IPv6 has a long way to go
Implementing the networking protocol will take huge investments
By Keir Thomas | PC World | Published: 16:05, 10 March 2011
There's maybe never been a better time to be a network engineer, that is provided you have IPv6 experience.
Businesses are crying out for individuals to help create next generation networking circuits, and the heat has been turned up as World IPv6 Day approaches in June, during which many of the Internet's most popular properties will open IPv6 entrances to give the technology the biggest test it's ever had.
Internet Protocol version 6, known as IPv6, is the new kid on the block when it comes to Internet addressing, the system by which computers can be uniquely identified on the Internet and data routed through to them. It offers so many Internet addresses that we need a word so little used that it doesn't even appear in dictionaries right now: IPv6 offers over well 340 undecillion addresses.
Related Articles on Techworld
However, IPv6 has actually been around since the late 1990s, when it was created to supersede the older v4 of the Internet Protocol scheme (IPv4). The rapid implementation of IPv6 at the moment has been made imperative by the recent depletion of IPv4 addresses.
Although IPv6 is arguably on its way (Verizon and Comcast recently began commercial trials, for example), there are a surprising number of bananas in the road.
In fact, it's extraordinary that we're at the stage of rolling out a world-changing technology when so many issues are yet to be addressed.
For example, a surprising number of home and small business routers still don't offer IPv6. You won't presently find a Linksys wireless router that supports IPv6, for example, although Cisco, the parent company of Linksys, says that the functionality will be added to new Linksys products real soon now, and firmware updates will be made available for products sold recently.
It's not clear whether the same updates will be made available for older hardware but, considering the amount of work involved and the fact that companies will receive no income from doing so, it seems unlikely. According to Cisco, there are also practical limitations, such as a lack of memory in older devices.
Of the routers that already claim to support IPv6, many of the implementations are buggy and might not even work. This even includes those certified as IPv6-compliant by the IPv6 Forum.
Security experts are also warning that the sheer number of addresses available via IPv6 will make spam blocking impossible. Currently, the most effective anti-spam technique is to simply blacklist certain internet addresses that are known to send spam messages.
This works because it's difficult for spammers to acquire additional IP addresses. However, because IPv6 has such a huge range of available addresses, spammers could easily acquire thousands if not millions of them. Each individual spam message could come from a unique IPv6 address, making accurate blocking impossible.
Because of this, there's talk of sticking to IPv4 for mail servers. Any incoming mail from untrusted IPv6-based mail servers could simply be bounced back. Until a reliable technology is invented, mail servers may never break free of IPv4-based addressing.
It isn't just spam that relies on blacklisting. Preventing denial of service (DoS) attacks, click fraud and preventing manipulation of search engines could also be hit, according to a story in The Register.
In fact, experts have pointed out that IPv6 is simply not yet as developed as IPv4 when it comes to security, and this is sure to increase costs, if not introduce episodic chaos.
The big problem is that IPv6 was designed by scientists to be a perfect implementation of a technology. It failed to take into account real world issues. In 2009, the Internet Engineering Task Force (IETF), the people behind IPv6, admitted that they hadn't ensured that IPv6 would play happily with IPv4.
IETF Chair Russ Housey admitted things weren't going to plan and added, "Our transition strategy was dual-stack, where we would start by adding IPv6 to the hosts and then gradually over time we would disable IPv4 and everything would go smoothly."
Sadly, as we can see today, this simply hasn't happened, despite additional technology being developed by the IETF to bridge the gap. We shouldn't be switching to IPv6 today. It should have happened years ago.
All of this could have been avoided if IPv6 had been made backwards compatible with IPv4, which would have made transition seamless for most organizations. Instead, it was decided that the internet addressing system should be rebooted from scratch.
It's because of issues like this that IPv6 is by no-means a shoe-in over the coming years. It may well simply be ignored. There are other ways of getting around the address depletion issue, such as carrier-grade Network Address Translation, where a neighborhood shares a single IP address.
This is vastly inferior to IPv6 in many ways but it has the advantage that it's easy to implement, and that's really all that matters in the commercial world of the Internet. Additionally, there's talk of a commercial marketplace for IP address blocks arising in the future, and the buying and selling of addresses might provide an additional incentive to simply ignore IPv6.