aefstoggaflmOpen Source FanPremiumReviews:
·Verizon Online DSL
|reply to skeechan |
Re: IPv6 is not needed
said by skeechan:I point to grc.com -> Services -> Security Now! -> Episode #371 | 26 Sep 2012 | 99 min. (Listener Feedback #151)
Enough not to worry about it. The crying "Wolf" is getting old. There is no crisis. IP4 offers enough address space for an IP for nearly every man, woman and child on the face of the planet.
End the hoarding of IP space, NAT addresses the rest.
and quote the part that is related to what you just said.
Leo: All right. Here's a complicated one from Tom Ribbens in Belgium. And just as a setup, last time we talked about the fact that the British, like, retirement system or something has an entire /8 block of addresses, which they apparently don't use publicly. Now...
Steve: More than one 256th of the entire Internet's address space.
Leo: And of course, as you all know, with the current system, the IPv4 system, we just are - we've run out, in fact. We are out of addresses. So we'll have to move to IPv6. But in the meantime, a significant number, millions of addresses, are being kind of, well, it seems, misused by the British trust: Steve, I thought your discussion last week about the 220.127.116.11/8 - this address space we talked about - was completely off. You said, "Well, all they have to do is change the 51 to 10," which is of course an internal designation.
This is not as simple as you might think. This would take weeks of planning and preparation and will cause issues along the way. And what for? When the IANA still had blocks of IPs to give to the RIRs, I believe they were crunching through them at the rate of two /8s a month. That would mean, even if we got that /8 back, it would only move the problem away for a couple of weeks. Seeing that there is no way any sizeable organization could renumber their whole network in two weeks, this is not a viable trade off. Even if we would find 10 such companies who could give back /8 blocks back - and I think there are probably that many - it would still only help for another half year.
You know just as well as I do that the real solution is IPv6, and that adoption will only happen when everybody is forced to adapt. It might cause a little mayhem when it really will be absolutely necessary, but delaying it another year is not going to help a thing because we'll hardly be better prepared, as there's almost no incentive currently to do so. Tom Ribbens. I think he's right.
Steve: Well, yes. I guess the point was that there has historically been a huge amount of waste.
Steve: Because we thought, oh, 4.3 billion IPs, we'll never use all of those. And so, early on, huge blocks were being handed out very easily. So the way I view this is sort of a - is a struggle with tension between competing interests and the need to implement IPv6. The problem is people who already have allocations of IPv4 IPs, they want to keep them. They've had them for a long time. They figure that they're entitled to do so. And they probably have a good point. They could make that case convincingly.
At the same time, we need to move to IPv6, but it is a pain. I mean, it requires the replacement and upgrading in some cases of entire networks and switches and routers within a company. And you could also argue that any company that already has IPv4 is disinclined to move away from four over to six. IPv4 addressing will never go away. I mean, probably never. It was first - it will continue to be supported. New allocations at some point will have to be IPv6. But at the same time, looking at huge blocks of unused IPv4 does create some tension because it would be easier to reuse that than it would be to make the move to IPv6.
So first of all, I mean, the one area where I disagree with Tom is the rate of consumption. It is no longer the case that /8s are being consumed at the rate of several a month. Remember a /8, as I mentioned before, is a huge chunk of the Internet. It's about a 200th of the Internet, of the entire address space. That's massive. So today, now that we know IPv4 IPs are so scarce, they are being managed far more carefully than they were in the past.
So I just sort of see this as a set of competing pressures. There is pressure to better use existing IPv4 space. There is pressure to move to IPv6. And we are running out of IPv4 space over time. Yet people who have large allocations of IPv4 that they are not using, there's some argument to be made for freeing some of that up to release some of the pressure. But, yes, ultimately, new people are going to have to be using IPv6. We'll get to a point where there will be no more IPv4 space.
Leo: Yeah, I mean, and it seems wasteful. I guess you could go back and forth on this. And there's a large camp of people that say, well, look, we've just got to have the pain, or it's not going to happen. And I'm not sure I disagree with that. Because, I don't know, I think what's going to happen is you're going to have ISP NAT. We're never going to go to v6 at home. It's going to be the ISPs who do it all. And we're actually...
Leo: Go ahead.
Steve: Well, imagine that a company said to another, a squatter, we'll pay you X amount of money for a chunk of your IPv4 space that you're not using because it's easier for us and more cost-effective for us to do that than it is to move our infrastructure to IPv6. So that may happen. I did see some dollar signs associated with the value of IPv4 space. And it was stunning.
Leo: Well, yeah. Especially as it gets more valuable as there's less of it.
Steve: Yeah, I just think that what we'll see is future, like, existing companies that have been around that have IPv4 probably get to keep it. Now, if they're offered chunks of money, where it makes sense for them to move, well, maybe they'll choose to give up some. But new allocations will probably, by virtue of the fact that there won't be a choice, will be in IPv6.
Please use the "yellow (IM) envelope" to contact me and please leave the URL intact.