Since IPv6 is going to replace IPv4, should I develop a new application to support it?

Question

I develop applications using many languages ​​and several platforms. From what I'm reading, IPv6 will be distributed over the next 5-10 years and that some Internet service providers even offer a direct IPv6 Internet connection. On my private network, all machines have a double connection to the glass and communicate with each other using IPv6 whenever possible.

My question is: should my new application be written to support IPv6? If so, what should I consider? Are there any flaws in IPv6 support now?

I don't know if this will help or hurt, but here are a few more specific details to guide the answers a bit:

My application will run on Windows / Linux / Mac, and it will consist of a platform-specific component written separately for each platform and a common “core” component written in an interpreted language.

My application should be able to find other instances running on any of the three platforms, and support both TCP and UDP between any two machines.

Answer 1

Yes: all your new applications must support IPv6. As you said, it is already used in many LAN environments.

No: you don’t have to change much to support IPv6. The underlying structure should take care of the differences for you. This assumes that you are not doing “bad” things, such as storing or otherwise manipulating IP addresses.

Answer 2

I see some of them citing Steve Gibson's opinion on this. Although Mr. Gibson is quite competent in security matters (I listen to every episode of “Security Now” and urge everyone interested in security to do the same), I think he’s a little unrelated to the current scheme of things when he’s not predicts future IPv6.

The use of IPv6 has already expanded beyond the scope of academic use and pays a lot of attention to the defense research industry, where it is no longer the question of “if”, but “when” it will replace IPv4 (in military networks, that is).

Of course, on the Internet, since we will be using v4 for a very long time, one of the reasons is that most of the hardware and software there is simply not compatible with IPv6.

I can’t come up with any flaws for implementing IPv6 now (besides the extra coding time and the extra complexity of saving two IP stacks, of course). In fact, with IPv6, things can be much simpler. For example, you may forget about NAT traversal issues.

Answer 3

Security researcher Steve Gibson said an amazing amount about this, noting that in the near future it will not appear. Here are some links that give a little more information:
http://www.grc.com/sn/sn-199.htm - Midway, he starts talking about IPv6 in the transcript / podcast.
http://www.grc.com/sn/sn-200.htm - Quick error correction in episode 199 and a bit more detailed information.

Answer 4

Your infrastructure should support both options, if possible, but if you have one choice, I would go with IPv4, but make sure you can add / switch to IPv6. But it will also depend on who the end user of the product is and what are the requirements for your project. Is this for a user who uses / will use IPv6, or are you going to accept into an existing database that mainly uses IPv4.

Based on my experience, the project I'm currently working on is considering migrating to IPv6, as there is a mandate to support IPv6. For the most part, we found out that hardware vendors (switches, routers, computers) and WAN manufacturers were able to support the transition to IPv6. There will be some work to integrate existing IPv4 equipment, but this can be done. Based on cost, it will be about the same, at least hardware.

However, some software products that we use cannot and cannot move to support IPv6, while others seem to be more convenient to support. Based on this, we upgrade to new versions of the software, but we will probably stay on IPv4 for this coin, or at least until the software catches IPv6 support.

Answer 5

I would at least start trying. I have already made some adjustments to some of my code. It has not been fully tested and more adjustments remain to be made, but even if it will never be used in these specific applications, I will have to support it at some point. Could also study now in front of him the situation with the sink or swimming.

Answer 6

The most recent news I've heard is that with NPA ... IPv6 really is not moving forward very quickly, if ever. (Steve Gibson - GRC)

It may not be worth your time.

EDIT ... obviously, it was a bad foresight that I should not have bought. IPv6 is coming and will be needed. IPv4 will probably always exist, but applications should be able to handle both in the near future.