-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi Ahmed, others,
Do we have any ISPs/ Carriers in the ME region that provide v6 to end-users (either as Tunnels, or native ). thanks - -gaurab On 6/9/11 1:41 PM, Ahmed Abu-Abed wrote: > My comments below with a minor correction in the 1st comment. > > -Ahmed > > > *From:* Brian Candler <mailto:[email protected]> > *Sent:* Thursday, June 09, 2011 12:54 PM > *To:* Ahmed Abu-Abed <mailto:[email protected]> > *Cc:* 'menog@menog. net' <mailto:'menog@menog. net'> > *Subject:* Re: [menog] Rapid IPv6 deployment for World IPv6 Day > > On Thu, Jun 09, 2011 at 10:51:26AM +0300, Ahmed Abu-Abed wrote: >> >> Until the whole internet AND web content AND networks AND >> applications move to IPv6 ONLY then there will be a need for tunneling. >> Dual-stacking is needed but it doesn't solve the IPv4 depletion issue, > > But client-side tunnelling relies on having an IPv4 address too, so it > doesn't solve depletion. > />> It may not solve depletion but tunneling accelerates IPv6 > adoption, while "carrier grade" tunnels present a more stable approach > than multi-level NATs. Future networks will likely be IPv6-only except > for the dual-stack hosts that tunnel IPv4-in-IPv6 , a reverse of today's > tunnels. Refer to the DS-Lite standards among others, and this approach > is part of the 3GPP/LTE standards for mobile networks migration to IPv6./ > // > Estimates I've seen so far from IPv6 day suggest that although traffic was > up, V6 accounted for between 0.02% and 0.3% of total traffic. Of that, 90% > was tunnelled (i.e. only 10% native). So basically: (1) there is no > signficiant IPv6 Internet today, and (2) if you want to join what there is, > you do indeed probably have to tunnel. > />> Tunnels solve the chicken and egg problem, it allows IPv6 content to > be accessible until IPv6 gains a wider installed base which may take > years. During the same time the burden of running IPv4 with no public > addresses will grow. Both RIPE NCC and ARIN have publicly endorsed > tunneling to speed up IPv6 deployment./ > > That doesn't mean that installing a tunnel client is a good idea for anyone > except network specialists who know what they're doing. > />> Protocols that automate the setup of carrier grade tunnels, such as > TSP, make installing tunnels a plug and play affair. Refer to my > original email and try it to see for yourself. For zero user > intervention needs, there are CPE IPv6 Adapters that plug in an ethernet > port on an IPv4 home router and automatically setup IPv6-in-IPv4 tunnels > (see the ARIN Wiki on IPv6 CPEs). All these are carrier grade solutions > and have been deployed by tier-1 carriers./ > > If random end-users start installing this stuff without understanding it, > then (a) they are probably opening up security holes into their network, and > (b) they may impede a later smooth rollout of native v6. > />> Whether users dual-stack or tunnel to IPv6 the security requirements > are mostly the same. Waiting for end-to-end dual-stack to be deployed > all the way to the home CPE to complete is a multi year project. And I > don't see IPv6-in-IPv4 tunnels impeding rollout of native IPv6 if the > tunnels clients AND servers are fully under control of the ISP and part > of their network. The rule here is for ISPs to avoid using Teredo, 6to4 > and ISATAP tunnels which, unfortunately, are everywhere./ > > > > > _______________________________________________ > Menog mailing list > [email protected] > http://lists.menog.net/mailman/listinfo/menog - -- http://www.gaurab.org.np/ -----BEGIN PGP SIGNATURE----- Version: GnuPG/MacGPG2 v2.0.16 (Darwin) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAk3y+z0ACgkQSo7fU26F3X1s2gCdGltM4CP9WI5TpN77JHD45ZaD UKIAoNsEQ1lmQ5x60svuQxP/O75TpBOO =UuY3 -----END PGP SIGNATURE----- _______________________________________________ Menog mailing list [email protected] http://lists.menog.net/mailman/listinfo/menog
