“You cannot manage what you cannot measure” the saying and the practical experience go.
So you enabled IPv6. Great! State that to the World and you better have a poncho on because the Gatorade will drench you. We in the IPv6 space got so little validation over time that we are happy when someone utters support for IPv6, let alone if she actually does something about it. It means our tribe is growing and another one finally decided to join the believers.
Of course, none of us will be there next to you when any and all new issues in your IT Universe are blamed on your IPv6 tinkering. None of us will be there to help you figure out what protocol is actually being used by the customer screaming at you. None of us will be there to even help you deal with the service provider who gave you the thumbs up on IPv6 when itself is far from being ready. Celebrations are over, real life sets in and it dawns on you that you now have TWO protocols to manage. Not only that but you still have to fend off the occasional: “Why did we do this in the first place?”
First, take comfort in the fact that you were absolutely right in attacking the IPv6 adoption head on. Smart people in your organization will thank you down the road but don’t hold your breath for a parade. Second, you knew from the beginning it is not just about flipping the switch, right? You knew all along it is about delivering services so, take the poncho off, sit comfortably at your desk and get cracking on monitoring IPv6. You still have time because, thankfully, the world doesn’t depend 100% on IPv6 just yet. You have time to figure out how this works in your environment, how to manage it during and after the transition.
You will come across the same issues you saw in IPv4. Below is a snapshot of us monitoring a service from multiple locations (www.v6sonar.com).
One location sees significant latency differences compared to other locations. It turned out to be an SP issue. We got it solved. Other locations dropped IPv6 access. We got that solved too. And ohh, by the way, we monitored IPv6 in parallel to IPv4 because that cannot suffer through this IPv6 enablement, not for the time being anyway.
Say, you had a maintenance window and IPv6 gets screwed up because the tech did not get IPv6 training or the procedures for IPv6 are not well oiled as they are for IPv4. You close the maintenance window, the tree fell in the woods and there was nobody to hear it (because users will use IPv4 as backup). Did it fall? Did it matter? It did and I bet you will find out at the most inconvenient of times as is always the case. This is a true story by the way.
The point is: Think about IPv6 in terms of production readiness, in terms of the quality of services that need to be delivered over it. To run it, you need the right tools. The good thing is that there is a little bit of time left to play with IPv6 before that customer is going to call you screaming and your boss will ask for metrics.
And as a parting thought: Good job on getting into IPv6, you saved yourself and your organization a lot of future pain!