Bad routes update

[Craig Metz]

In message <Pine.GSO.3.96.990719221751.22806C-100000@iscone>, you write:
>3FFE:F00:2::0/48		11261

  This is not a bogus route per se. I told them to set it up this way. I have
been recommending that non-pTLA multi-homed sites use a /48, which happens to
mesh nicely with the way I've been managing tunnel address space at NRL.

  I can see a very very strong case for making it /32 and setting a hard and
fast rule that prefixes >/32 are not allowed into the cloud. I think that it
would be a Very Very Good Thing if no backbone router ever had to look at
beyond the first 32 bits of the address, as this would make life a Whole Lot
Easier for hardware that is designed for the best-performance case being 32
bit addresses (like, oh, say, most backbone routers).

  Now, these other prefixes are bogon.

  I'd like to add one I saw in my routing tables:

  3FFE:F00:A:1133::0/64
         *> 3FFE:F01:0:FFFF::5 FE80::C8E:50C2:7 Tunnel1 1225 1103 65502 5623 559 1717 1835 1849 109 3462 3263 49 i
         *  3FFE:F01:0:FFFF::19 FE80::E0:1E8E:C2C1:18 Tunnel6 5609 1225 1275 1103 222 5623 559 1717 1835 1849 109 3462 3263 49 i
  3FFE:F00:A:11E1::0/64
         *> 3FFE:F01:0:FFFF::5 FE80::C8E:50C2:7 Tunnel1 1225 1103 65502 5623 559 1717 786 1849 109 3462 3263 49 ?
         *  3FFE:F01:0:FFFF::19 FE80::E0:1E8E:C2C1:18 Tunnel6 5609 1225 1275 1103 222 5623 559 1717 786 1849 109 3462 3263 49 ?

  NIST guys, please fix.

>3FFE::0/16			2497 (prepended)

  Sigh.

>When Ipv6 goes live, unless business is more good-willed than it is now,
>this is going to break things, and one pTLA may not have much motivation to
>fix the problem (unless flames on the 6bone mailing lists really hurt).

  What's basically going to have to happen soon is that we're going to have to
turn on the basic BGP knobs. I'd like to see routing transit policy stuff in
place on the 6Bone as well as reasonable path metrics and such, as it would
stop much of the current tunnel routing insanity.

  Perhaps the Cisco guys could elaborate, but I had the impression that only
a very limited subset of the knobs are really built for IPv6. This might
complicate things.

									-Craig