[6bone] Who respect RFC2772 ? (4. Routing Policies for the 6bone)

[Daniel Austin]

Hi 6bone folk,

Just for the record, please dont use our looking glass or ASpath tree to
identify unaggregated prefixes as we filter *outbound* only.

However, should anyone see any unaggregated prefixes *from* us, please email
ipv6@kewlio.net and/or here and we'll fix it right away!


With Thanks,

Daniel Austin,
Managing Director,
kewlio.net Limited.
<daniel@kewlio.net>


----- Original Message -----
From: "Robert J. Rockell" <rrockell@sprint.net>
To: "Paul Timmins" <paul@timmins.net>
Cc: "Nicolas DEFFAYET" <nicolas.deffayet@ndsoftware.net>;
<6bone@mailman.isi.edu>
Sent: Thursday, October 31, 2002 5:53 PM
Subject: Re: [6bone] Who respect RFC2772 ? (4. Routing Policies for the
6bone)


> Nicolas,
>
> that's a view of our backbone. On our backbone, we'd probably want to
carry
> our customer routes (the /48's within 3ffe:2900::/24).   Forwarding would
> get weird if we didn't carry a RIB with our customer's prefixes in them.
:)
>
> You'll want to check an extrenal looking glass to see what is being HEARD
> from Sprint, and I'm hoping (if I can still configure a router; I think I
> can) that nothing is leaking. If it is, please let me know, and we'll be
sure
> to fix it.
>
> thanks for the reminder though. Please feel free to write me directly with
> any future infractions you see (or you can contact our group at
> ipv6-support@sprint.net).  I'll make sure to go through them all, and see
> what is getting exported outside of our ASN.
>
> I think 2772 leaves room to hear some more specifics, even from a pTLA, if
> it is for a load-sharing purpose (especially if people have set up their
> space geographically based), especially as the 6bone is still all ICMP,
and
> predominantly 'friendly'.  When we see more traffic, we will most likely
go
> back to a strictly 'hot-potato' enviroment for our peers, as is
predominant
> practice in bi-laterals today at the highest levels.
>
> remember, it's all about constraining non-aggreagtion so it doesn't
> contribue to a global problem.  I think that was the intention of the
> writing.
>
>
>
>
>
> Thanks
> Rob Rockell
> SprintLink
> (+1) 703-689-6322
> -----------------------------------------------------------------------
>
> On 31 Oct 2002, Paul Timmins wrote:
>
> ->On Thu, 2002-10-31 at 05:34, Nicolas DEFFAYET wrote:
> ->> Sprint: http://www.sprintv6.net/aspath/odd-routes1.html
> ->
> ->> The pTLA said in their pTLA request that they agree to all current and
> ->> future rules and policies !
> ->
> ->If you are trying to suggest that SprintV6 doesn't aggregate and
> ->announces long prefixes, I take issue with that.  Sprint does filtering
> ->very well. I recieve transit from them, and recieve routes via a private
> ->ASN, and not only has my block never been announced to their peers, I
> ->don't recieve the specifics of other transit customers of sprint.
> ->http://ipv6.timmins.net/bgp/odd-routes1.html
> ->In fact, the only specific I carry is my own. I am a transit customer of
> ->sprint, and I exchange prefixes with Viagenie as well.
> ->I have inbound filters, but sprint's announcements pass through 100%.
> ->I didn't want to get into this, but Sprint has always complied with RFC
> ->2772, and even as an end point, they've been willing to help me comply
> ->where applicable.
> ->That, and as my mom always said, "If everyone else jumped off a bridge,
> ->would you?" Because others break the rules doesn't give you permission
> ->to.
> ->
> ->-Paul
> ->
> ->--
> ->Paul Timmins
> ->paul@timmins.net / http://www.timmins.net/
> ->H: 248-683-7295 / C: 248-379-7826 / DC: 130*116*24495
> ->A: noweb4u / R: KC8QAY
> ->
> ->_______________________________________________
> ->6bone mailing list
> ->6bone@mailman.isi.edu
> ->http://mailman.isi.edu/mailman/listinfo/6bone
> ->
>
> _______________________________________________
> 6bone mailing list
> 6bone@mailman.isi.edu
> http://mailman.isi.edu/mailman/listinfo/6bone
>