15:40:02 EST

Brian, I am completely gob-smacked by some of your assertions below. You have been heavily involved in the work we have done in NENA is arriving at both requirements and a series of possible solutions for DSL networks. You are aware that location in these networks is dependent on tracing identifier chains through the network. That at the chain starts at the ISP with the IP address and works its way back through various identifiers, which may include L2TP tunnel and session identifiers etc. You have also seen the suggested mechanisms for passing this information back from the ISP LIS to the RANP-LIS. I agree with Barbara, in the DSL world the IP address is a good identifier, but it is ultimately not the only identifier required to provide location, it provides links to other identifier that ultimately allow location to be determined. The issue is that some of these intermediary identifier are dynamic, so you cannot simply provision location against IP address or user-identifier in the LIS. The document Barbara and I have both posted links to now show how this can be done. I would urge those that have not read it. Cheers James > -----Original Message----- > From: Brian Rosen [mailto:br@brianrosen.net] > Sent: Thursday, 15 February 2007 12:48 AM > To: 'Otmar Lendl'; 'Andrew Newton' > Cc: geopriv@ietf.org; 'Stark, Barbara'; hgs@cs.columbia.edu > Subject: RE: [Geopriv] Geopriv L7 LCP: New Requirement > > Well, nothing is foolproof. The Cable Modem guys are stuck with using MAC > address of the cablemodem, but it is possible on some systems to take your > CM to your friends house and plug it in. As long as (in most systems) you > are in the same CMTS area, it will not know that you moved it. > > I had been under the assumption that the IP address correlated to a > customer > which correlated to an address, but now that you point out that a > username/password is the key to the customer ID, it's obvious that won't > work. What is needed is something akin to the cablemodem MAC address. Is > there something? > > For some time now, several of us have been suggesting that IP address is > not > a good identifier for many networks. Can it be that DSL is among the > networks where IP address cannot reasonably be used to determine location > from a LIS? I would guess that there is something the client could get > at, > which would be a good identifier. Even if that were something tied to the > DSL modem, it wouldn't be much different than the cable systems. Best, of > course, would be something that could be correlated to the DSLAM port. > > Brian > > > -----Original Message----- > > From: Otmar Lendl [mailto:lendl@nic.at] > > Sent: Wednesday, February 14, 2007 5:29 AM > > To: Andrew Newton > > Cc: geopriv@ietf.org; Stark, Barbara; hgs@cs.columbia.edu > > Subject: Re: [Geopriv] Geopriv L7 LCP: New Requirement > > > > On 2007/02/13 21:02, Andrew Newton <andy@hxr.us> wrote: > > > On Feb 13, 2007, at 3:01 PM, Stark, Barbara wrote: > > > > > > >While DSL lines don't (or shouldn't) move, the L2TP tunnel, or > > > >router port, where traffic for a given DSL customer comes into an > > > >ISP's network, does change. PPPoE authentication info isn't tied to > > > >a DSL line (I can use my login and password from anybody's > > > >bellsouth.net DSL line). So, the ISP can't assume, that just > > > >because they see traffic from a particular customer, that the > > > >customer's traffic is coming from his DSL line. The ISP needs to > > > >ask the access provider where that traffic physically originates > from. > > > > > > I was attempting to note that this information is static. Regardless > > > of how the traffic flows in the network, the location of the DSL end > > > point remains the same. Therefore, this information could be passed > > > on the forms filled out in triplicate when the customer signs up. > > > Even if this does not happen, some identifier relating to the > > > customer has to be provisioned that can be used as a reference. > > > > > > That being said, if you feel the information is dynamic and in need > > > to a protocol so that it can be updated, I'll take your work for it. > > > > Andy, > > > > the location of a DSL line ist static in the sense that once > > a customer has logged in via that line, his location remains > > the same. > > > > What Barbara (and I) tried to explain is that DSL lines with > > username/password based authentication are more similar to > > dialup accounts than to leased lines. > > > > When a reseller of such an DSL service adds a new customer, > > two things happen: > > > > * the customer's line is connected to a DSLAM port > > * a username/password pair is generated, communicated, and activated > > > > There is *no* check in the system which restricts this username/password > > pair to this specific DSL line. > > > > The LIS of the ISP thus CANNOT operate like this: > > > > 1) Map the source IP address of the RELO/HELD query to the customer > > account. > > 2) Look up the address of that customer's DSL line. > > 3) Return that address. > > > > This is just as misguided as using the customer's home address of a > > dialup-user as the location. > > > > /ol > > -- > > < Otmar Lendl (lendl@nic.at) | nic.at Systems Engineer > > > > > _______________________________________________ > > Geopriv mailing list > > Geopriv@ietf.org > > https://www1.ietf.org/mailman/listinfo/geopriv > > > _______________________________________________ > Geopriv mailing list > Geopriv@ietf.org > https://www1.ietf.org/mailman/listinfo/geopriv ------------------------------------------------------------------------------------------------ This message is for the designated recipient only and may contain privileged, proprietary, or otherwise private information. If you have received it in error, please notify the sender immediately and delete the original. Any unauthorized use of this email is prohibited. ------------------------------------------------------------------------------------------------ [mf2]

_______________________________________________
Geopriv mailing list
Geopriv@ietf.org
https://www1.ietf.org/mailman/listinfo/geopriv
Received on Wed, 14 Feb 2007 14:40:02 -0600

This archive was generated by hypermail 2.1.8 : Wed Feb 14 2007 - 15:39:54 EST