I guess that means vendor C has no excuse on the 7200 VXR series (and I believe a few of the newer models). But I still don't see anthing fantastically IPv6 happening there. Daryl G. Jurbala Introspect.net Consulting Tel: +1 215 825 8401 Fax: +1 508 526 8500 http://www.introspect.net <http://www.introspect.net/> PGP Key: http://www.introspect.net/pgp <http://www.introspect.net/pgp> -----Original Message----- From: stephen@sprunk.org [mailto:stephen@sprunk.org] Sent: Friday, June 13, 2003 12:48 AM To: eddy+public+spam@noc.everquick.net Cc: nanog@merit.edu Subject: Re: IPv6 [.....] Most L3 switches shipping today (e.g. the product in question) have particular ethertypes and destination address offsets hardcoded into their ASICs. It's not a matter of supporting 128-bit addresses -- they simply doesn't understand IPv6's header any more than they do DECnet or AppleTalk. While allocation policies may have an effect on how IPv6 FIBs are most efficiently stored, address length is a fairly small part of the problem when you're talking about redesigning every ASIC to handle both IPv4 and IPv6. [....]
Daryl G. Jurbala wrote:
I guess that means vendor C has no excuse on the 7200 VXR series (and I believe a few of the newer models). But I still don't see anthing fantastically IPv6 happening there.
The 7206VXR (along with all 7200/7400) supports IPv6 in IOS 12.2S, 12.2T and 12.3. 12.2S is not yet the recommended IOS for ISPs using 7200 series routers, but is expected to be the recommended ISP IOS in the future. 12.2T/12.3 are the current recommended IOS for LNSs. The support seems to gradually increase with time. As a 7xxx-based network, IPv6 came as a pleasant surprise when we upgraded IOS for other reasons. Some current significant missing items: NetFlow of IPv6 (...sure we can do the traffic, but can we bill it?!...) Communication by the router to IPv6 hosts for RADIUS, tacacs, NetFlow, etc (...ok, so we can use IPv6, just as long as it's not for anything our routers want to talk to...) But for that matter, NetFlow doesn't work well with MPLS either (in a VPDN environment, where you can't identify a virtual access interface based on SNMP and hoping it was the same when the traffic happened, there's no way currently to identify which VRF the traffic occurred in and thus no way to map traffic back to a customer when multiple VRFs contain the same IPs). David.
participants (2)
-
Daryl G. Jurbala -
David Luyer