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

[Jeroen Massar]

JAKO Andras wrote:
>>> Apparently there are quite a few ASN's who still filter the /48 PI's
>>> from ARIN.
>>>
>>> Please verify your filters are up to date or bug your transits when your
>>> ASN appears in the missing list, as can be checked live at:
>>> http://www.sixxs.net/tools/grh/lg/?find=2620:0:230::/48
>> Another ARIN PI /48 to double check against is 2620:0:c0::/48
>> I've been announcing that prefix via Level3/AS3356 for a few months
>> and you'll see it suffers from the same reachability problems
>> Marc Blanchet is seeing with his 2620:0:230::/48 block.
> 
> Are there any pingable addresses available? I think this is one of the 
> usual methods nowadays to help resolve such issues (at least in IPv4). 
> Even the RIRs themselves set up pingable hosts in fresh v4 /8s. (Although 
> I'm not sure whether all RIRs do that.)
> 
> See also http://www.ris.ripe.net/debogon/debogon.html

Although it would be useful, RIS does not do debogon's yet for IPv6, but
I am pretty sure that if it would be nicely requested from them that
they could set something up.

The main way to check it, is by checking GRH to see if the announcement
comes through. In general, for IPv6, folks do BGP filtering, not packet
based (at least from my limit knowledge on this).

What works most of the time, is a traceroute6 towards <prefix>::1, you
will then at least end up at the end of the BGP path.

There is no 'standard' saying that ::1 has to be pingable. Maybe an
extra attribute in an inet[6]num could specify a server which can be
used for reachability tests in that prefix?

Greets,
 Jeroen


-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 311 bytes
Desc: OpenPGP digital signature
Url : http://lists.cluenet.de/pipermail/ipv6-ops/attachments/20070319/cef959bf/signature.bin