Forgot your password?
typodupeerror
Upgrades

IPv6: Japan Leads 115

Posted by michael
from the IP-address-in-every-pot dept.
Incongruity writes: "ZDNet, in an interactive week article examines the progress towards acceptance of the IP version 6. The Japanese government has set a deadline for its information technology sectors to run on IPv6 by 2005. Other than that deadline set by the Japanese government, acceptance and implementation has, according to the article, been less than full steam ahead. This despite the fact that IPv6 have been available for allocation since mid-1999."
This discussion has been archived. No new comments can be posted.

IPv6: Japan Leads

Comments Filter:
  • by Anonymous Coward
    In Denmark and I guess in rest of Europe it is getting harder and harder to get a static IPv4 address at your ISP. All xDSL "allways on" connections are put behind NAT. and NAT is not a good thing if you like end to end connectivity.

    As of IPv6 I can run it through a tunnel, but his requires a static IPv4 address, so IPv6 for end users is first realistic when your ISP upgrades.

    Right now I am behind a 1:1 NAT at home but this will change soon accoring to my ISP. They will provide me with more local adresses (so you can add your toaster) at the cost of a static address.
    So goodbye IPv6 tunnel, and services at home.

    /Andreas Bach Aaen
  • by Anonymous Coward
    If you want to test IPv6 functionality easily, look into using 6to4 [kfu.com]. Every IPv4 address has around 2^80 IPv6 addresses associated with it (I can't recall the split into networks). That page gives instructions for BSDs, and some Linux instructions are available from Debian [debian.org]. I believe MS has instructions somewhere as well; check Google [google.com].

    It's nigh trivial to set up. However, the public gateways listed aren't terribly reliable. Don't plan on running useful servers behind a public 6to4 gateway. It is very useful for testing programs.

  • by Anonymous Coward
    Connections need not be encrypted. However, IPv6 does support IPSEC more cleanly. There is a major remaining IPSEC hurdle: the public key infrastructure. Deploying the necessary infrastructure is very difficult. Not only is there a good deal of technical work remaining, but allocating the top-level trust is also highlycontroversial. Do you trust VeriSign? I sure don't, and they're likely to be the center of trust for most practical purposes. And IPSEC is not set up for simple, peer-to-peer trust relationships. Those are difficult to maintain.

    The protocols in IPSEC are insanely complicated, as well. There will be security-destroying bugs for quite some time. Plus, most users will hose it. How often do users check the certificate authority of certs presented through web browsers? If users have to make decisions on trust all the time, they'll make trivial ones.

    So the encryption aspects will likely come later, and it won't be completely transparent in many situations. Having a future path to secure communications is great, but IPv6 doesn't translate into a huge security benefit over SSH right now. In a tightly controlled environment and in network cores, you can use IPSEC now, but many people believe the network edges will consist primarily of ad-hoc networks. Those induce really strange trust relationships, not all of which have been fully explored.

  • Hey! That's *exactly* what cisco want! A reason to sell router upgrades to customers!

    -Dom
  • A portable class C isn't worth the hassle that comes with trying to actually use it. Even if you find an ISP to route it (which shouldn't be hard), the problem is that several large network operators refuse to accept rouutes for networks that small. Verio is the best example of this. So you end up being unreachable from portions of the internet, which sort of defeats the purpose of being on the internet.

    The other problem with portable addresses is that is means a mess in the routing tables. Getting a block from your ISP means that they can aggregate your route with the routes of their other customers and then they need only advertise one summary route for a large group of networks.

    One of the things they got right when they designed IPv6 was to emphasize that small networks are connected to larger networks, which are connected to very large networks, which in turn interconnect to the other very large networks. The IP addressing scheme should reflect that and emphasize the need for the IP addreses to match the network topology (small IP block fits into a larger block upstream, and son on). This allows for easy summarization of routes.

    The only exception to this rule is for people or organizations that need multiple connections to different providers and even then there are ways to mitigate the need to advertise multiple routes (Cisco has an excellent white paper on this issues).

    The last company I worked for had a portable /16 range and I thought it was the coolest thing to have a "B" with only 2000 machines. Now I know better.
  • You are more likely to be pitched about Voice over IP than IPv6 from a vendor salesperson.

    I'm responsible for product support of a major networking vendor, across Europe, Middle East, Africa, & India (EMAI).

    plcurechax is correct--while I'm wholly on the post-sales side of the vendor equation, all our future plans revolve around VOIP solutions, with nary a mention of IPv6. I've plugged I2, IPv6, Linux support for our client software, etc. to those in engineering who would listen, but ultimately the market (and by extention, our products) are driven by what the customer requests. NAT, and the multitude of other options to alleviate the address allocation crunch, make IPv6's benefits secondary concerns to QOS, price per port, VOIP, redundancy, etc., etc.

    End result? Don't expect to see IPv6 deployed in EMAI or the US in the immediate future. It's simply not on customer's radar. Not to mention most network admins are so poor in knowledge about networking fundamentals, that the leap to IPv6 won't happen for a long time yet.
  • by jjr (6873)
    How much time you give people there will always be people who will not be ready for the change. I say that the US government follows Japan's example. because it will increase it jobs and the economy. But hey let us see what what happens. In the next few years.
  • It's almost impossible to get your own IP range these days. Almost everyone leases them off an upstream ISP.

    --
  • A Net Engineer friend of mine claims that Cisco are reluctant to support IPV6 because the amount of memory required to hold the routing tables for IPV6 is huge. Until memory prices come down it won't be worthwhile implementing it in routers (especially since there is little demand, chicken and egg problem).

    --
  • You are wrong, most IPv6 address space is not portable.

    Multi homing is one of the problems with IPv6.
    IPv6 is designed to make it much easier to renumber than IPv4 though.
    And IPv6 hosts may have two ip addresses from separate providers providing multi homing that way (I don't think this is exactly how it's supposed to work but it's something like this, portable address space won't be used for small blocks).

  • by zyklone (8959) on Tuesday May 22, 2001 @12:53AM (#206634) Homepage
    It's fairly easy to see that they will run out in a few years.
    This document [iana.org] lists the current allocations. There are not too many /8s left unallocated.
    There are a few allocated to large corporations that probably don't need that many addresses though.

    RIPE (Europe) were just allocated another two /8s so they must have a need.
  • I think Microsoft is waiting for when will the various domain registrars (e.g., Network Solutions) start supporting IPv6 addresses on a large scale.

    Once that happens, don't be surprised that Microsoft will offer an update for Windows 98/ME/2000/XP that will change the network support to include IPv6 addresses.
  • Unless ARIN has changed policy very recently, they charge an arm, a leg, and your neck to get an address block.
  • my view is that they are only charging a lot now, so vendors like cisco can get off their asses and build a new series of backbone router.....

    Perhaps Linux 2.6 will be powerful enouf that we will be able to have backbone routers running Linux.

    Then.... Trully, Tux will rule the world.
  • I think you hit the nail on the head there. IPv6 won't be a reality until Microsoft's implementation is no longer experimental, and is actually usable.

    Sad, but true.

    domc
  • Folks are working on the multihoming issues now, and it's possible they may come up with a method that doesn't have the scaling problems inherent in the current method of IPv4 multihoming (advertising the same prefix through multiple uplinks).

    There is an IETF working group with a charter for this: Site Multihoming in IPv6 (multi6) [ietf.org]

    cjs

  • by cjs (12969) <cjs@cynic.net> on Tuesday May 22, 2001 @12:18AM (#206640) Homepage
    In Asia, the situation is pretty bad, and has been for a while. It's extremely difficult to get more than a handful of IP addresses from your ISP, and NAT is more common than in the US. This is one of the reasons why folks in Japan are further ahead with IPv6.

    IIJ has been offering IPv6 service (not tunnelled over IPv4) for a while, and some vendors in the US (such as Panix in NYC, I believe) are also starting to offer this.
  • I agree, the question shouldn't be "Why change" but "Why NOT change?"

    To that end, at the NOC of the Academic Insitution I work for as a net/sys admin, we just made it an informal requirement that anything new being setup (either a new service, or upgrading of an existing one) should be IPv6 capable. Simple as that. Sure, it does restrict your choices a bit, but the impact was minimal to us since we use BSD for the majority of our services.

    It's been a few months now, and *all* the basic services that we maintain (primary & secondary DNS & MX, http/ftp proxy, a cluster of mailbox hosts hidden behind a POP3/IMAP4 redirector, a large FTP archive and all our web pages) are IPv6 capable. I really like the fact that in all our hosts, all the services are binded to both IPv6 and IPv4 sockets and have both IPv6 and IPv4 addresses pointing to them via DNS.

    The result is that, since I use FreeBSD at my workstation, like many other colleagues, we only use IPv4 for connections outside our network.

    Granted, we're currently using an extra router and tunnels for IPv6, but it's only a matter of time until we upgrade our border router to handle IPv6 and get rid of the tunnel and speak IPv6 with the backbone we peer with.

    I believe that the situation is similar in other countries too - once again it is the Academia that will lead the way, just like it did with IPv4. This is nor surprising. If you ask me *WHY* we converted to IPv6, I cannot give you an answer. Really, there's no answer. We just *DID*. This is not the kind of answer that management of a corporate entity likes to hear from their engineers, especially when it restricts choices somewhat and requires extra work to iron out bugs and problems, and all that for apparently no reason (as far as THEY are concerned).

    I also get the impression that the shortage of IPv4 addresses and the difficulty one faces when seeking an allocation, is a status that many corporate entities actually *LIKE*.

  • Everyone will get ONE address block in IPv6 now, instead of a zillion routeable pieces of IPv4. The problem is, the definition of everyone is now much larger. If more people (and companies) have a portablely routeable address space, they're gonna want to be routed to. And that means you (owner of an IPv6 portable routed block) are probably gonna "own" about 20 bytes in every core and border router. How many of these blocks do you think there will be?

  • This is exactly why IPv6 currently sucks. There's almost no benefit to it unless you can get portable space. And the allocation process for IPv6 is even more difficult than for IPv4. Sure you can get a lot more numbers ... if you can get anything at all. The problem is you can't even get portable address space.

    I'd like to try out IPv6, probably using tunneling for now. But I want to get the address space NOW that I will keep FOR ALL TIME. They are not letting that happen. And that is what I think will be the biggest roadblock to IPv6 acceptance.

  • So can I get my portable life-time IPv6 allocation from NTT?

  • Let's see. How about a bargain basement price of US$0.01 per address. A small block of IPv6 has 4294967296 addresses. That's $42,949,672.96 Quite a killing there. Too bad it's IPv6 itself that's going to be killed.

    All I need is a block of about 256 addresses in IPv6. Why is that so f***ing hard for the allocators to do? They need to stop thinking in terms of IPv4 to allocate IPv6 space.

  • Lack of portability isn't inherint in the design of IPv6 ... it's a function of the backwards thinking by bureacrats left over from IPv4. Most businesses only want portable space (at least once they understand the issues).

  • I want portable permanent IP space w/o an archaic routing system. IPv6's routing does not appear to be the solution.

  • The proper way to route should never have a big fat routing table. Apparently IPv6 didn't solve the classic routing problems that IPv4 has, probably because IPv4 was hitting other limitations first. If we're going to have non-portable address space to limit the size of routing tables, then what's the point of even going to IPv6 at all?

  • Well, they can't really make it free, but it could be very low cost, charging for the administrative cost, not the amount of space.

  • by Petrus (17053) on Tuesday May 22, 2001 @01:18AM (#206650)
    The IPv4 shortage has many dire implications. I would hope that I have a right to have my personal mail server and my personal web server and ftp server. I feel quite uncomfortable with my personal stuff being kept anywhere outside my locked house. With current IPv4 is is not always possible. Assingning dynamic IP became the norm and static IP are either unbearably expensive, or even prohibited in residential areas.

    Owners of the static IP ranges seem to be the king of internet universe, that can dictate price, conditions and force you to run your server off their premisses (for a fee).

    Can somebody post details, how bad can be the censorship implications ov IPv6? I think, that the contents tags ccould be actually bogus, so that contents-based censorship might become ineffective.

    How difficult would it be to stop a packets on the border? How many paths out of the country are there?
  • Does anyone know which operating system support IPv6, or have patches to provide IPv6 support? This is an important factor, along with software expects a non-IPv6 IP address. Unless the OS support and application support is there, I can expect a lot of problems.
  • And you know what?

    I want to kill myself.

    I remember about ~8 years ago, i was reading about network connectivity and stuff, and it said "do not just pick IP numbers out of thin air. Email xx@xx to request your own IP block". (it was email, the web didn't really exist back then, so there was no website to go to for IPs)

    I could have actually gotten my own Class C or whatever, free, back then. :(((

    *sighs*.

    I'd kill for that now, i really would.

    Will IPv6 ip's be given out free? How much are they in the Australia region?

    I just wanna get a block now, i wanna get in early on things now :) Never wanna miss another oppurtunity like that again!

    BTW, IPv6 network connectivity works *perfectly* between FreeBSD, OpenBSD (and Linux, according to a friend that uses it). I haven't got it to work in NT4 or Win2k yet, but i haven't tried IPv6 in NT for a few years now. (The Microsoft Research website has an 'experimental' research IPv6 stack)

    D.
  • That is sad :(

    Fortunately, MS said Windows Whistler/XP/NT6.0(NT5.1?) will contain full IPv6 capabilities.

    So we might finally make some progress with IPv6 adoption....

    I still wanna know where i can get public static IPv6 ips.

    D.
  • he end user needs only to have v4 nat happen - and have the v4 to v6 translation happen upstream. so - the end user has a 10.x private - which goes upstream to his isp, the isp has v6 peering relationships and has a block of legal v4 classes assigned to them. keep v6 out at the core backbone level for as long as possible - but each tier 1-3 has a certain v4 and v6 blocks that they own - and dole them out as needed v4 first.

    This sounds like the "end user" would not be able to have a "real" IP address for running things such as a Web server...

  • by Grit (18830) on Tuesday May 22, 2001 @06:36AM (#206655) Homepage
    You shouldn't believe everything the IPv6 people tell you. Sure, they _claim_ they will reduce the size of routing tables, but only by renumbering fairly often--- a scheme that has not been demonstrated on a large scale. (How often? Nobody knows.) Most of the recent growth in the size of routing tables has been from increased multihoming--- which IPv6 does not yet provide a good solution for.

    IPv6 requires you to have a distinct range of IP addresses from each of your upstream ISPs. The addressing/routing architecture does not allow these ISPs to advertise your "other" prefix to their backbone providers (or, possibly, to their peers.) This negates much of the benefit of multihoming, since any particular address is tied to one ISPs--- and possibly to one ISP and one of that ISP's providers.

    As far as I understand it, the current wisdom on IPv6 multihoming is to use tunnels between the various ISPs you have addresses for; this doesn't completely solve the problem, since you still have a dependence on the ISP which "owns" that particular address. And tunnelling, of course, adds extra overhead and an additional routing table entry in the ISP's routing tables.

    IPv6 doesnt "solve" current problems with routing, it just attempts to legislate them out of existence. And yes, I _do_ subscribe to the IPv6-haters mailing list.
  • Customers are always going to be ignorant of the options. Sooner or later, everyone will hype it up and demand it. Look at what's happening these days with wireless. Everybody seems to want it but few ISPs can explain why it makes sense.

    Not to mention most network admins are so poor in knowledge about networking fundamentals, that the leap to IPv6 won't happen for a long time yet.

    MOST. Some admins are actually quite knowledgeable. Kids, study up on your IPv6 NOW and you'll have a big advantage.

  • Cisco is fixing to get set up the bomb if they aren't ready to support IPv6 very soon. IPV6 is already here. Some other company could steal Cisco's #1 spot by getting industrial strength IPv6 hardware on the market before Cisco can react.
  • Yup. Supporting IPv6 is a good thing. IIJ is setting a good example by offering IPv6. I checked the American branch [iij-america.com] of IIJ to see if they offer it in the states, but it seems that they don't. Their rates are hideously expensive, too. Oh well.
  • Try the KAME project [kame.net].

    Here's one guy's experience [echidna.id.au] setting up a tunnel to the 6bone with OpenBSD. By doing it this way you get a connection the IPv6 backbone and you can run IPv6 in your local network without needing IPv6 services from your ISP.

    Note that KAME is for BSD. If you really want Linux, try USAGI [linux-ipv6.org].

  • Wow. I got a reply from IIJ-America within 30 minutes.
    The prices are out of my league for a simple home ADSL hookup, but I'm pretty impressed with their response time.

    > First of all, thank you very much for requesting the
    > DSL information. For your location, we can provide SDSL(1Mbps/1Mbps).
    > For the price is below, installation(Including Router):
    > 1yr. $1020, 2yr. $660, or 3yr. $480
    > Monthly charge = $444.
    >
    > Regarding IPv6, please Contact us either phone at
    > XXX-XXX-XXXX or e-mail at info@XXXXX.com. Thank you.
    > Thank you for contacting us. Sincerely,
    > ===================Shigeharu Miyazaki

    Shortly after getting that message, a rolling blackout in California took out an m-l.net router and half of the 'net vanished for about an hour for me. Doh!
  • by ianezz (31449) on Tuesday May 22, 2001 @12:56AM (#206662) Homepage
    If you read the documentation about IPV6, its adoption should greatly reduce the size of routing tables. So, perhaps, it's the case of researching the thing a little more.

    AFAIK (from reading the IPV6 docs), it's the current inefficient allocation of IPV4 networks/addresses that leads us to large routing tables.

  • by akb (39826)
    The Napster of IPv6 is the fact that its multicast native. Multicast will let anyone be able to stream live multimedia to an unlimited number of end users. In my opinion this is the most important feature of ipv6.
  • As a consumer, I have to agree with you. When the end user no longer has the ability to have permanent reachability, the most important consequence of the Internet is in danger of being stifled; it's fundamental freedom.

    Users will be limited by the courage of hosting companies and the like. If I could (and I can't) get my home cable modem to run "lronhubbardisanalienslugmonster.com", I have the choice to criticize Scientology with that site. If the end user loses all hope of running their own services, then his freedom of speech will be limited by the most cowardly tendencies of hosting providers. Great, cable companies and Geocities will be the arbiters of content. Blech.

    How does IPv6 fit into this? It's critical! Until the core internet becomes completely IPv6, the holders of addresses currently - ISPs, generally speaking - hold the limiting property for the medium. I'm guessing that as addresses become scarcer, and therefore more valuable, the ISPs find LESS incentive to upgrade.

    It also looks like a truly portable address under IPv6 - say, tacocellphone.slashdot.org - has to rely on dynamic DNS with VERY low refresh...

    Now let's look at the home user in the future. People on mass broadband - the type with dynamic addresses, or the type not meant for "real" use - your basic peon connectivity - might be the first to be stuffed behind IPv6. Their ISP maintains external v4, but of course you can't really be reached at home from pure non-upgraded v4 customers. If this happens, then some whole new layer of peer-to-peer services become critical.

    But I can't see how Junior can run a quake server under this scenario, so we've got problems. On the other hand, I'm sure Time Warner would love for the net to become a passive medium, but for the sake of the argument, let's assume that they can't go v6 like this. Now we're stuck with v4 addresses becoming like broadcast licenses. Increasing censorship, high cost prevents newcomers, amateurs, hobbyists from participating, so the internet, while it has more "channels" than cable ever will not die, it will just become more and more boring, as the massive amount of content becomes more and more scrutinized.

    The only way out that I see, of course, is smaller ISPs - how are they going to get you connected? Some kind of high-speed wireless, large cities only, I'm guessing. But the point is, the transition path might be that as v4 begins to suck, some customers will jump ship to v6 ISPs. They will accept becoming client-only for v4 net in exchange for greater freedom - v6 ISPs won't be tracking your P2P actions and snitching the way TimeWarner probably will, eventually. They won't care all that much what people do, it will just be a rebirth of mom-and-pop ISPs. The situation will be alleviated somewhat by application-aware routers that take a v4 address, look at the application layer - Host: headers, for example, and translate into v6 addresses. Lots more "port 80 tunneling" in that future. But eventually, the freedom to occupy space (all the addresses you can eat), crazy hobbyist content, special interest IPv6 ISPs, etc

    So what happens? My guess is that Japan will have the first large-scale version of the v6 ISPs. They will figure "whatever, v4 internet is mostly english. If we all switch to v6, we can access Japanese content, good enough." Their government won't be terrorized, as ours is, by claims of too much government interference, so they will create incentives. The US may stay IPv4 for a long time, trying to use the v4 address privilege to maintain an aristocracy of content production.

    Of course, all of this supposes a migration of the hip to v6 to create enough "cool" for the scenario to go to completion.

    Boss of nothin. Big deal.
    Son, go get daddy's hard plastic eyes.

  • We've been hearing stories for a while now (3 years? longer?) that IPv4 addresses in certain ranges will be running out. Has anyone actually had any problems getting one. Does anyone have a public IPv6 address yet.

    No-one in the States, no, because the States has grabbed more than half of the world total. Plenty of people in East Asia and Africa, because they came late to the table and got hardly any. There are more people in China alone than there are addressable IPv4 addresses.

  • A Net Engineer friend of mine claims that Cisco are reluctant to support IPV6 because the amount of memory required to hold the routing tables for IPV6 is huge...
    I was wondering that myself, but I had heard that proccessing power was a large part as well.

    In Japan last year I saw native IPv6 routers on sale from lots of different makers. Yamaha [ohnolab.org] had them from the equivalent of US$400 upwards. The interesting thing about this is that the prototype was built for Yamaha by some students at a technical university as part of a (?) Masters course.

    CISCO aren't making them because they don't want to, not because they can't - and if they don't move fast they'll lose this market to the Japanese.

  • Or perhaps IPv6 integrated into the current desktop OSes. Wait, isn't it already (Windows 2000, Linux, OS X)?
  • Maybe IPv6 isn't widely deployed because of the lack of compatible applications. Many programmers don't make a step to have their apps work with IPv6.
    I've written a french IPv6 programming HOWTO [claranet.fr] to help these people port IPv4-only apps to IPv6.
    IPv6 is something really worth to look at.
  • I have been wanting to try it too - but simply couldn't get past the setup stage with the howtos (I am probably missing some fundimental knowledge here, but if a fulltime Firewall/LAN technician for a multinational has trouble setting it up, what chance does a normal user have?)
    --
  • Part of the handback problem is the policies have changed.
    Assume you had a portable 'B' which you "own" from the early days.

    if you hand back part of that, then you make routing difficulties for yourself. that is why they recommend you hand back the whole block, and accept a replacement (smaller) block.
    The problem is, even after you have pushed though the renumbering, got everything working, and are happy.. the rules have changed. The new block you get will not be portable, and you will not own it - you will be allocated it which makes a difference. For a large company, it does not make sense to do the "right thing" and hand back an address range you are using less than half of, only to find you are given back something less flexable, with routing and multihoming issues, and expected to go cap in hand back to them if you need another class C in the future (and are probably turned down as you already have enough if you NATted them into your existing range)

    under V6, things are worse - you have no rights at all in your IP range, to the point you can be asked to renumber into another range at any time if it makes routing easier. even leaving aside the chaos that will cause in the DNS, for a large organisation the renumbering alone could work out very expensive indeed... so I imagine most will try to hold onto their legacy V4 subnets until they are forced to give them up.
    --

  • by DaveHowe (51510) on Tuesday May 22, 2001 @03:18AM (#206671)
    Just how static do IP addresses have to be?
    Very. It can take some hours for DNS changes to trickle down to distant parts of the net, and until you can resolve the new address, the website is "broken" for your customers.

    Why would anyone want an IP address space which is not a subspace of the provider's address space?
    Two reasons - portability and multihoming.
    Multihoming is where you sign up with two or more providers, so that if one has network problems or goes under financially, you are not out in the cold
    Portability means you can get a better price from your isp. Consider the following two possiblilities;

    1. Moving to another ISP for cheaper prices means just moving your IP allocation to another ISP
    2. Moving to another ISP for cheaper prices means renumbering your entire externally visible IP range, updating (and moving) your DNS servers, and waiting for the changes to trickle down (with loss of connectivity for your customers)
    Which of these two customers does the ISP salesforce stick that extra 2% price increase on this year?
    --
  • by DaveHowe (51510) on Tuesday May 22, 2001 @03:54AM (#206672)
    The new DNS may well happen - one of the failings of the current system is that it does not support non american-english characters; while from certain points of view this is fine (after all, if you can't type an URL on your machine, how many hits will they get?) support for the japanese charset in email and webpages has been standard in IE/OE for some time. The most obvious solution to this (encoding DNS names in non-US as the unicode multi-char representation, as web pages can do has been *PATENTED* in the us. I am sure I don't have to start the usual stupid-us-patents thread again though...
    --
  • by DaveHowe (51510) on Tuesday May 22, 2001 @03:09AM (#206673)
    Yes, of course he would.

    It is common practice for companies to hide an entire RFC1918 subnet behind a small number (8 or 16) of internet addresses. One or more of those will be allocated to internal addresses (so if your webserver (say) is 192.168.1.2 but your external webserver address is 200.100.50.5, then packets both ways will be rewritten to hide the internal address behind the externally visible one)

    Given how large the available IP address range is for V6 (the *minimum* allocation would be a class B by the old standards) There is no reason you can't have a 1:1 mapping from IPV6 external addresses to internal V4 addresses; further, you probably will want to static-map the lower two bytes of your 1918 to that address range rather than the recommended (which is the MAC of the card) due to the fact that swapping out a faulty network card would then force-renumber your webserver to a different V6 IP address.....

    I fully expect to see Hybrid mode firewalls in the near future, which in addition to mapping the small number of externally visible V4 addresses to Internal hosts, also map V6 (autotunnelling to the ISP) for both internal hosts and outbound browsing traffic.
    --

  • I would say that dynamic addresses do provide a significant increase in anonymity over static. True, "The authorities" can unravel a dynamic assignment iff they work quickly enough that the logs haven't been rotated into oblivion. And they'll need court orders and such. Nothing is automatic, so this will only happen occasionally.

    Contrast this with IPv6 where even "dynamic" IP assignments (as you point out) are very likely to have a static component -- some bits to identify your userid. Mask out the appropriate bits, and anybody will be able to track you. Employers, insurers, ex-spouses, marketers, etc.

  • Nothing against the Japanese, but if they want to lead on IPv6, let them! Although it doesn't seem like even they are picking it up all that fast.


    Myself I don't much like IPv6. 'Way too much overhead with 128 bit addresses. That's 24 extra bytes per packet, ~5%. Also a significant reduction in anonymity (fixed IPs vs current dynamic IPs).


    I'm also not convinced that IPv6 will solve real (vs imagined) problems or bring compelling new features. The current IPv4 routers seem to be able to keep up, and if they have trouble, they should drop straggling routes (addrs away from their heirarchy). Most of the current Inet problems are more related to poor software (DNS, SMTP). QoS sounds like a neat feature, but I doubt it will be widespread because of the difficulty of cost charging.

  • by spinkham (56603) on Tuesday May 22, 2001 @06:21AM (#206676)
    As a security dork, I feel the need to point out something you all are forgetting...
    IPsec is a part of the IPv6 standard, meaning when we all move to IPv6, all traffic will be encrypted, not just specific VPN links like we do now.. That's a HUGE benefit, at least in my eyes...
  • by Phizzy (56929) on Tuesday May 22, 2001 @05:30AM (#206677)
    I'm frankly getting sick of all of this IPv6 hype. With NAT, BGP and classless routing protocols, IPv4 still has plenty of life left in it. The change to IPv6 isn't going to happen soon, and it doesn't need to. Besides, if you really want to run IPv6 right now, just to prove that you are so much r3373r than your sys-admin buddies, go ahead and run it, and tunnel it through IPv4. It's perfectly feasible, and probably what early-adopters of IPv6 are going to have to do anyways, because as far as I know, there isn't a single backbone provider who is even seriously discussing implementing IPv6 in their network. We have loads of IPv4 space left, the IPv4 network that we're all using to post on this great site is obviously working quite well, and a load of new address space isn't going to help the internet in any really useful way. IPv6 is going to be a whole lot of work, a lot of hassles, a lot of connection problems, and with little short-term gains. Everyone always preaches not to upgrade your kernel if there isn't anything you're going to gain from it, so why upgrade your logical network addresses if it's not going to provide better service to you? IPv6 will come, but not until we need it to.

    //Phizzy
  • by Phizzy (56929) on Tuesday May 22, 2001 @07:01AM (#206678)
    Mac addresses are relatively local. When your computer sends out a packet, it wraps the data in a layer 2 header, w/ the Src and Dst MAC addresses, and then a layer 3 header, w/ dest and src IP addresses. Now, say you're sending a request to /... since your computer has no way of knowing /.'s MAC address, and no need to, it uses the MAC address for the gateway that the host has assigned to it. Once this packet goes across the ethernet to the gateway router, the router strips the layer 3 header, leaves it pretty much intact, looks at the layer 2 header, sees it's mac address, and knows it has to forward it. When it does forward it, it uses the MAC address of the next-hop host as the dest MAC address and it's outbound port's MAC address as the source, and the same thing happens at the next-hop, all the way down the line and back. SO, the lesson is the MAC address need only be unique within a broadcast domain, and the broadcast domain ends at the router. And besides, IPv6 isn't going to change the layer 2 addressing, we'll still be using (likely the same) Mac addresses.

    Plus, I don't see anywhere you can buy internet-enabled garage doors OR fridges. So all of this is pointless, just like the whining about IPv6.

    //Phizzy

  • Sadly, I've yet to see a IPv6 implementation that supports IPSEC. In fact, when I tried to use freeswan w/ipv6 the machine died :
  • Interestingly enough, the data center for NTT Communications [ntt.com] ( a subsidiary of Japan's massive telco, NTT) is ready to roll with IPv6. Apparently they are the first and only data center capable of this. A sign of the times when a slow moving behemouth like NTT can be so forward thinking. Must be the influence of DoCoMo.
  • ... But IPv6 won't be widely deployed until the consumer version of Windows supports it, and can transparently proxy for the old Windows apps that don't understand it. Until BillyBobWinUser can be assigned a IPv6 address and still play EverCrack, it's not going to happen.

    Question for the audience: does DirectPlay support IPv6? Does .Net?
  • Does anyone know which operating system support IPv6

    Beside the OSses mentioned elsewhere in this thread, Solaris has supported IPv6 for a while. And it'll happely run IPv4 and IPv6 on the same interface.

    I do not know whether all their applications support IPv6 though.

    -- Abigail

  • Take a look at my website for a description on how to have IPv4 clients be put on an IPv6 network. I just finished research on it and it's not completely finished, but the base idea is there. All you need to do is translate at a gateway. On the inside looking out it looks like IPv4 and on the outside looking in it looks like IPv6.
  • It's all well and good saying "allocations have been available since 1999" but in actual fact it's quite difficult getting an allocation. Why?

    Most of this is to do with the Local-IR requests [ripe.net] which fail (at least at RIPE [ripe.net]) because you need three separate peers before they'll even consider it.

    Then of course your upstream should be allocating from their PA block anyway. And since most upstreams aren't allocating IPv6 to end users...

    ...it's all a bit much really.

    --

  • Dynamic IP addresses don't provide you with any less anonymity than static ones. If you that they do, then you are a fool. Its pretty damn easy for the appropriate authorities to find out who you are in either case. Regardless, there is nothing that says that you have to have a static IP address with IPv6. Consider that a dialup user will be doing IPv6 over PPP, the isp is still going to assign them a dynamic address. As far as cable modem or ADSL users go, I suspect they will still end up with dynamic addresses as well, as this gives the ISP more freedom to renumber their networks and sufficently annoy their userbase into not using their connection to run a mp3 leech server..

  • I think an important factor here is that Microsoft isn't fully supporting IPv6 in its 9x or NT operating systems. I don't think we will see companies migrating over until MS gives it the green flag.

    Don
  • Dynamic IPs aren't the only problem with holding a internet services. The problem is the ISP and if they want to give the user ability to do so. I have a dynamic ip (changes after 12h of no connection) and it most certainly doesn't make me unable to create a web server and a ftp. And as a I keep the connection alive actively, it only drops on network breakdown, it isn't really a problem.

    I suppose the problem here is that ISPs want to preserve upstream as it's harder for them to control it's cacheing than downstream. So they put up firewalls to cut it down.

    You can get static IPs here and actually from some ISPs you get it if you ask for it. For the ISPs that I can get for my DSL, there is only one possibility and that would cost some 150FIM/month ($23). Additional to the $75/month (It's a 512/256 ADSL). It's not unbearable, but I consider it a waste of money.

    So for many people IPv6 won't bring atleast more of that freedom, it's all about bandwidth. But for me it would likely allow me to get a static ip for no additional charge.
  • A number of years ago, a friend of mine (Curt) got his own (personal) routable C class subnet assigned to him (it was something of an 'oh hell, why not' kind of thing. Nowadays, for a company to get a class C range takes a good bit of work. As was said -- now it's usually borrowed from their ISP.

    Just this weekend a friend of mine (John) mentioned that his Co-Location provider was charging $4/year per IP address. Not much, on the surface, but this means that the class C that Curt got permanently assigned for free a decade ago is would cost John $1K/year now.

    In 1992, the University of British Columbia department of Computer Science got it's own Class "B" range assigned (the UBC, generally, already had at least one "B" range assigned to it). This was for a network of, maybe, 400 machines. I challenge you to find me someone who's been assigned a class B in the last few years for as few as 1000 machines. In some cases, a 1000 machine network might only get one or two class 'b' blocks and be expected to NAT most of their machines through a firewall. "I mean, you don't really need all of those addresses, do you?"

    So, yeah, I do think that IP addresses are getting scarcer these days.
    --

  • Ahh...

    * 2001-05-09 19:38:06 USA lags behind in IPv6 deployment (articles,internet) (rejected)

    I tried to post this same story a few weeks ago, about how the USA is falling behind in the deployment of IPv6. Basically, the reason for this is that the USA has got the lion's share of existing IPv4 addresses, so the incentive to convert has not been as high. So, we're letting ourselves lag behind, as usual. It will be sad when everyone else is speaking IPv6 and we're still stuck behind 10.x.x.x NAT's...


    Super eurobeat from Avex and Konami unite in your DANCE!
  • The difference between the $400 Yamaha router and the one people would expect from Cisco is about three orders of magnitude in the amount of traffic they have to handle. Cisco routers are used in backbones for Gigabit connections. If they can't offer IPv6 routers able to handle that amount of traffic, it's better for them to not offer IPv6 routers at all, or they'd lose their "top dog" aura.
  • Microsoft, or *shudder* AOL. MS's "experimental" IPv6 stack, standard in Windows XP, works quite well for me. Experimental doesn't mean unusable...

    AOL though, they have the money to buy up as many IPv4 addresses they may ever need...
  • I read that list posted elsewhere, I guess Stanford recently gave back 36/8, and other /8's had been given back. But there are I believe more than 50 /8's still unallocated, and I don't think anyone but RIPE, APNIC, and ARIN can get them anymore.
  • I use IPv6 with a tunnel to the 6bone [6bone.net]. My web and email servers, as well as others, are at this moment IPv6 ready. Here is a very good site for IPv6 information: hs247.com [hs247.com].
  • Cisco is indeed pushing forward with their IPv6 support, as seen here [cisco.com].
  • you have to justify it or else cop for a virtual
    .oO0Oo.
  • CISCO announced support of IPv6 on May 14, 2001 [cisco.com]. It's a software upgrade, and will be in Cisco IOS Software release 12.2(1)T, available at the end of May. Support will be available for the Cisco 800, 1600, 1700, 2500, 2600, 3600, 4500, 4700 routers and various other devices by the end of May 2001, says the press release.

    That should push availability up considerably.

  • My only experience of obtaining static IPs is over here in the UK, where things are always expensive. However the ISP I use provide routed ADSL with 13 static IPs for GBP99 a month. It works out far cheaper than what I was paying for my old ISDN connection, which provided just the single static IP address and meant I had to implement NAT myself.

    It is possible to have v.cheap internet access but if there are certain features you require that impact the bottom line of the ISP, you have to expect to be charged a premium for them.

    I suspect that even with IPv6 ISPs will continue to prefer to offer dynamic addresses. From what I have read IPv6 address ranges will not be free and so minimising the number of required IPs will help keep down costs.

    Tim Sansom
    http://www.samoa.co.uk/whereis.html [samoa.co.uk]

  • you can find all sorts of info on IPv6, including how to connect to an IPv6 over IPv4 network that exists now at 6bone.net [6bone.net]
  • rather than making themselves incompatible with the rest of INTERNET I don't really see what are they trying to achieve. Oh yeah, shooting themselves in a foot. Of course.
  • Fortunately, MS said Windows Whistler/XP/NT6.0(NT5.1?) will contain full IPv6 capabilities.

    Hopefully they'll get it fully integrated in, like IPv4 for the final release. I'm running a beta of XP (NT5.1, not 6.0 :) right now, and to install IPv6, you run "ipv6 install" from the commandline. If you want to configure static addresses and routes, you do it from the commandline too. But it does work... I got to see the Dancing KAME [kame.net] from IE6.0 :)

    I still wanna know where i can get public static IPv6 ips.

    http://ipv6tb.he.net [he.net] runs a tunnel broker and gives out /64 blocks. I've got 3ffe:1200:3028:81e7::/64, which gives me 2^64, or 18446744073709551616 addresses :)

  • In terms of helping out the chicken-and-egg problem of routers not supporting v6 until there is demand, and there being no demand until routers support v6: 3GPP [3gpp.org] has decided that it will deploy the next-generation mobile internet on IPv6 exclusively. Hitting v4 internet sites will be done through gateways.

    If this network is sucessfully deployed (think 2002 to 2004), it should give IPv6 a huge shot in the arm.

  • We've been hearing stories for a while now (3 years? longer?) that IPv4 addresses in certain ranges will be running out. Has anyone actually had any problems getting one. Does anyone have a public IPv6 address yet.

  • I wonder, wouldn't it be very well possible for universities to start the change? It wouldn't be the first good thing they'd do, look at BSD and Sun.

    For a uni it could well be worth the effort to migrate, after all, managing your network should become easier. Furthermore it would be a nice opportunity to teach students something about networks. Sure, it could be costly if routers have to be replaced because they don't support IPv6 yet (I don't know about that), but there will be some government fundings, no doubt. And if more and more IPv6-clouds appear, the threshold for others to migrate will become smaller and smaller.

    I personally would welcome IPv6 with open arms. Not a chance here to get a decent connection to the Internet without some form of NAT, which means you can't run most services you'd like to.

  • --Current predictions place the final IPv4 addresses to run out on Thrusday.

    --IPv6 is currently illegal as people use it for Quake and Quake kills High School Students.
    ---
  • Japan jumps to IPv6, Japan create a new DNS sceme, everyone jumps to the new DNS and Japan Internet. Corparations get fair treatment (and MS gets to own .com version of the new DNS sceme and impemet thier entire .net crap) Current internet dies becosue the new internet supports 1GB bandwith per user, and static IP. US cries becouse no one is using the current internet...

    And then I wake up.
  • I've been reading through the FAQs but can't find any where to register my own set of E-Class(not C-Class since) IP addresses. Is there a particluar FAQ, RFQ, or website I should look for? Is there a particular organization I should go to?

  • I have wanted to try out IPv6 on my LAN, but not sure it will cause more problems. I know few applications can handle it, and how backwords comptiable is it ? Since all of these IPs are behind a firewall, it won't make that much of a difference.

    I think its great that they have created a deadline. I think more places in tyhe world should do the same. Its kind of everyone else is waiting for everyone to start.


    until (succeed) try { again(); }
  • Chello in Sweden provide 4 real ip addresses per customer even if the customer only needs one. RIPE must be fairly mad over this (unless of course they don't know)?
  • Cisco do have IPv6 images availible but yeap, you're correct there are no general deployment images with IPv6 support.
  • > In Denmark and I guess in rest of Europe it is getting harder and harder to get a static IPv4

    In Sweden (part of Europe :) there are serveral DSL solutions available for customers. All of those provides a real ip-address. Two of the major ones also provides static ip's. Some cable-model-companies uses DCHP, but you still get a real ip-address.

    It's mostly the universities who provides internet access through nat, since that reduces the amount of servers with illegal content on it.

  • Just how static do IP addresses have to be? Why would anyone want an IP address space which is not a subspace of the provider's address space? When the finer routing decisions are kept at provider level, the routing tables for the big pipes can be made lean and fast.
  • That's the conflict between IPs seen as routing tools (non-portable) and IPs seen as abstract addresses (portable). Both multihoming and portability (as well as DNS-related downtime) are non-issues for almost all users who are now in the situation that they can't get a static IP address. "Static, until routing changes" is a good tradeoff between routing table size and user experience. Skapare implied that most people would get portable addresses and in that case, the price increase will go to them, for causing routing table bloat.

  • by YKnot (181580) on Tuesday May 22, 2001 @02:09AM (#206713)
    IPv6 is not the tool for giving us more NATed 10.x.x.x networks. Users will not benefit from IPv6 if it's only used as backbone technology and the endpoints of communication keep calling eachother 32bit names. What's the advantage of having bazillion addresses free for everyone if you can't enter them into your latest first person shooting game? Don't let people mislead you: The key for quick migration is not backbone providers making a start. It isn't some remote tunnel possibility either. It's IPv6 "Napster" which will do the trick.
  • No-one in the States, no, because the States has grabbed more than half of the world total.

    Did you stop to think about why that might be? :)

    You are making a bad assumption if you think it's as easy as it used to be to get address space in the US. You'd be wrong; we had to beg and plead to get a measly /27 from our ISP, after we filled our /24. And forget getting portable space.

    I think some networks in Europe use lots of IP addresses as well. Let's take Demon Internet of the UK for example. They assign static addresses to all of their dialups. They are somewhat famous for doing so, but how many addresses could they conserve by assigning dynamically like every other large dialup ISP in the world (I'm assuming, but you get the idea :)? I know they have 193.195.0.0/16, 194.222.0.0/16 and 194.217.0.0/16 at the least. That's a lot of IP addresses.

    RIPE just began allocating 80/7 (if memory serves) to European networks, as well.

    That being said, it has always bothered me greatly when there are places like MIT, who has legacy space of 18/8, yet hardly needs millions of addresses. They won't give it back! And of course I am aware of the difficulties and expense involved, but their unwillingness to play fair, and ARIN's insistence that we must "conserve, conserve, conserve!" isn't helping us poor fools who can't even multihome effectively. Sigh, this has all been said before :)

  • Follow the link in the original story where the link says 'available for allocation'.

    Or just click here. [arin.net]

  • This is from the IPV6 Policy Document:

    4.1 IPv6 Addresses not to be considered property

    All allocations and assignments of IPv6 address space are made on the basis that the holder of the address space is not to be considered the "owner" of the address space, and that all such allocations and assignments always remain subject to the current policies and guidelines described in this document. Holders of address space may potentially be required, at some time in the future, to return their address space and renumber their networks in accordance with the consensus of the Internet community in ensuring that the goals of aggregation and efficiency continue to be met.

    So, for example, someone could force all of Japan to change their IPv6 addresses for "administrative reasons"? I suspect this could get very political; imagine a governing agency of the IPv6 addresses wanted to sock it to a given area of responsibility.

    Or perhaps I'm not reading this correctly.

  • by _ph1ux_ (216706) on Tuesday May 22, 2001 @01:36AM (#206719)
    well kinda - but here is what needs to happen for widespread adoption of v6:

    the major backbone providers need to adopt v6 - not the end user. the reason is as follows:

    the model is this: tier 1-3 providers need to implement v6 on a backbone level - which will allow for major availability in the v6 arena when it comes to allocation.

    the end user needs only to have v4 nat happen - and have the v4 to v6 translation happen upstream. so - the end user has a 10.x private - which goes upstream to his isp, the isp has v6 peering relationships and has a block of legal v4 classes assigned to them. keep v6 out at the core backbone level for as long as possible - but each tier 1-3 has a certain v4 and v6 blocks that they own - and dole them out as needed v4 first.

    this allows for a "trickle down" approach to adoption of addy's in the new space.

    then as the net grows - you can still use v4 and v6 so as to maintain layers of complexity.

    re-allocate all v4 addys as class C.

    then as an end user client you only have a C net at best to allocate for dmz/external addy's - and make it semi-manditory that companies implement nat on a 10.x net. this will allow for almost unlimited flexibility in the corp - and very very flex environs for the ISP from 3 to 1 tiers.

    if i am wrong let me know - it is just an idea - what do you guys think.

    however I will admit that it will require a large renumbering of the net - but I as an admin have no complaints about incurring such a change - as it would be a fun project (to delegate ;) and would give a lot of experience to all people. and could be promoted as national v4 to v6 implementation month etc... it is about time we had such a large scale project anyway - for community purposes.... ??????

    let me know. I still will like it no matter what anyone says :)

  • It's probably most dependant on Router manufacturers. IPv6 addressing is backward compatible, however the internals of the packats make for certain incompatibilities that would need to be handled internally to the routers. Some manufacturers are developing smarter routers [slashdot.org] but not even these are setup to handle IPv6 yet as far as I know...

    --CTH

    --
  • So I like some of the ideas behind ipv6, but at the same time I dont like other things. I personaly see why evently we need to leave dotted quad, but the ability to censor seems to be beyond reason. To be able to stop a packet at the border, to be able to tell the type of media being transmited, to be able to cap users bandwidth useage, etc.

    I have heard that one of the reasons that people cant get ipv6 out there fast enough is because of companys like cisco and others not having ipv6 supported well as of yet, is this true?
    If its not, why is it taking so long?
    What are the bennifits to staying with dotted quad?
    Where is a good lamens description of ipv6?



    The Lottery:
  • I was wondering that myself, but I had heard that proccessing power was a large part as well. I have also heard that parts of the cisco OS just dont have support for it, but this isnt coming from a cisco certifed person just some local geek friends I know.


    The Lottery:
  • Hm. Forgive me, but which IPv6 policy is this from? There are plenty. Most of them are still being worked upon. Haven't heard of this one. Although, I can say that it'd be quite simple to change someone's IPv6 prefix; in ISP-ville, they just send a message to the router, and it does everything necessary without human intervention (gotta love machines taking all our jobs and performing them better than us, eh?). Now, I don't quite think this'd work very fast over all of Japan, nor do I think anyone would reassign Japan's entire IPv6 prefixes...

    Anyway, in all my rambling, I still wanna know which spec this is from.

  • by plcurechax (247883) on Tuesday May 22, 2001 @12:24AM (#206729) Homepage
    I've wanted to change to IPv6 for a long time now, but it seems that major upgrades that break things need to have a tangible benefit to end-users. Some application needs to be updated or replaced to handle IPv6. IPv6 doesn't appear to have a benefit to the end-user, only the network admins, so it the users are not demanding it. So far there isn't end-user application that users are screaming for.

    Makes me think it's a customer-driven world we live in.

    OS vendors and network hardware vendors are treating IPv6 as experimential, which is why people are not deploying it. People like the network to work with as little work as possible. You are more likely to be pitched about Voice over IP than IPv6 from a vendor salesperson.

    Another major concern is hardware compatiblity, people don't want to scrap older routers. IT departments have to watch their budgets these days. Most routers do support IPv6 or can be updated to do so.

  • There's so much misinformation about IPv6 in the replies I felt like commenting. (It's probably to late to do any good for slashdot, but at least I'll feel better for having done it.)

    IPv6 uses 128 bit (16 byte) addressing.

    The minimum allocation is still 1 address of course.
    The minimum network allocation is a /64. that's 2^64 addresses, or 281474976710656 class B address blocks. In theory, no ISP should ever have less, but clearly there's a market segment that has been ignored - ISP customers, and it will be serviced. I'm guessing that most home networks will get a /96 (4 billion address) but that's just a guess. Every ISP is probably going to do it differently.

    IPv6 packets have a standard for encryption, which arguably means they will be easier to encrypt than IPv4 packets, but they aren't all encrypted by default. Also, encrypted IPv6 packets can encrypt the source address, making traffic analysis more difficult. However, packets encrypted using the standard encryption are easy to identify as encrypted packets. This would make traffic analysis of encrypted traffic easier.

    Although technically no one owns IPv6 address space, it's extremely unlikely that anyone will ever be asked to return address space until we are close to running out. According to the IPv6 specs., renumbering should be a simple task, and it also shouldn't be necessary. I'm not sure I believe either of those statements, but that is what is claimed. The real reason for this clause is to remind ISPs to tell their customers that they can't take their address space with them when they switch ISPs. (I do think it's reasonable to assume this could happen again if it wasn't prevented.)

    Some Windows IPv6 support already exists. (I'm using it right now.) the website hs247.com/ [hs247.com] has a lot of information, go slashdot them. ;)
    FreeBSD and Linux already support IPv6. There are bugs, but then there are bugs in IPv4 too.

Any program which runs right is obsolete.

Working...