You can still carry the v6 NLRIs in MP-BGP though right? Joshua Moore Network Engineer ATC Broadband 912.632.3161 - O | 912.218.3720 - M From: Mel Beckman [mailto:mel@beckman.org] Sent: Monday, July 06, 2015 10:49 AM To: andrew Cc: Lee Howard; Josh Moore; nanog@nanog.org Subject: Re: Dual stack IPv6 for IPv4 depletion MPLS requires an IPv4 core. You can't run an IPv6-only infrastructure because neither CSCO or JNPR have implemented LDP to distribute labels for IPV6 prefixes. -mel via cell On Jul 6, 2015, at 7:15 AM, andrew <andrew@ethernaut.io<mailto:andrew@ethernaut.io>> wrote: Pardon my ignorance - what do you see missing in MPLS in regards to support for IP6? -------- Original message -------- From: Mel Beckman <mel@beckman.org<mailto:mel@beckman.org>> Date: 07/06/2015 9:44 AM (GMT-05:00) To: Lee Howard <Lee@asgard.org<mailto:Lee@asgard.org>> Cc: Josh Moore <jmoore@atcnetworks.net<mailto:jmoore@atcnetworks.net>>, nanog@nanog.org<mailto:nanog@nanog.org> Subject: Re: Dual stack IPv6 for IPv4 depletion And let's all complain to the MPLS working group to get IPv6 support finished up! -mel beckman
On Jul 6, 2015, at 6:27 AM, Lee Howard <Lee@asgard.org<mailto:Lee@asgard.org>> wrote:
Some thoughts. . .
³Native dual-stack² is ³native IPv4 and native IPv6.²
³Dual-stack² might be native, or might by ³native IPv6 plus IPv4 address sharing.²
Your IPv4 address sharing options are CGN, DS-Lite, and MAP. There are operational deployments of all three, in the order given. You need them close enough to your customers that traffic will return over the same path. You can¹t share state among a cluster of boxes, but that¹s not the end of the world; a device failure sometimes causes loss of state. MAP is the hot new thing all the cool kids are doing.
Look to your router and load balancer vendors for devices that do these. CGN is the only one that doesn¹t require updates to the home gateway. The more IPv6 your customers use, the smaller your CGN/AFTR/MAP can be.
Think about how you¹ll position it to customers. It¹s difficult to change a customer¹s service mid-contract. At some point, a customer is no longer profitable: if NAT costs and service calls add up, you may be better off buying addresses or losing the customer. You may need to buy some IPv4 addresses to give you time; contact a broker.
You may be surprised how hard it is to root IPv4 out of the system. Don¹t buy anything you can¹t manage over IPv6, including servers and applications. Sorry, vendor, I can¹t buy your cluster, I don¹t have the IPv4 address space to provision it.
Lee
On 7/4/15, 8:09 AM, "NANOG on behalf of Josh Moore" <nanog-bounces@nanog.org<mailto:nanog-bounces@nanog.org> on behalf of jmoore@atcnetworks.net<mailto:jmoore@atcnetworks.net>> wrote:
Traditional dual stack deployments implement both IPv4 and IPv6 to the CPE. Consider the following:
An ISP is at 90% IPv4 utilization and would like to deploy dual stack with the purpose of allowing their subscriber base to continue to grow regardless of the depletion of the IPv4 space. Current dual stack best practices seem to recommend deploying BOTH IPv4 and IPv6 to every CPE. If this is the case, and BOTH are still required, then how does IPv6 help with the v4 address depletion crisis? Many sites and services would still need legacy IPv4 compatibility. Sure, CGN technology may be a solution but what about applications that need direct IPv4 connectivity without NAT? It seems that there should be a mechanism to enable on-demand and efficient IPv4 address consumption ONLY when needed. My question is this: What, if any, solutions like this exist? If no solution exists then what is the next best thing? What would the overall IPv6 migration strategy and goal be?
Sorry for the length of this email but these are legitimate concerns and while I understand the need for IPv6 and the importance of getting there; I don't understand exactly HOW that can be done considering the immediate issue: IPv4 depletion.
Thanks
Joshua Moore Network Engineer ATC Broadband 912.632.3161