Re: IPCP option for SIP server
James Carlson <[email protected]>
| Newsgroups | gmane.ietf.pppext |
|---|---|
| Message-ID | <[email protected]> |
Ashutosh Dutta writes:
> > Note that DHCP itself needn't necessarily be involved in the IP
> > address allocation strategy at all. It's possible to use DHCP to
> > distribute configuration details without doing stateful address
> > allocation. See the DHCPINFORM messages (cited earlier) for details.
>
> James agreed, PPP server itself can keep the IP address pool and
> dispense it and DHCPINFORM can probably get other server details using
> DHCPINFORM. Only requirement is then to have a DHCP server in the
> backend and have a DHCPclient of some sort in the NAS server.
If you instead put that client in the remote node ("N1" in my diagram)
rather than in the NAS ("N2" in my diagram, though I was intentionally
showing PPP itself as a symmetric link), then you don't need any PPP
changes at all, and you can get *ALL* of the other configuration
features of DHCP for free without needing to change IPCP for each one.
That's the point. There's an existing solution that works well, is
widely deployed, and is quite easy to implement. We thus don't need
another.
In addition to that, adding SIP configuration here would just beg
other questions. Should Ethernet have a way to do this as well,
perhaps via 802.1X extensions? Why shouldn't all L2 protocols do
this? What happens if the remote node supports both DHCP and the
proposed IPCP option, and queries to these return different answers?
Which one is considered authoritative and why?
Adding new features to an existing protocol isn't a matter of "could,"
but rather a matter of "must." In other words, we do it as a last
resort because nothing else can reasonably be made to work. That's
not the case here.
--
James Carlson, KISS Network <[email protected]>
Sun Microsystems / 1 Network Drive 71.232W Vox +1 781 442 2084
MS UBUR02-212 / Burlington MA 01803-2757 42.496N Fax +1 781 442 1677
_______________________________________________
Pppext mailing list
[email protected]
https://www1.ietf.org/mailman/listinfo/pppext