[LACNIC/Politicas] Nueva propuesta LAC-2018-14 / Nova proposta LAC-2018-14 / New proposal LAC-2018-14
Alejandro Guzman
alejog at google.com
Wed Feb 13 15:53:22 -02 2019
I would actually add that such policy won't ease the depletion of the
remaining address space...
On Wed, Feb 13, 2019 at 9:48 AM Arturo Servin <arturo.servin at gmail.com>
wrote:
>
> >Should we release a policy that will ease the depletion of our
> >remaining address space? In exchange of what exactly? A cleaner RIR
> >database with coherent data? That's not enough reason in my opinion.
>
> Short answer yes.
>
> There is no much space left anyway. And we need to keep the database
> clean, otherwise we face larger problems.
>
> To me it is enough reason, sorry to be so opposite to your opinion.
>
> Regards
> as
>
> On Wed, Feb 13, 2019 at 4:36 AM Alejandro Guzman via Politicas <
> politicas at lacnic.net> wrote:
>
>> Luis y Fernando
>>
>> Que pasa si no hay una política... seguimos como estamos... las
>> transferencias han ocurrido, están ocurriendo y seguirán ocurriendo. Aún
>> los pioneros en IPv6 cuya red soporta IPv6 hace años siguen necesitando
>> IPv4 por la realidad actual... y al haber demanda, pues se darán las
>> transferencias.
>>
>> Que pasa si tenemos una política que las permita con cierta regulación y
>> orden? Pues al menos ganamos que estas se den en el marco de las
>> políticas,
>> a la luz y registradas. De hecho el que ya o deban ser ocultas permite que
>> en muchos casos no se necesite un broker...
>>
>> Esta política no causa ningún "harm" pues el supuesto "harm" existe aún
>> sin
>> la política... y al menos produce algún bien.
>>
>> Alejandro
>>
>>
>>
>> On Tue, Feb 12, 2019 at 6:46 PM Luis Balbinot <luis at luisbalbinot.com>
>> wrote:
>>
>> > I'm not being mean, I'm being real. I don't see how this policy will
>> > do more good than harm.
>> >
>> > Should we release a policy that will ease the depletion of our
>> > remaining address space? In exchange of what exactly? A cleaner RIR
>> > database with coherent data? That's not enough reason in my opinion.
>> >
>> > I'd like to hear the opinion of other operators.
>> >
>> > Luis
>> >
>> > On Tue, Feb 12, 2019 at 10:45 PM Nicolas Antoniello
>> > <nantoniello at gmail.com> wrote:
>> > >
>> > > Hi Luis,
>> > >
>> > > As you may probably know, lots of traffic does not mean lots of
>> > addresses.
>> > > Normally the ones that consume more IP addresses are those who provide
>> > > connection services, not those who provide data alone.
>> > >
>> > > And the fact that you imply that RIR registries (like PKI
>> repossitories,
>> > IP
>> > > allocation database, reverse registries and whois itself) are not
>> being
>> > of
>> > > much use nowadays reflect the need for fixing that and for maintaining
>> > > coherency about data stored.
>> > >
>> > > Lastly if you just don’t care about all that (not your concern you’ve
>> > > mentioned) let me ask you why you’re being so mean on not having a
>> policy
>> > > that may, at less, better that.
>> > > ‘Cos as far as I’m concerned we -don’t care- protecting an IP space
>> > that’s
>> > > already exausted, right?
>> > >
>> > > Regards,
>> > > Nico
>> > >
>> > >
>> > >
>> > >
>> > > El El mar, 12 de feb. de 2019 a las 21:33, Luis Balbinot <
>> > > luis at luisbalbinot.com> escribió:
>> > >
>> > > > >
>> > > > > But still, while we don’t have that policy, transfers occur and so
>> > the
>> > > > > RIR’s registries are inconsistent... and RPKI fails, and BGP
>> seems to
>> > > > mess,
>> > > > > and so many things that we may avoid !
>> > > >
>> > > >
>> > > > And exactly why does that should concern me or anyone else? It's a
>> good
>> > > > thing that RPKI fails and BGP has been a mess for a very long time
>> > thanks
>> > > > to cloud providers and CDNs. RIR registries are of very limited use
>> > today.
>> > > > All we get are abuse mailboxes that nobody reads and very limited
>> and
>> > > > outdated owner information. People trust IRR databases, PeeringDB
>> and
>> > > > geolocation data more than whois data.
>> > > >
>> > > > Again: I'd like to see examples of such transfers in action because
>> I
>> > don't
>> > > > think they are relevant. I've worked with Goodle, Akamai,
>> CloudFlare,
>> > > > Netflix, Microsoft, Facebook and Amazon. They are all using
>> legitimate
>> > > > address space in the LAC region. And those alone account for *a
>> lot* of
>> > > > traffic.
>> > > >
>> > > > It's not common practice to advertise address space from other ASNs
>> in
>> > the
>> > > > LAC region with LOAs like it happens in the ARIN region. It should
>> be
>> > easy
>> > > > to track and make reports on that.
>> > > >
>> > > > Luis
>> > > > _______________________________________________
>> > > > Politicas mailing list
>> > > > Politicas at lacnic.net
>> > > > https://mail.lacnic.net/mailman/listinfo/politicas
>> > > >
>> > > _______________________________________________
>> > > Politicas mailing list
>> > > Politicas at lacnic.net
>> > > https://mail.lacnic.net/mailman/listinfo/politicas
>> > _______________________________________________
>> > Politicas mailing list
>> > Politicas at lacnic.net
>> > https://mail.lacnic.net/mailman/listinfo/politicas
>> >
>>
>>
>> --
>>
>> Alejandro Guzman | Interconnection Director - LATAM | alejog at google.com
>> | +1
>> (650) 4268561
>>
>> If you get an email from me outside working hours it is not because I am
>> always on or expect an immediate response from you, it is because work
>> flexibility is key for me. Evening and weekend emails are a sign that I
>> prioritized personal obligations during the week.
>> _______________________________________________
>> Politicas mailing list
>> Politicas at lacnic.net
>> https://mail.lacnic.net/mailman/listinfo/politicas
>>
>
--
Alejandro Guzman | Interconnection Director - LATAM | alejog at google.com | +1
(650) 4268561
If you get an email from me outside working hours it is not because I am
always on or expect an immediate response from you, it is because work
flexibility is key for me. Evening and weekend emails are a sign that I
prioritized personal obligations during the week.
More information about the Politicas
mailing list