Consequences of switchover to IPv6?

Over the past year I've heard many things about IPv4, and mainly heard that the available addresses are practically exhausted as of 3 days ago!

Everything I know about Layer 3 of the OSI model (or Internet Layer for TCP/IP Model) is based on IPv4. I've never assigned IPv6 addresses nor do I know how to assign them to gateways, end devices, etc.

I know very little about IPv6, and it makes me wonder if the majority of us even realize that we've almost ran out of IP addresses for the Internet.

Are most current devices able to support IPv6? That's the biggest concern I have with the switchover to IPv6.

I learned about this once... doesn't this just mean the IP address will become longer?

Originally Posted by Maverick Nikohesus

I learned about this once... doesn't this just mean the IP address will become longer?

Yes, but it doesn't only mean they will become longer. Obviously, they will also become more complex as a result of an increased size of bits.

IPv4 address: 192.168.1.1

In Binary:

(11000000.10101000.00000001.00000001)

IPv6 address: 21DA:00D3:0000:2F3B:02AA:00FF:FE28:9C5A

In Binary:

(0010000111011010000000001101001100000000000000000 01011110011101100000010101010100000000011111111111 11110001010001001110001011010)

//disregard the space between the two zeros, the string of numbers should be together

As you can see, you have many more combinations with IPv6. We need IPv6. I'm concerned as to whether or not the majority of devices are able to handle the switch to IPv6.

My assumption is that a lot of this is going to be carried out at a hardware level by ISPs, and some sort of routing/emulation technology will be used so that old software doesn't conflict. It's pretty much impossible to implement IPv6 without totally breaking tons of existing software. However, I have been unable to find anything really concrete on the subject.

IPv6 isn't much different from 4. You, IIRC, no longer need subnets, but other than that it's pretty similar.

Alas, most ISPs are still unprepared to roll out IPv6. I know mine isn't. So the switch is still years away, at best, so we'll have no free IP addresses for quite a while.

As to backwards-compatibility: There sort-of-is some. Since old software sends all network data to the OS, where packets are actually formed, the OS can create v6 packets for software originally written to use v4 without much trouble. That's one of the benefits of using the OSI model. The trouble is, of course, if software has hard-coded IPs (rather than domain names) that it needs to connect to. In that case I don't know that anything can be done.

The big problem is that all the hardware in a connection, endpoints and midpoints, needs to support IPv6 before anything on the connection can use it. So to get a v6 connection to Facebook your OS needs to support v6 (they all do now), your home router needs to support it (most don't), your ISP's routers need to support it (most don't) the internet backbone routers need to support it (some do), Facebook's ISP's routers need to support it, Facebook's routers need to support it, and Facebook's webservers need to support it.

If even one of those doesn't, then you can't connect using IPv6. Right now, a small handful of ISPs support it, some backbone routers do, and a tiny number of websites. So it's functionally impossible to use v6 for almost anything, because techs refused to upgrade for 30 years despite knowing full-well that this was coming.

Originally Posted by John

If even one of those doesn't, then you can't connect using IPv6. Right now, a small handful of ISPs support it, some backbone routers do, and a tiny number of websites. So it's functionally impossible to use v6 for almost anything, because techs refused to upgrade for 30 years despite knowing full-well that this was coming.

For the most part, I would agree and say that techs have been lazy for 3 decades.

But I will say that no one could have predicted our love for iPhones, Android devices, and other mobile devices that use up our IP addresses.

When IPv4 was developed, I'm sure that we knew it would probably run out. We just didn't know how fast they would be used up and what would make them run out.

I suppose if someone from a long time ago really looked into a theory like Moore's Law, then we would have had an idea of what the scale of our technology would be today.

No, IPv4 was developed at a time when it was thought that there would be, at most, millions of computers on the internet, not the billions we have now.

Further, IPv6 was developed as soon as it became clear that everyone would have at least one computer on the 'net and that IPv4 was not future-proof. But no one upgraded because v6 couldn't be made compatible with v4 and because running out of IPs was always a decade or two away. Until suddenly it wasn't.

What should've been done was an install of parallel v6 connections over time. Nothing huge, but say 5% of a network's capacity should've been mirrored every year, sort of thing. Then the switch could've been thrown a decade ago, when we still had IPs to spare.

It's much like the Y2K bug. It was known that it was poor programming and that it would cause problems, but since those problems were a few decades away, no one cared.