ARIN PI space filters (2620:0000:/23)

[Jørgen Hovland]

Hi Bjoern,

-----Original Message-----
From: ipv6-ops-bounces+jorgen=hovland.cx at lists.cluenet.de [mailto:ipv6-ops-bounces+jorgen=hovland.cx at lists.cluenet.de] On Behalf Of Bjoern A. Zeeb

> why shoudn't it? Clueful people did have filters from close to day 1.

Why did this thread start again? 


> Gert Doering has been maintaining his public filter lists[1] since 2002.

No offence regarding Gert, but he is maintaining a manual list that requires human intervention. That is not particularly clueful. Just check the archives.


> Why would someone allow anyone to go where v4 ended and let people
> announce any &$#&*()* prefixes, 

Because the implementation permit such things.
It is however the inconsistent policies also being modified all the time that are causing the problems.

> highly deaggregated stuff, 
True

> hijacked netblocks
Can easily be done with prefix filtering too. I have successfully tested it using two of our ISPs seperate networks in two different countries. 

> unallocated prefixes, ...?
There will always be unallocated prefixes in your accepted prefix filter. If you claim anything else, then you must be updating your prefix filter every minute.


> If it's not stopped now it will never be and v6 will become the same
> v4 is now. Can you do the math?

And you want to stop it with a prefix filter list?