On 03/04/2010 10:52 AM, Thomas Magill wrote: > 2. Longer than /24 prefixes in global BGP table. The most obvious > answer is that some hardware may not handle it... How is that hardware > going to handle an IP6 table then? I have had several occasions where > functionally I needed to advertise for different sites but only needed > 20-30 addresses which is a complete waste of a /24. How hard would it > be to start allowing /25s when compared to trying to roll out IP6?
prefix deaggregatation beyond /24 is probably inevitable but that doesn't mean you want people to burn routing table slots on your equipment for /28s. That routing table slot is an externality that everyone has to pay for. By holding the line to the extent that it is held, a cap of the growth rate of your dfz fib that is roughly congruent with rir policy. handling the v6 table is not currently hard (~2600 prefixes) while long term the temptation to do TE is roughly that same in v6 as in v4, the prospect of having a bunch of non-aggregatable direct assignments should be much lower...
