RIPE Region Runs Out of IPv4 Addresses

New submitter 8-Track writes "The RIPE NCC, the Regional Internet Registry for Europe, the Middle East and parts of Central Asia, distributed the last blocks of IPv4 address space from the available pool. This means they are now distributing IPv4 address space to Local Internet Registries (LIRs) from the last /8. An ISP may receive one /22 allocation (1,024 IPv4 addresses), even if they can justify a larger allocation. This /22 allocation will only be made to LIRs if they have already received an IPv6 allocation from an upstream LIR or the RIPE NCC. Time to move to IPv6!"

Comments

The internet is full. Go away.

[plover]

Don't we already have enough people on the internet? Why do we keep encouraging more? :-)

Note: to all you humor-impaired people, the smiley face indicates this is a JOKE.

Re:The internet is full. Go away.

[jonadab]

> 1. Fuck IPv6. Let's keep the IP addresses a rare and highly desired commodity;
> 2. Charge an exorbitant fee every time a DHCP request is serviced;
> 3. Profit!

The problem with this is, IPv4 addresses are not rare. They're not anything like rare. There are approximately ten thousand times as many of them as are actually needed.

We only ran out because they were systematically over-allocated, handed out like free candy, based purely on requests, with no regard for actual need or common sense. My employer, for example, currently has more _unused_ global IPv4 address than we have employees. Our upstream provider did not even inquire how many addresses we needed or even wanted; they just handed us a block of the things. Something like 80% of the allocated global IPv4 addresses are not currently being used on the public internet.

More to the point, in excess of 99.9% of the public IPv4 addresses that *are* actually being used on the public internet, in the sense of packets actually traversing public networks to or from systems assigned those addresses, aren't *needed*, because they're being used strictly for the client side of client/server networking (mostly in the form of DNS, HTTP, and HTTPS) and would if anything be better off behind NAT (because it would reduce the risk of worms, and there's no downside for systems that are not servers and do not actually need peer-to-peer, i.e., most home systems and virtually all business desktops).

IPv6 is not a solution to this problem. If we allocate IPv6 addresses the way we have allocated IPv4 addresses, we'll run out of them in just a few more years. Then what? IPv8, with 1024-bit addresses, so we can start allocating entire /256 blocks and run out again?

The correct solution is to stop allocating public IP addresses that aren't needed. This can easily be done by charging a *small* amount (per month or per annum) for each address. Honestly, as many addresses as there are available, a dollar a month retail, marked up from less than half that in bulk, would be more than enough to charge. That way people can go ahead and get addresses they *might* actually need and not feel too bad about the expense, but it's enough to keep most people from grabbing ridiculously more addresses than they could ever possibly find a use for, as has been the case so far.

When people sign up for the internet at home, the ISP can ask, "For an extra dollar a month, do you want a public IPv4 address for peer-to-peer networking or to access your computer remotely from another location?" Most people will say no and can go behind NAT. Small businesses, instead of getting a /24 just because they can, can get as many addresses as they actually need for their servers plus one NAT gateway to service all the desktops. (Business desktops *need* to be behind some kind of hardware firewall or gateway anyway, for security reasons. There's no reason it can't do NAT as well -- the extra 0.002% of CPU cycles will put the electric bill up by, what, three cents a month?) Large international megacorporations, similarly, instead of nabbing a /8 for each major national subdivision of their company just because it costs almost nothing to do so, can scale down their allocation request to something more in keeping with what they might potentially actually need.

I believe this will naturally happen over the next few years (assuming IPv6 adoption goes about as far as I think it will). Nothing particular needs to be done (other than perhaps the usual anti-trust stuff in areas where competition between ISPs is artificially restricted e.g. by only one local phone company being allowed to maintain lines). The situation will sort itself out. ISPs that try to charge completely unreasonable fees for public IP addresses will go out of business, because people will just go find another ISP (assuming there's another ISP to go find -- see previous note about anti-trust issues). ISPs that charge too little (which I think would just about have to be nothing at all) will run out of addresses to allocate. Sorted.

Re:The internet is full. Go away.

[joostje]

The problem with this is, IPv4 addresses are not rare. They're not anything like rare. There are approximately ten thousand times as many of them as are actually needed.

Well, with only 32 bits of address space, that's only 4,294,967,296 possible addresses, and there are already more people on the planet. We do need more.

Re:The internet is full. Go away.

[shutdown -p now]

Fuck NAT. Don't you dare preach that ugly hack as the Right Way to solve the problem.

time to do... something

[alphatel]

I will soon run out of underwear (I have been told this since 2009). I still have not done anything about it despite holes in them. Count on my continued responsiveness to this problem.

Personally?

[kiriath]

I'm going to wait it out and skip straight from IPv4 to IPv8... IPv6 could be the Windows Vista of the IP world.

Re:Personally?

[SJHillman]

Why not just use IPv9

RFC 1606: http://tools.ietf.org/rfc/rfc1606.txt

Re:Recyle Recyle Recyle....

[Anonymous Coward]

or, you know, just use ipv6.

Re:Not unexpected

[Cimexus]

Yeah agreed. I've been on native IPv6 (dual stack, obviously) for, hmm, approaching two years now (I'm in the APNIC area so they ran out of IPv4 a while ago) and honestly I'm only reminded of the fact when someone brings IPv6 up in an article or something. The changeover was easy from the user's perspective - it just works. Indeed I suspect many users of my ISP don't even know they are on IPv6.

The resistance and heel-dragging on the changeover in many places/companies is a bit mystifying to me. It's not really that hard.

Re:spammers

[doshell]

Sigh. We've been over this countless times. Even if you managed to reclaim all IPv4 ranges that are not being completely used presently, you would buy yourself only a few more months (at current growth rates) until you ran out of addresses again.

I seriously have a hard time trying to understand why so many people on Slashdot seem to be militantly against IPv6. You'd expect more of an allegedly technologically literate audience.

All cool sites are already running IPv6.

[Mikaelk]

Like youtube, google, facebook and slashdot.
ok, all except slashdot.

Re:spammers

[Nofsck Ingcloo]

I guess the reason I'm dragging my heels is my complete mystification and annoyance that the designers of IPV6 didn't do something sensible like make some small corner of the V6 address space map to the V4 address space. So instead of being simple and seamless, I have to spend some time fooling around with my equipemnt and software to work around that omission. A pox on the designer's heads.

Re:Why aren't we on IPv6 yet?

[Midnight Thunder]

This is something the ISPs, the upstreams, well the big guys in general have to do. As an end user I couldn't care less.

As an end user you shouldn't have to care, but when the upstream guys haven't done their work and you can't access newpopularsoscialsite.com, which is IPv6 only, then you start getting annoyed and start trawling the net to see why things are broken. The problem is many of up the stream guys, at least in North America, have dropped the ball and aren't even offering options for techs who do care and are interested in being early adopters of native IPv6. Just don't get me started on some of the incompetent replies I have got from some ISPs.

As a savvy end user, for my home network, I will want to continue to use NAT or something equivalent. I don't want my printer, my desktop, my laptop and my phone that connect to the WiFi to have an externally approachable address.

If you configure your devices to only use link-local IPv6 addresses, then there is no reason they will be seen by the outside world. Even then, with a routable IPv6 address you can configure you firewall rules to only expose certain devices to the internet. In the IPv6 world the firewall will be your friend and I believe as it becomes a more important component people will work out ways of making it simpler to configure.

Re:IPv6?

[Blue Stone]

>The rapture is here!

It's the IPocalypse!

Re:spammers

[SmilingBoy]

Bullshit. I have followed IPv4 exhaustion in detail for the last 5 years. The prediction was always that IPv4 will run out at the global level between 2010 and 2013 (it happened in February 2011), and run out at the regional level in the years after that (it happened in April 2011 in Asia-Pacific and today in Europe-Middle East). So no surprises at all. If you are a European ISP, and you stuck to the rules of RIPE NCC, you now have IPv4 stocks that should satisfy your growth needs for the next three months. After that, you cannot grow your network anymore without resorting to the mess that CGN is.

Re:spammers

[SmilingBoy]

Guess what, they did: ::FFFF:111.222.111.222 is IPv6 for 111.222.111.222. But you still need to "fool around" with equipment because there is no way that an IPv4-only device can address an IPv6 device.

Re:spammers

[doshell]

I don't know where you get that IPv6 is a "full protocol rewrite" of IPv4. For the most part it does exactly the same as IPv4 except with more address bits, and in some cases it even simplifies its predecessor (e.g. no IP header checksum). Any person able to understand or implement IPv4 ought to be able to understand or implement IPv6, because there are no fundamentally new concepts. (I would venture that most people who criticise IPv6 don't even understand fully what IPv4 does, so they don't really know what they're talking about.)

I am also interested in hearing what a "simple extension of IPv4" would be, in your opinion. Odds are you will propose something to the tune of keeping the original IPv4 header and semantics, and tacking some extra address bits at the end. Except in that case you'd still have to teach every fucking router and end system in the world how to decode the new-fangled packets, which is not any different from IPv6 from a cost perspective. You might as well do it right and fix some of IPv4's warts (header checksum, autoconfiguration, node mobility, etc) instead of applying a band-aid solution.

NAT is hardly an acceptable extension of the IPv4 addressing space because NATted clients do not have the same capabilities of non-NATted clients. (Yes, I know about hole-punching techniques; they do not solve the problem fully, and in respect to what they do, they are defeated by many real-world NAT implementations.) If you don't understand the importance of this, I encourage you to read about the end-to-end principle. Finally, it is ludicrous to suggest that implementing NAT at the scale that will be required by the ever-growing Internet would be any cheaper than IPv6. Carrier-grade NAT doesn't exactly come for free.

Re:spammers

[grumbel]

how much longer would they have?

We currently use around 12 class A networks per year of which there are only 255 in total (many of which are unrelocatable due to being reserved for localhost, multicast and so on) . Whenever you hear people complaining about IBM or whoever holding a large chunk of IP addresses, that refers to a single class A network. So getting IBM, HP or Xerox to restructure their network and give back their IPs would buy you one month each time. There aren't a whole lot of companies holding class A networks, so you could at maximum get probably 2 years or so, realistically much less.

A little extra time to shake out the bugs from any infrastructure upgrade seems couldn't hurt, too.

We already had 14 years to do that, another one or two won't make a difference. IPv6 doesn't need time, it needs something that forces people to make the switch, running out of IPv4 seems to slowly building up to be that force.

Re:IPv6?

[kasperd]

And since I'm running a small webserver from home, which I presume will remain IP4 indefinitely, what's the easiest way to tell if somebody with an IP6 address can access it?

If you don't know the answer to that question, then the answer is no. You need to actually update your DNS records for the domain to include an AAAA record with your IPv6 address. Without that record an IPv6 only client will have no way of even trying to reach your domain. So, you need to get an IPv6 address, and then put that IPv6 record in DNS. If your Internet provider doesn't provide IPv6, then you have three options. Use a tunnel (tunnelbroker.net is the one I have the best experience with), switch to a better Internet provider, or wait for your current Internet provider to catch up.

Users who have both IPv4 and IPv6 will be able to reach your website, even if your website only has IPv4. The users, which will experience problems, are those who have only IPv6. There is still a couple of ways ISPs can make those users reach your site, but they involve NAT, which will reduce the reliability. Those NAT solutions come in two flavours. There are the CGN solutions, which are just doing IPv4 and work similar to the typical NAT people have at home, just at a larger scale. The other option is NAT64, where the NAT translates between IPv6 and IPv4.