With ARIN’s announcement on September 24th 2015, that their Free Pool of IPv4 addresses has officially been depleted, comes the end of an era. The end of the IPv4 vs IPv6 battles, the end of tip-toeing around the subject of deploying IPv6, the end of IPv4. Will the world now finally accept IPv6 into it’s loving embrace and begin wide-spread deployment of this taboo protocol?
Probably not anytime soon. At least, in my opinion.
The problems you run into are the same problems that prevented most organizations from deploying IPv6 from the get-go. I’d like to illustrate some of these points, and in doing so, perhaps spread some awareness as to WHY it is taking so long for IPv6 adoption to become mainstream.
1. IPv6 Deployments Cost Money
Money, Money, Money, must be funny. Not really, it’s actually kind of sad. I say money before hardware and software because it really does boil down to this. IPv6 has been around for a while, with the IETF releasing several RFC’s defining Internet Protocol version 6 (IPv6) starting with RFC 1883.
Now people will tell you things like, “oh IPv6 isn’t or wasn’t supported by such and such hardware” or “but IPv6 doesn’t work with X software”. Those people were and are right. IPv6 WASN’T supported by many hardware and software platforms. Why though? Well, think of it in these terms; would YOU want to invest countless amount of $$$ in order to make a transition from IPv4 to IPv6 when instead you could simply keep buying up IPv4 address space? Changing out hardware, developing new software, updating existing software, working with any 3rd party vendors to introduce cross-compatibility, the list of things you’d have to do goes on and on!
This problem continues to exist today. As of this writing, there are only 25168 active IPv6 prefixes being announced via BGP. Compare that to the 574541 active IPv4 prefixes being announced, and you start to see the bigger picture. Of the 25168 active IPv6 prefixes, the vast majority are being announced by the major telecom and communications companies; including; Sprint, Time Warner Cable, Cox, Comcast, AT&t, Verizon, T-Mobile, and well, you get the idea. Why though? The answer is simple, money!
IPv4 is a valuable commodity now. Why would a for-profit organization waste that commodity on it’s end-users, when a cheaper long-term solution is available? By shifting towards IPv6-Only deployments, these guys free up valuable IPv4 address space, which they can now sell, lease out, or return to local RIR’s, FOR PROFIT.
The same trend exists with software developers. The slow adoption rate of IPv6 within certain softwares can be directly attributed to a lack of interest in spending money on research, development, and deployment. Take WHM/cPanel for example. This is a control panel which dominates the industry in terms of usage. Most web hosting companies use WHM/cPanel to provide a platform for their users to build websites on. It wasn’t until October 1st 2013 that WHM/cPanel introduced IPv6 support into their 11.40 branch, and it STILL isn’t 100% stable for use.
Overall, IPv6 adoption rates are on the rise. While the trend towards adoption still leans heavily towards major Telecom and Broadband providers, we’re seeing upticks in adoption by a variety of other services. End-user demands have played a major role is pushing IPv6 adoption forward, and as demand continues to rise, we’ll start to see IPv6 deployments become commonplace in corporate infrastructures.
2. IPv6 Deployment Takes Time and Resources (Which equal what? Money!)
The next major obstacle in my mind is the fact that there is never enough time, and never enough staff to get the job done. Do you focus on fixing bugs, introducing new features, and maintaining end-user support? Or do you focus on redesigning your application and/or network to support that IPv6 thing?
The reality of things is that time and resources are finite, you have to keep your customers happy and you can’t overwhelm your staff. Everyone was aware that IPv4 depletion was imminent, they simply chose to put off IPv6 adoption until the very last moment. It’s a human trait, we don’t change until pushed to the very brink of destruction.
3. Software Support for IPv6
What about legacy platforms whose developers have either closed shop, or are otherwise missing in action? Unfortunately, unless those platforms are open-source, they will continue to rely on IPv4. The IPv4 and IPv6 protocols can not directly communicate with each other, thankfully, there are mechanisms developed to deal with this situation such as Dual IP stack implementation, Tunneling, and Proxying and translation for IPv6-only hosts.
You again run into the problems of money and time/resources. While these technologies to enable transition exist, many organizations either lack the money and time, or lack the desire to expend money and time towards deploying these technologies to facilitate the transition to IPv6. It’s not as simple as flipping and switch and, VOILA!, IPv6 is now supported. You need to update your underlying network, your server configurations, your software configurations, your DNS settings, the list goes on and on.
In my opinion, we’ll see widespread adoption in software once end-user demand reaches critical mass. Many newer networked programs are shipping with IPv6 support baked in, but for those legacy and stubborn pieces of code, it’s going to take end-user’s kicking and screaming to force developers into rewriting their code.
4. Hardware Support for IPv6
Hardware support goes last on my list because we’ve had hardware support since almost as long as IPv6 was defined. Many hardware implementations back then were software based, but nowadays, most modern networking devices have support built in. The likes of Cisco, IBM, Nortel, to name a few, have been actively involved with IPv6 deployment since the 90’s.
The major excuse when it comes to hardware is again, related directly to money. Upgrading hardware costs time and money, and when you’re a for-profit organization, you don’t necessarily have the inclination to spend either. Like I mentioned earlier, there is the potential for profit to be made by simply reclaiming IPv4 address space, and turning around and leasing, selling, or returning said space. You could even cover your hardware upgrade costs.
As stated several times in this article, IPv6 support and deployment IS on the rise. It’s going to take continued end-user demand to push it forward, and it’s going to take some far-sighted individuals within their organizations to help facilitate that deployment.
While IPv4 isn’t going anywhere anytime soon, IPv6 is the future. Submit a ticket, write to your Senators, call the President, do whatever you need to do to enable an accelerated adoption of IPv6.
You can actively view the status of IPv6 route tables by visiting http://www.cidr-report.org/v6/as2.0/
Sprint, Time Warner Cable, Cox, Comcast, AT&T, Verizon, T-Mobile US,etc are not deploying IPv6 so they can sell it. They’re doing it because IPv6 has more addresses so scales better. With the exception of T-Mobile US they all use dual-stack for Internet access so they need to retain their IPv4 addresses.
Hey Ross,
I think you may have misunderstood what I was saying. I wasn’t talking about the re-sell value of IPv6 addresses, rather, the resell value of IPv4 addresses.
Not only can you “sell” your allocations, but you can lease them out indefinitely.
Check this out to see what I’m talking about: https://www.ipv4auctions.com/
If you’re a medium-large ISP and have a bunch of allocations, you can make a killing off your IP’s.