Today is IPv6 day. Nothing happened, no machines were broken and no web sites died. The test is pointless and it is going to take a lot more commitment and a real plan to get users moving to IPv6.

Today is IPv6 day.

Or if you are reading this later June 8th 2011 was IPv6 day.

Nothing happened, no machines were broken and no web sites died.

A non-event is the only way to describe it. Of course everyone involved will claim that this was exactly the point - IPv6 being switched on is a non-event.

Well given that most of the Internet is currently IPv4-only, yes indeed, switching on IPv6 on one or two high profile sites is hardly going to make a difference.

The clients who try to connect to the sites using IPv4 will still just connect via IPv4 and completely ignore the existence of IPv6 addresses, DNS and the whole IPv6 infrastructure for that matter.

The point is that the hardware that carries IPv6 is the same as the hardware that carries IPv4 and both protocols can coexist.

An IPv6-connected client can use an IPv6 or an IPv4 service but an IPv4 connected client can only use an IPv4 service.

So let's think about this for a moment. If all of the servers in the world suddenly switched to dual stack IPv4 and IPv6 the actual use of IPv6 would change very little as most of the clients would remain as IPv4 users and ignore the IPv6 facility. IPv6 day has just simulated switching a few high profile sites to dual stack IPv4 and IPv6 so it really isn't a big deal.

There really isn't much point in servers, i.e. web sites, switching to IPv6 until the number of IPv6 clients reaches a critical mass that mans that an IPv6 only website doesn't miss out on a significant proportion of the potential traffic. Of course there is no point in ISPs upgrading their connections to produce that critical mass until there are some servers actually using IPv6 and only IPv6. Until there are IPv6 only server running services that people want to use there is no pressure on the ISPs to spend the money on the upgrade.

Chicken and egg?

Not really - it's more like deadlock.

So happy IPv6 day and remember after this most of the sites are going back to IPv4 only. Why? I have no idea but without a more serious commitment and without a real plan for getting clients moved over to IPv6, i.e. getting the ISPs to move in the IPv6 direction, it is going to take a long, long time to see IPv4 as old technology.

I'd start breaking out the NAT routers and trading in your unused and underused IPv4 addresses for money if you can...

To learn more about IPv6 and why it will be needed in the future see IPv6: The Programmer's View