World IPv6 Day: Most-watched Tech Event Since Y2K

alphadogg wrote in with a fairly extreme bit of hyperbole saying "The nation's largest telecom carriers, content providers, hardware suppliers and software vendors will be on the edge of their seats today for World IPv6 Day, which is the most-anticipated 24 hours the tech industry has seen since fears of the Y2K bug dominated New Year's Eve in 1999. More than 400 organizations are participating in World IPv6 Day, a large-scale experiment aimed at identifying problems associated with IPv6, an upgrade to the Internet's main communications protocol, IPv4. Sponsored by the Internet Society, World IPv6 Day runs from 8 p.m. EST Tuesday until 7:59 p.m. EST Wednesday. The IT departments in the participating organizations have spent the last five months preparing their websites for an anticipated rise in IPv6-based traffic, more tech support calls and possible hacking attacks prompted by this largest-ever trial of IPv6."

Comments

Fingers crossed

[elrous0]

I haven't gotten much use my well-stocked bomb shelter since Y2K. Sure, religious types keep predicting the end of the world, and guessing wrong every time. And bad predictions aren't going to justify the money I've put into this goddamn thing. Did you know that a generator's gaskets will dry-rot over time, even if you don't use it? Well guess what, they will--and that shit is expensive to fix too.

Man, if only we could have one nuclear war. Then the neighbors might finally stop laughing at me.

Re:Fingers crossed

[drinkypoo]

Did you know that a generator's gaskets will dry-rot over time, even if you don't use it? Well guess what, they will--and that shit is expensive to fix too.

You joke, but those (possibly fictional but still representative) gaskets were designed to rot. There exist compounds which do that job, which cost very slightly more money, which will last longer than you will. In this particular case, the problem could be solved with all metal gaskets. In other cases, substituting silicone, nylon, or viton (depending on the application) for neoprene is what is needed.

One of the selling points of my motherboard (GA-MA770-UD3P 1.0, I.I.R. all the little letters C.) was that it had Japanese solid capacitors which are supposed to last longer than the usual kind in or out of normal use. Hopefully the other components match.

There are jillions of other examples, such as automakers using silicone-filled rubber (dunno what kind of "rubber") bushings instead of air-filled polyurethane, even though these will typically wear out during the lifetime of the vehicle. Anyone know of any product lines (of any kind) designed specifically for durability? How about a unifying resource for finding them?

Re:Fingers crossed

[Kjella]

If we really went to all out nuclear war, you might be better off just biting the bullet. What would happen afterwards is a massive temperature drop, worse than the last ice age.

A global average surface cooling of -7C to -8C persists for years, and after a decade the cooling is still -4C (Fig. 2). Considering that the global average cooling at the depth of the last ice age 18,000 yr ago was about -5C, this would be a climate change unprecedented in speed and amplitude in the history of the human race. The temperature changes are largest over land ... Cooling of more than -20C occurs over large areas of North America and of more than -30C over much of Eurasia, including all agricultural regions.

-20C is enough to turn Florida into Alaska, one thing is that people can live in Alaska temperatures but the world's food supply would utterly collapse. Billions would likely die of starvation or freeze to death, not war.

Re:Fingers crossed

[gnick]

Anyone know of any product lines (of any kind) designed specifically for durability?

Pyramids? Maybe not an ideal design, but that's certainly what the engineers of the time were shooting for. You know, durability and difficulty to assemble.

Re:Fingers crossed

[elrous0]

I don't think you appreciate the vast amount of canned peaches I have at my disposal.

Re:Fingers crossed

[Anonymous Coward]

Anyone know of any product lines (of any kind) designed specifically for durability?

Lemmy.

Re:Fingers crossed

[drinkypoo]

The problem is, I can't exactly call up the local contractor and have one put in. Maybe Halliburton would come out and pour me some shitty concrete.

Re:Fingers crossed

[Anonymous Coward]

Polyurethane bushings squeak a lot, and automotive consumers care way too much about reductions in noise, vibration, and harshness, to the point where every car carries around hundreds of pounds of needless sound-deadening material.

Re:Fingers crossed

[drinkypoo]

-20C is enough to turn Florida into Alaska, one thing is that people can live in Alaska temperatures but the world's food supply would utterly collapse. Billions would likely die of starvation or freeze to death, not war.

The hard part for the prepared is not being discovered and eaten (out of house and home, if not literally) by the unprepared. So the question is, how many years. It's not that difficult or expensive to stock three or even five years' worth of food if you're willing to be very bored with your meals. Better to be bored than hungry, however.

Incidentally, anybody not stockpiling food right now is going to either be hungry or poorer by next year. Fuel costs are projected to rise, and crop failures are currently ongoing and massive. There will be shortages. There will be price hikes. Food has gone up hugely in the last few months already and the shortages haven't even hit yet, this is just on spec and due to rising fuel costs.

Re:Fingers crossed

[drinkypoo]

Polyurethane bushings squeak a lot, and automotive consumers care way too much about reductions in noise, vibration, and harshness, to the point where every car carries around hundreds of pounds of needless sound-deadening material.

Polyurethane bushings greased with the proper kind of grease really do not squeak. Further, when the bushing is permanently affixed to a sleeve as are the polyurethane bushings I installed at the pivots of the Dana 50 TTB in the front of my 1992 F250 7.3 4x4, that connection will squeak even less. I ripped most of the asphalt (except from the toe pan, but yes from the floor) and all of the interior out of my 1989 Nissan 240SX — yes, it got hot in there — and I couldn't hear the poly bushings at all even with the silencer in the "fancy" (cheap on eBay, though) exhaust, which brought it down pretty much to stock levels since I had no header. But then, they came with the proper grease and I used it.

If vehicles are designed to take a poly bush with a metal sleeve then this is a non-issue. And using hollow poly bushings provides superior ride to silicone-filled rubber in every way as they are more consistent. This actually has the effect of improving ride quality under the hands of competent engineers, because they may tighten up the suspension design and do the damping in the shock absorbers where it belongs. Then all you have to do is drop the typical OE Tokico crap for some Bilsteins or similar...

Re:Fingers crossed

[Anonymous Coward]

One mans 'bomb shelter' is another's 'Nag-Proof Man Cave'. If properly stocked, well provisioned and buried deep enough. Well, you get the idea.

Now about the neighbors, get one of those page sized 'foreclosure' window stickers. The ones with the red background and the word FORECLOSURE in big bold lettering. Wait till they are all away and post one of them up in a large picture window over looking the street. It won't stop them from being dickish about your hole in the ground, CORRECTION Your Man Cave, but it'll open up new topics for all to talk about.

Re:Fingers crossed

[gnick]

OK, LOCAL contractor I'll grant you. But with sufficient real estate and a sufficient supply of indentured laborers, I don't see the hurdle. However be warned, the folks that have opted to use these for vacation homes have taken severely extended vacations. In the words of the prophets, "You can check out any time you'd like, but you can never leave."

Re:Fingers crossed

[elrous0]

Yep, Lemmys are able to withstand decades of incredible amounts of abuse, and still keep going.

Re:Fingers crossed

[elrous0]

Incidentally, anybody not stockpiling food right now is going to either be hungry or poorer by next year.

On the upside, those of you looking for that little extra incentive to finally go on that diet you've been meaning to go on are in luck.

Re:Fingers crossed

[Eunuchswear]

Incidentally, anybody not stockpiling food right now is going to either be hungry or poorer by next year.

Start the hoarding now to bring on the collapse earlier.

Let the panic buying commence!

Re:Fingers crossed

[Toonol]

Nuclear winter is was mainly a propaganda device. It would have a cooling effect, but the claimed magnitude was terrifically overstated. It's a bit like global warming; real, but it's not going to result in 20 meter ocean risings.

Re:Fingers crossed

[mellon]

Dude, it's way worse than that. Slashdot doesn't work unless you enable IPv4. I'm posting this using IP-over-carrier-pigeon!

(I'm really disappointed that slashdot, supposedly a geek site, didn't bother to enable IPv6 for World IPv6 day.)

Re:Fingers crossed

[xaxa]

Anyone know of any product lines (of any kind) designed specifically for durability?

Anything at all that's designed for durability?

Anything where replacing it is more expensive or inconvenient than paying for something engineered to last as long as possible, especially anything where safety is critical. Parts of power grids, railway tracks/stations/vehicles, things on large buildings, satellites, parts of continuously-run factories, bridges/tunnels, ...

Re:Fingers crossed

[Bob the Super Hamste]

To be fair gaskets, especially automotive ones, are in most cases easily replaced with much more durable ones. When ever I replace automotive gaskets I get the copper ones if available, or the ones with the "lifetime" warranty. If looking for something designed to last I suggest the Model 500 phone, it seems that when a company actually has to support something they do a better job designing it.

Re:Fingers crossed

[ilsaloving]

Now I know how my non-technical friends feel when I start going on about network topology or the benefits of proper normal form in a database.

Re:Fingers crossed

[shutdown -p now]

Don't forget to buy that silver either - you'll need real money for the upcoming apocalyptic society! ~

Re:Fingers crossed

[lytithwyn]

Did you know that a generator's gaskets will dry-rot over time, even if you don't use it? Well guess what, they will--and that shit is expensive to fix too.

They dry rot especially if you don't use it. By running the generator fuel/oil (depending on which gasket) are splashed/pushed onto the gasket, moisturizing it and prolonging it's life. Crank it up every now and again and be sure to use fuel stabilizer if you don't drain the gas tank; otherwise the gas turns sort of green and rotten.

Re:Fingers crossed

[hjf]

Same reason I chose it for my home server. That, and the fact that it supports ECC memory.

Re:Fingers crossed

[hjf]

Durability? Try induction lighting. Rated for 50.000+ hours. Shitty light, but you can't get any longer life than that.

Re:Fingers crossed

[ravrazor]

Can you shut up? I was interested in reading some comments about IPv6 day.

Re:Fingers crossed

[Vegeta99]

There are jillions of other examples, such as automakers using silicone-filled rubber (dunno what kind of "rubber") bushings instead of air-filled polyurethane, even though these will typically wear out during the lifetime of the vehicle.

Just replaced control arm bushings and strut cups on my VW. It's advised to not replace them with polyurethane replacements because it will make the handling too stiff.

Re:Fingers crossed

[onepoint]

If this type of war did happen, and let's just say that there was no radioactive fallout. I would venture to guess that all agriculture zone 9 would become about a zone 5 http://www.willisorchards.com/zone ... with that being the case, in North America, we would then move southerly to Mexico and the west coast of the USA.

I don't know much about South America in general, but I would guess that most of the amazon lower basin would become the worlds bread basket after a large investment in fertilizer

I don't even want to think about water level drops, and change in weather patterns

Re:Fingers crossed

[Anonymous Coward]

Millions of peaches? Peaches for free? Millions of peaches? Peaches for me?

look out!

Re:Fingers crossed

[drinkypoo]

Just imagine how I feel when people try to make car analogies.

Re:Fingers crossed

[drinkypoo]

Just replaced control arm bushings and strut cups on my VW. It's advised to not replace them with polyurethane replacements because it will make the handling too stiff.

Too stiff for what? Anyway, that doesn't contradict what I said; your suspension was designed for squishy bushings. Ostensibly, so was the front suspension on my Ford, but it's like a whole new truck with poly up there. That's partly due to the fact that one of the rubber (without silicone) bushings had let loose of the metal sleeve inside of it, and partly due to the fact that it's a 6,800 lb truck (wet and ready) and it really needs a firmer bushing than was specified to begin with.

Re:Fingers crossed

[Vegeta99]

Too stiff for day-to-day driving, but mechanically speaking, too stiff for the control arms to take. So, replacement with a poly bushing essentially leaves you with a more expensive repair 25,000 down the line.

I used to be a Ford guy, and they make everything too squishy. I had to re-valve my T-bird's automatic transmission because the shifts were so soft. Like your truck, the T-bird's suspension seemed to be designed for it's Mustang cousin, simply not equipped to handle a fatass boat.

Re:Fingers crossed

[Eunuchswear]

No, no. You need the silver to disguise yourself as a smurf.

Re:Fingers crossed

[ilsaloving]

Let me guess.... About the same as I felt when I was watching Jurrasic Park and the one character exclaimed, "I know this! It's UNIX!"

Re:Fingers crossed

[drinkypoo]

Or seeing a "sun4m i386" ... uh yeah. Don't think the Sun386i was a sun4m. Not sure what it actually called itself, though.

Re:Fingers crossed

[Anonymous Coward]

That's not what your GF said.

Guess who's not taking part?

[squiggleslash]

Big names like Google are:

$ host www.google.com
www.google.com is an alias for www.l.google.com.
www.l.google.com has address 74.125.45.147
www.l.google.com has address 74.125.45.104
www.l.google.com has address 74.125.45.99
www.l.google.com has address 74.125.45.105
www.l.google.com has address 74.125.45.103
www.l.google.com has address 74.125.45.106
www.l.google.com has IPv6 address 2001:4860:800a::6a
$

But one tech website you'd expect to want to dabble in the new and good for some reason isn't:

$ host slashdot.org
slashdot.org has address 216.34.181.45
slashdot.org mail is handled by 10 mx.sourceforge.net.
$

Well, of course!

Re:Guess who's not taking part?

[Anonymous Coward]

They'll get around to it when they get to adding Unicode support. To be fair, Unicode is only 20 years old and IPv6 only 13 years old, so they aren't much later with these technologies than they are with their stories.

Re:Guess who's not taking part?

[sakdoctor]

Dabble in the new? Slashdot can't even deploy Ajax correctly.
Slashdot still has its place, but it's definitely a 'legacy' website.

Re:Guess who's not taking part?

[Anonymous Coward]

This may be true but CmdrTaco can still write one hell of a comic book movie review. That's got to count for something. Doesn't it?
 
Slashdot hasn't been a real tech site for over half its life.

Re:Guess who's not taking part?

[ledow]

And yet they've posted four IPv6 stories in the past month (and dozens before) where EVERYONE brings this up too.

You'd think they'd have bothered to do it by now. Shows you how "with the geeks" this site really is.

Re:Guess who's not taking part?

[squiggleslash]

In fairness, Unicode requires quite a bit of testing to make sure it works, even if you're using tools that, out of the box, should support Unicode transparently. In Slashdot's case, a legacy of building the site on what was considered top of the line in the 1990s has left them with a lot of things that can go wrong.

IPv6, on the other hand... well, if you're using virtual hosting (and /. is), all that it takes is to turn on IPv6 on the front facing server, give it an IP address (which could just be a 6to4 address), update DNS, and, well, it either works or it doesn't. A half competent sysadmin should be able to do all that in less than ten minutes. I say that, because I am a half competent sysadmin, and adding IPv6 to the websites I host (on a third party VPS no less) took just that. I enabled 6to4 on the VPS itself, assigned the 6to4 address, and added the DNS record. And everything "just worked". Took me less than 15 minutes.

I'd be very interested to know why CT hasn't done this.

Re:Guess who's not taking part?

[drinkypoo]

To me the sad thing about slashdot is that it doesn't do that much. All the hard stuff is in scalability. The interface hardly does ANYTHING. Well, there is a really really bad interface that does some stuff, but anyone serious has disabled big chunks of it because it blows chunks. Why haven't we seen a complete rewrite by now? Answer, because people are actually paying for this crap! You subscribers are the problem.

Re:Guess who's not taking part?

[Anonymous Coward]

Slashdot doesn't even have https, what are you expecting?

Re:Guess who's not taking part?

[Abstrackt]

Dabble in the new? Slashdot can't even deploy Ajax correctly.
Slashdot still has its place, but it's definitely a 'legacy' website.

<getoffmylawn>And that's the way we like it!</getoffmylawn>

Re:Guess who's not taking part?

[tepples]

Slashdot has HTTPS as long as you log in and subscribe.

Re:Guess who's not taking part?

[hairyfeet]

Question: Has anybody figured out how to keep the fact that IPV6 will give everyone a unique and easy to trace address from making the switch into a *.A.A and big brother government wet dream? Because frankly I kinda like how dynamic addressing makes for a PITA to *.A.As and big brothers everywhere. Even though I'm not doing anything I sure as hell don't want some daily report handed to some pencil pusher, nor do I want to explain that the BT I was downloading was the latest ultimate driver DVD to some *.A.A monkey who thinks I have nothing better to do than download his latest Titney Spears 'epic".

So frankly if the eggheads haven't figured out this basic problem (which considering their answer to NAT was "you don't need it" frankly I don't trust them to wipe their noses) then personally i hope it takes a hell of a long time to switch. Let the third world countries that were late to the party have IPV6 while we that already have addresses use something like 6to4 to talk to them. Because honestly I really really REALLY don't trust my government not to sell us out right now, nor do i trust them not to label someone a terrorist for say putting up a mirror of Wikileaks. I know that with determination you can still be hunted down, but lets not make it any easier for them, shall we?

Re:Guess who's not taking part?

[IamTheRealMike]

The problems with Unicode aren't to do with legacy. They're to do with a poorly implemented anti-abuse system. Some people were using control characters to screw up the page rendering. Their solution was to simply break Unicode rather than come up with blocks for the different types of abuse.

Re:Guess who's not taking part?

[TheCycoONE]

http://www.regular-expressions.info/unicode.html#prop

Re:Guess who's not taking part?

[SockPuppetOfTheWeek]

The real reason Unicode isn't supported is because we'd have too much Zalgo spam on Slashdot if it was. (Well, that, and the control characters problem that Mike mentioned. The real Mike, not the fake one.)

Re:Guess who's not taking part?

[SockPuppetOfTheWeek]

The Ajax works just fine; it's the CSS that's always broken.

Re:Guess who's not taking part?

[shutdown -p now]

I wouldn't say that "Working ..." popup getting stuck on every third comment post is "working just fine" (this is in Chrome stable).

Oh, and what about the bug where every time you left-click on a comment (including in the reply textbox... good luck selecting anything!) scrolls back to the next minimized comment up in the thread and opens it? It's still there, and has been reported countless times ever since "Slashdot 3.0".

Re:Guess who's not taking part?

[Anonymous Coward]

And yet they've posted four IPv6 stories in the past month (and dozens before) where EVERYONE brings this up too.

You'd think they'd have bothered to do it by now. Shows you how "with the geeks" this site really is.

Or maybe it shows how much IPv6 matters in practice. To wit: not much.

Re:Guess who's not taking part?

[magamiako1]

Of course, the egg-heads have figured it out.

It's called IPv6 Privacy Extensions. Look it up.

Re:Guess who's not taking part?

[SockPuppetOfTheWeek]

Yeah, I probably should have said "HTML and CSS". Although that "Working" thing is just always there on Firefox (right from when the page first loads), so I assumed it was a CSS glitch with its display or visibility not getting set correctly. Its classname is currently "busy genericspinner hide", but it's visible.

The comment-opening... I think it's due to the comments nesting inside the parent comments, which are supposed to be clickable, but since the children comments are inside it they're clickable too and the parent gets the click event. So I don't know whether to call that CSS or HTML brokenness. It's a nesting error.

Re:Guess who's not taking part?

[Anonymous Coward]

Oh, and what about the bug where every time you left-click on a comment (including in the reply textbox... good luck selecting anything!) scrolls back to the next minimized comment up in the thread and opens it? It's still there, and has been reported countless times ever since "Slashdot 3.0".

I had a nice litte rant going about how awful this problem is, but I got so tired of wrestling with this very issue that I actually gave up on it.

I have realized that if you click the checkbox for Anon, then tab down to the comment box, it gets you around the problem. For some reason that checkbox doesn't trigger whatever that ridiculous JS is.

Re:Guess who's not taking part?

[hitmark]

something like this? http://www.faqs.org/rfcs/rfc4941.html

Re:Guess who's not taking part?

[Anonymous Coward]

As opposed to our current situation where every broadband connection has a more or less static IP for months, if not years, at a time? Of course, see the siblings on why IPv6 is actually better than IPv4 here.

Re:Guess who's not taking part?

[jc42]

Slashdot can't even deploy Ajax correctly.

Hey, nobody can deploy Ajax correctly. ;-)

Re:Guess who's not taking part?

[Raenex]

In Slashdot's case, a legacy of building the site on what was considered top of the line in the 1990s has left them with a lot of things that can go wrong.

It's just a bunch of homebrew Perl scripts. And it isn't that hard to get Unicode working.

Re:Guess who's not taking part?

[hairyfeet]

Uh huh. Question: Has ANYONE actually implemented that, or made it a required part of the standard? Because I can't find that anywwhere in your link, nor has any bunch I've seen offering IP V6 offered anything like that. It sounds like a nice idea for sometime in the future but I don't see any routers or other devices I can actually buy or use allowing this, do you? So far everything I've seen they just hand you a block and you are on your own pal.

Now at least currently with the larger regional ISPs you can have a new dynamic address sometimes two or three times a week (In my own case a simple power cycle will get me a new address assigned) and frankly with all the nasty things coming out about governments everywhere I really don't want to change out a harder to trace setup for an easier to trace one, do you?

And folks can waste their mod points all the want but in case they ain't noticed we are getting laws like PROTECT IP rammed down our throat that will get you time in PMITA prison just because some troll company paid by the record companies (and in the case of media defender actually supplying CP on the side to give busts to the NY DA) says it was you. This is NOT a nice time we are living in folks, frankly it is the tech guys VS the jackboots. I just want to make sure the tech side comes out on top, don't you?

Re:Guess who's not taking part?

[hitmark]

This would not the done on the router level in any case.

The basic thing about IP6 is that as it uses hex it can take the MAC of a card and use that as the basis for a IP. Iirc, IPX had a similar feature.

However, it is still possible to tell the os to use something different then a MAC as a basis.

https://secure.wikimedia.org/wikipedia/en/wiki/MAC_address

note how if bit 2 of byte 1 in a MAC address is 1 rather then 0, the address is "locally administered". This means that the MAC on the chip have been overridden by the admin.

This can already today, in combination with something like a collection of public wifi hotspots visited at "random", be used to stay relatively anonymous online.

Re:Guess who's not taking part?

[Anonymous Coward]

Slashdot doesn't even support SSL. And trust me, they should.

Re:Guess who's not taking part?

[hairyfeet]

Now you seem to know a hell of a lot more about this than I do, since I mainly deal with SOHOs, SMBs, and home users, so maybe you can fill in the blanks. How exactly would one use this if one were at home?

One of the nice things about dynamic addressing is the "face in the crowd' problem it gives the *.A.As, since with a large regional ISP like mine you are literally talking about half a million people, a good 90% of which are a single power cycle away from a new address and which (at least in my area) gets a new address weekly whether you power cycle or not. That means you are literally talking about mounds of records to go through to figure out who had address X at time Y which is why ISPs refuse to process more than a dozen or two a month which keeps these massive fishing expeditions by the *.A.A and the total scum they hire (if you didn't hear Media Defender or whatever they call themselves now was actually seeding CP on P2P sites to give the NY DA people to bust) from just grabbing huge chunks of people for their little extortion racket.

Now so far all I've seen when it comes to IP V6 is the ISP gives you a set chunk and that's it, you're stuck with those addresses. Now while using some bits from the Mac or whatever might help to obfuscate the IP address a little bit, I just don't see how it won't be trivial for the *.A.A types to get the ISPs just to hand over a list. After all you will be having the same IP address regardless since there really isn't any point to dynamic addressing when you are talking trillions of addresses.

While I'll admit I haven't been able to completely wrap my head around IP V6 yet (sadly there is no "IP V6 for dummies" and my area is still strictly IP V4 so no hands on) I'm gonna assume there is some sort of prefix that gives you the general area/ISP of the address, yes? Wouldn't it then be trivial to use that to find the exact person, since without dynamic addressing they won't have those mounds of data to go through? After all you still have to have an address that can be routed from A to B so wouldn't it be trivial to backtrace? After all it is only the huge amounts of data that make it a PITA for the *.A.As now and that would be gone if everyone had static.

While I like the idea of having my own block to do with as I will frankly after watching some of the horseshit like PROTECT IP (BTW if you haven't signed the petition against it you really should, the petition apparently scares the *.A.A enough they are trying to claim now we must not exist so all of those like me that signed have been sending emails to congress to show we most certainly DO exist) and the crazy laws being supported by our government frankly anything that makes it even a tiny bit easier for those in power to monitor you scares the hell out of me.

Then add in the fact the eggheads behind IP V6 refused to make any kind of backwards compatibility (stupid) and then refused to even entertain the thought that there might actually be decent uses for NAT that some might want, like systems management and having machines only outside accessible under certain conditions (arrogant) really makes me wonder if they have thought this thing through. After all if all governments were fair and just and listened to the people we wouldn't even be having this conversation yet we are. Sadly it is gonna end up with our freedoms not in OUR hands, but in the hands of the techs, for it is ONLY them who can bake in enough countermeasures and obfuscation to keep our last free resource from turning into an Orwellian nightmare. And sorry if that sounds a little melodramatic, but I have a friend that works forensics in the state crime lab and you'd be surprised how quickly one of those guys can ruin someone, and with PROTECT IP giving them even MORE power I really don't like the looks of things ATM.

And...

[Anonymous Coward]

And even less is going to happen.

Re:And...

[bmo]

The only reason why Y2K /wasn't/ a disaster was because people worked their asses off for it to not happen.

Idiots everywhere...

--
BMO

Re:And...

[Marillion]

I agree a lot of people worked very hard. But another very important factor that I feel is under-appreciated is that problem wasn't really as dire as the mainstream was lead to believe it was.

At the time, I was working for an air carrier. Passengers can book a flight 330 days in advance and past flights are kept for 30 days. The interface with the reservation system uses month and day. This led to my mantra, "You can't have a Y2K problem if you don't have a Y."

Re:And...

[Anonymous Coward]

Let me guess - you were a baggage handler.

Re:And...

[JourneymanMereel]

Y2K was definitely overblown by the mainstream media. I knew many people who were scared their car wouldn't start on Jan. 1, 2000.

Re:And...

[jc42]

The only reason why Y2K /wasn't/ a disaster was because people worked their asses off for it to not happen.

Where "people" is defined as "Cobol programmers". ;-)

There were any number of analyses (before and after 2000-1-1) that concluded that around 99.5% of the actual Y2K problems were in vintage Cobol code. There were any number of jokes created based on this curious fact. But in other languages, date handling was long-since packaged in library routines, so a single upgrade could fix all their potential problems without the developers even being aware that there had been a problem. I work in several languages, and I was unable to find any Y2K problems in any of the code I was working with.

Cobol, on the other hand ... I recall reading one "study" of one company's code, in which the writer said that when his count of different date formats in the database passed 180, he decided he understood why there was a problem. And he understood why no centralized solution was possible; he and his buddies just had to hunt down every single date manipulation, figure out what sort of hokey calculations it was doing, and in most cases, fix it. In many cases, there weren't enough spare bytes in the fixed-field-width DB records to fix it without finding and rewriting every line of code in every program that accessed such fields.

And many people pointed out that the Cobol language wasn't the problem. The problem was the Cobol programmer community, which has long had such practices that should qualify them for a diagnosis of legal insanity and institutionalization at the expense of their employers that encouraged and rewarded such behavior. Nobody seems to have found any other programming community that did things this insane. There is a conjecture that this is because Cobol was almost strictly a "business" programming language, and in a business setting, dealing with anything that's more than a year in the future is generally considered irresponsible and grounds for firing. But anything that will produce a short-term saving, such as saving a byte of disk space per record by using a compact date/time format is strongly approved. And this is the main thought process that led to Y2K problems.

Re:And...

[Kjella]

Obviously there were a lot of critical systems that had to be fixed and a lot of people that worked very, very hard to fix them. That's not what we're talking about. The point is that so was everything else, every insignificant little auxiliary system had been given the full testing and overhaul. The press was running around like crazy on the 1/1/2000 trying to find stuff that didn't work and practically came up empty.

I've seen bog standard OS upgrades go less smooth than y2k, simply because you hadn't done that much dry-run testing on every weird hardware/software/test case combination out there. It's sorta expected you have tested the 99% critical and no-so-critical things and work out the 1% kinks afterwards, but through the whole "the sky is falling" attitude y2k was tested to 99.99999% or there abouts. Great in ensuring it all went silky smooth, but just maybe quite a bit of overkill. Both in that many spent ungodly amounts on excessive and very expensive y2k compliance and consulting services and that many stuffed the great y2k budgets with other things.

Hardly the most-anticipated 24 hours

[Ruvim]

This event had been very unpublicized for this to be the most-anticipated 24 hours in tech industry for the last 10 years.

Re:Hardly the most-anticipated 24 hours

[gnick]

It doesn't need to be - Very few people need to know to avoid it affecting their normal routine. My ISP (f'n Comcast) isn't helping me out with IPv6 and neither is my employer (a major national lab), but I expect zero effect. I suspect that I'm just a typical example of the vast majority of the population.

Re:Hardly the most-anticipated 24 hours

[FST777]

Not for lack of trying, but with the vast majority of the people, even the tech-aware ones, the typical reaction to anything "IPv6" is: "Wha..?", leading to lack of interest at first mention.

Which 24 hours sine Y2K were more anticipated? The launch of the iDevices? Is that really the tech industry? I work in networking (at a large web-based content-provider), and in "the field", this is a very, very important day (which we all hope shall pass relatively silently).

Re:Hardly the most-anticipated 24 hours

[Sancho]

It doesn't need to be - Very few people need to know to avoid it affecting their normal routine. My ISP (f'n Comcast) isn't helping me out with IPv6 and neither is my employer (a major national lab), but I expect zero effect. I suspect that I'm just a typical example of the vast majority of the population.

You're actually among the most likely people to notice, then. Depending upon your OS/browser/configuration, sites can fail or become quite slow if they advertise a AAAA record but you only have v4 connectivity. In fact, that's one of the things that's being measured today.

Me? I have v6 and I don't notice a difference. I have to go looking with tcpdump to see if I'm actually connecting to the v6 address or to the v4 address. As far as I can find, there's no place in my browser which tells me one way or the other.

Re:Hardly the most-anticipated 24 hours

[davew]

It's been pretty hard to miss in networking circles specifically. Reason I say this is:

A lot of people here seem to be missing the point of the event. It's not really about boosting IPv6 traffic for the day; there are still other links in the chain to get sorted out before we can do that (most visibly, users' LANs and internet connections.) But one big thing that's been holding up the dual stacking of BIG websites, the kind participating today, is a really tiny proportion of users who don't know they have IPv6 configured and it's broken.

The numbers are in or around a fraction of a percent, but for a really big site, that's too many users. We need to find these guys and get them to fix it.

So the target for this one hasn't been mainstream users or even system administrators, but ISPs and IT support departments, so that they can find the problems in advance. (Maybe you fall into this category and missed it, in which case, sorry, but I saw it in pretty much all the networking channels I was aware of over the past seven months.)

So far, eighteen hours in, I've not seen many reports of problems. This is EXCELLENT NEWS, because if the perception of problems turns out to be much greater than reality, some of the participants might decide to leave IPv6 on permanently. That's one more link in the chain, so that ISPs that do deploy IPv6 to their users will actually begin to see some more take-up of traffic. Step by step, such is how the chicken/egg problem is unravelled.

Slashdot has no AAAA address

[simoncpu was here]

I was happy to see xkcd, Google, Yahoo, Facebook, and Plurk turn on their IPv6 capabilities, but I was quite sad that Slashdot didn't take part in the World IPv6 Day.

Re:Slashdot has no AAAA address

[bill_mcgonigle]

I was quite sad that Slashdot didn't take part in the World IPv6 Day.

That's on the list right after getting UNICODE to work.

Re:Slashdot has no AAAA address

[Anonymous Coward]

xkcd has had ipv6 for a long time, and has been running in dual stack.

I only know this because I screwed up my ipv6 configuration a long time ago and forgot about it. I then was unable to reach xkcd for a long time, until I remembered.

I don't think that xkcd is participating, same with heise.de

Re:Slashdot has no AAAA address

[Anonymous Coward]

I am happy to see Slashdot not using IPv6 as it is one of the few sites that I am not having problems trying to connect to. Wikipedia also seems to work ok. Google has been almost impossible to use today.

Re:Slashdot has no AAAA address

[tom17]

Did Facebook & Yahoo turn theirs off again, or are there actually problems with this whole thing?

~$ host yahoo.com
yahoo.com has address 67.195.160.76
yahoo.com has address 69.147.125.65
yahoo.com has address 72.30.2.43
yahoo.com has address 98.137.149.56
yahoo.com has address 209.191.122.70
yahoo.com mail is handled by 1 e.mx.mail.yahoo.com.

~$ host facebook.com
facebook.com has address 69.63.189.11
facebook.com has address 69.63.181.12
facebook.com has address 69.63.189.16
facebook.com mail is handled by 10 smtpin.mx.facebook.com.

Re:Slashdot has no AAAA address

[simoncpu was here]

I don't know why, but many sites have AAAA records for www.example.org but not for example.org:

$ host -t AAAA yahoo.com
yahoo.com has no AAAA record

$ host -t AAAA facebook.com
facebook.com has no AAAA record

$ host -t AAAA www.yahoo.com
www.yahoo.com is an alias for fpfd.wa1.b.yahoo.com.
fpfd.wa1.b.yahoo.com has IPv6 address 2001:4998:f011:1fe::3000
fpfd.wa1.b.yahoo.com has IPv6 address 2001:4998:f011:1fe::3001

$ host -t AAAA www.facebook.com
www.facebook.com has IPv6 address 2620:0:1c08:4000:face:b00c:0:2

Re:Slashdot has no AAAA address

[tom17]

Ahhh oops. That explains it.

Also, I noticed that although host was not giving an ipv6 addy, wget was, but that's just because facebook.com redirects to www.facebook.com that does have ipv6. Missed that first time.

~$ wget facebook.com
--2011-06-08 12:00:04-- http://facebook.com/
Resolving facebook.com... 69.63.181.12, 69.63.189.16, 69.63.189.11
Connecting to facebook.com|69.63.181.12|:80... connected.
HTTP request sent, awaiting response... 301 Moved Permanently
Location: http://www.facebook.com/ [following]
--2011-06-08 12:00:04-- http://www.facebook.com/
Resolving www.facebook.com... 2620:0:1c00:0:face:b00c:0:2, 69.63.189.26
Connecting to www.facebook.com|2620:0:1c00:0:face:b00c:0:2|:80... connected.
HTTP request sent, awaiting response... 302 Found
Location: http://m.facebook.com/?w2m&refsrc=http%3A%2F%2Fwww.facebook.com%2F&_rdr [following]
--2011-06-08 12:00:05-- http://m.facebook.com/?w2m&refsrc=http%3A%2F%2Fwww.facebook.com%2F&_rdr
Resolving m.facebook.com... 66.220.147.43
Connecting to m.facebook.com|66.220.147.43|:80... connected.
HTTP request sent, awaiting response... 200 OK

That's a good dual stack test for the client end right there... 4 and 6 in one instance of wget.

Re:Slashdot has no AAAA address

[xaxa]

Try:

> host www.facebook.com
www.facebook.com has address 69.63.189.11
www.facebook.com has IPv6 address 2620:0:1c18:0:face:b00c:0:2

> host www.yahoo.com
www.yahoo.com is an alias for fpfd.wa1.b.yahoo.com.
fpfd.wa1.b.yahoo.com has address 69.147.125.65
fpfd.wa1.b.yahoo.com has address 67.195.160.76
fpfd.wa1.b.yahoo.com has IPv6 address 2a00:1288:f006:1fe::3000
fpfd.wa1.b.yahoo.com has IPv6 address 2a00:1288:f006:1fe::3001
fpfd.wa1.b.yahoo.com has IPv6 address 2a00:1288:f00e:1fe::3000
fpfd.wa1.b.yahoo.com has IPv6 address 2a00:1288:f00e:1fe::3001

I think the bare domain does a redirect to the www. anyway.

Re:Slashdot has no AAAA address

[instagib]

I followed your advice, but now it says "No space left on device".

Re:Slashdot has no AAAA address

[tom17]

What advice?

Did you reply to the wrong post or something?

Re:Slashdot has no AAAA address

[tepples]

That's on the list right after getting UNICODE to work.

As I understand it, the decision to make Unicode not work was intentional. Slashdot started to use a character whitelist soon after vandals started to post comments with a directionality control character. The vandals used this to spoof moderation scores and break the layout of following comments. I'm willing to provide citations on request.

Re:Slashdot has no AAAA address

[Anonymous Coward]

There are relatively few control characters in Unicode. It'd make far more sense to use a blacklist.

Not to mention the whitelist is absurdly small and doesn't include such things as &hellip; (HTML notation since it won't let me post the character, obviously).

Re:Slashdot has no AAAA address

[bill_mcgonigle]

I'm willing to provide citations on request.

Do you have them handy? I'd don't quite remember the time frame - I'm thinking this was about a decade ago, when the primary use of Unicode was to break Slashdot, but today that's much different.

Everybody else seems to have figure out how to handle Unicode input and re-display.

Re:Slashdot has no AAAA address

[RabidMonkey]

you are the target of IPv6 day then - it's a chance for you to try and find out whats wrong. contact your isp, look at your OS version, do upgrades, look at your router. If you're having problems now, you'll have them later - don't wait and hope they go away magically.

Re:Slashdot has no AAAA address

[Anonymous Coward]

*whoosh*

Re:Slashdot has no AAAA address

[tepples]

I wrote about the character whitelist earlier in this comment.

Re:Slashdot has no AAAA address

[tom17]

Yeah pretty much!

recursive redirect joke?

Re:Slashdot has no AAAA address

[Anonymous Coward]

Google and Facebook have had IPv6 enabled for quite a while. This was just enabling it globally rather than edge cases.

Re:Slashdot has no AAAA address

[bill_mcgonigle]

Yeah, and that was a fine quick-n-dirty fix nearly a decade ago, when Unicode wasn't really important. Fast-forward to 2011 and everybody else is successfully accepting Unicode with blacklists for the troublesome bits. I don't see the value to the Slashdot community by continuing to use the 2002 fix.

Re:Slashdot has no AAAA address

[Anonymous Coward]

I was happy to see xkcd ... turn on their IPv6 capabilities

xkcd has supported IPv6 for a long time, it has nothing to do with IPv6 day.

Re:Slashdot has no AAAA address

[Anonymous Coward]

Did Facebook & Yahoo turn theirs off again, or are there actually problems with this whole thing?

~$ host yahoo.com
yahoo.com has address 67.195.160.76
yahoo.com has address 69.147.125.65
yahoo.com has address 72.30.2.43
yahoo.com has address 98.137.149.56
yahoo.com has address 209.191.122.70
yahoo.com mail is handled by 1 e.mx.mail.yahoo.com.

~$ host facebook.com
facebook.com has address 69.63.189.11
facebook.com has address 69.63.181.12
facebook.com has address 69.63.189.16
facebook.com mail is handled by 10 smtpin.mx.facebook.com.

Yeah, the whole idea was that participants in the survey would turn it off once the day was over. According to the Internet Society FAQ

Why are websites turning IPv6 off after 24 hours?

One of the main purposes of the test flight is to enable website owners to assess the ability of end users to reach their websites. It is essential for websites to analyze impact before and after in order to make a longer term impact assessment and overall plan. This 24 hour period will help the Internet community continue to improve IPv6.

Re:Slashdot has no AAAA address

[tom17]

Yeah, no. I posted that before they turned it off lol.

They only had it turned on for the www.*, you see. That was my error.

Legacy devices

[Anonymous Coward]

Some people with legacy devices, which may be so because of legacy softwate, are stuck with IPv4 for some time to come.

Did Skype participate?

[bill_mcgonigle]

I had calls getting dropped every 5 minutes or so last night. Then again, Skype's entire network seems to go down on occasion, so perhaps an IPv6 test is an unlikely cause.

But, I'm safely small enough that my ISP is just starting to talk about offering an IPv6 trial in a city far far away. I'm signed up for them to let me know in 4 years that IPv6 is available for testing...

Whatever happened...

[Anonymous Coward]

to IPv5?

Re:Whatever happened...

[lxs]

Wiki is your friend.

tl;dr;
They skipped 5 to avoid confusion with the Internet Stream Protocol.

So whatever happened to IPv5?

[SpryGuy]

That's my question.

Re:So whatever happened to IPv5?

[RazzleFrog]

Odd numbers are development releases - duh. Only even numbers are for stable releases.

Re:So whatever happened to IPv5?

[Laser_47]

IPv5 was originally designed as a streaming protocol, but they abandoned it now that everything is encapsulated in HTTP.

Re:So whatever happened to IPv5?

[i kan reed]

It was strictly conjecture from when IPv4 was just released. The realities of how the internet was actually used defined IPv6 instead.

Re:So whatever happened to IPv5?

[fedux]

ST2 distinguishes its own packets with an Internet Protocol version number 5, although it was never known as IPv5.

See: http://en.wikipedia.org/wiki/Internet_Stream_Protocol

Re:So whatever happened to IPv5?

[Stween]

http://youshottheinvisibleswordsman.co.uk/2011/01/29/ipv4-5-6/

Re:So whatever happened to IPv5?

[Anonymous Coward]

It's like Star Trek movies!

Funny/interesting addresses

[Sinus0idal]

I've seen a few already today!

www.facebook.com has IPv6 address 2620:0:1c18:0:face:b00c::
cisco.v6day.akadns.net has IPv6 address 2001:420:80:1:c:15c0:d06:f00d
www.luns.net.uk has IPv6 address 2a01:8900:0:1::b00b:1e5
www.bbc.net.uk has IPv6 address 2001:4b10:bbc::1

Does v6 kick off 'IP addresses as a marketing tool'? :)

Re:Funny/interesting addresses

[elPetak]

IP addresses in IPv6 are ugly so I doubt they can be used that way.
OTOH, saying some site is "IPv6 certified" or some other BS like that may be used as marketing tool, and people would buy it just like they buy anything in a green package thinking it's healthy.

Re:Funny/interesting addresses

[joeytmann]

I find it rather funny that www.facebook.com has face in part of the address.

Re:Funny/interesting addresses

[Anonymous Coward]

It's "faceb00c", if you read the next part. Or "facebook", when you account for a bit of 1337 spelling and pronounce it phonetically.

Re:Funny/interesting addresses

[Anonymous Coward]

And, if you're paying slightly better attention, b00c is an approximation of "book".

Re:Funny/interesting addresses

[BrentH]

If you look really closely it even has a booc!

Re:Funny/interesting addresses

[xMrFishx]

it actually says "face:booc", as there's no K in hex. Luns.net says boobies though, which is more important.

Re:Funny/interesting addresses

[imbaczek]

i think i like b00b1e5 more.

Re:Funny/interesting addresses

[ceced]

Actually it has "Facebook" (face:b00c), Cisco has "Cisco Dog Food" (c:15c0:d06:f00d), lunk.net.uk has "Boobies" (b00b:1e5) and BBC has "bbc" :)

Re:Funny/interesting addresses

[Anonymous Coward]

Don't forget dog food.

Re:Funny/interesting addresses

[IAmGarethAdams]

Yes, that was the point of the parent post. Hence the CISCO/"cisco dog food", LUNS/"boobies" and BBC/"bbc1" examples too

Re:Funny/interesting addresses

[mykdavies]

Nicely spotted. In case anyone else missed the full thing as I did the first time round, Cisco's address ends c:15c0:d06:f00d = cisco dog food, very clever.

Re:Funny/interesting addresses

[rwa2]

Yeah, surprised Google didn't manage to get anything with "8008" in it. Maybe in IPv4space they already spent too much money buying 8.8.8.8 and 8.8.4.4 from L3 that they didn't want to spring extra for 8.0.0.8 as well.

Re:Funny/interesting addresses

[Fenris Ulf]

www.sprint.net has IPv6 address 2600::

Re:Funny/interesting addresses

[Anonymous Coward]

I like the BBC's effort too.

Little pressed to explain what b00bs have to do with LUNS though...

Re:Funny/interesting addresses

[Anonymous Coward]

b00bs are related to everything!

Re:Funny/interesting addresses

[Anonymous Coward]

Hey, they figure, why waste any opportunity. Maybe I'll make my home network start with the ascii code for my last name.

Re:Funny/interesting addresses

[InsertWittyNameHere]

lol b00b:1e5

Re:Funny/interesting addresses

[John Hasler]

They can put anything they want in the rightmost 64 bits. That's all part of their local address.

Re:Funny/interesting addresses

[antdude]

Is there a joke on "dog food"? Is Cisco eating their own dog food for IPv6?

Re:Funny/interesting addresses

[petermgreen]

I suspect google are actually big enough to get a /32 which means they can put whatever they like in the rightmost 96 bits. The BBC is probablly in a similar position.

Re:Funny/interesting addresses

[petermgreen]

Just checked and it seems I was wrong in the BBCs case, looks like bogons own the address and suballocated it to the BBC (i'd guess the allocation range was 2001:4b10:bbc::/48 ). I was rather surprised at this given the size of the BBC I expected them to have a direct allocation from RIPE. Maybe this shows that they are just dipping their toes in the IPv6 waters.

Re:Funny/interesting addresses

[John Hasler]

> Is Cisco eating their own dog food for IPv6?

The DoD told them they had to if they wanted to continue to bid on defense contracts.

Re:Funny/interesting addresses

[Anonymous Coward]

4b10 is allocated to RIPE - see here So RIPE apparently gave BBC 2001:4b10:bbc::/48

Unlike in the 70s, I don't see why Google or BBC need a /32. If their providers do what's standard and just use the 45 bit global routing prefix themselves, it will give both these orgs 16 bits for a possible 65536 subnets, and then 18 quadrillion addresses within each of those subnets. So the address that the BBC has is all they'll ever need.

In the 70s, when IPv4 addresses were being distributed like confetti, you had companies like IBM, DEC, HP et al being given public Class A addresses, which would have allowed them to have 16 million addresses. I doubt that throughout its history, any of them have had 16 million employees, devices and everything even worldwide, even though departed employees wouldn't have taken any w/ them. But this time, the IETF is pretty conservative about how it's distributed the addresses - only 2001::/16 has been given to the IANA so far, and from that, each RIR has been allocated a fair amount. And none of them will have any reason to run short, since every organization will need only one /48 global routing prefix, and from there, it would be good to go for 'ever'

When the word's population is 1 quadrillion, it will be time to start designing IPv7, assuming that other technologies haven't made IP totally redundant by then.

Re:Funny/interesting addresses

[Anonymous Coward]

2600::/16 is ARIN's, not Sprint's. ARIN probably gave Sprint something like 2600:x:x::/48, which Sprint can use.

If your name is less than 4 letters, and uses only A-F, Z, S, G, you should be okay. Just beat everybody else w/ your last name.

Re:Funny/interesting addresses

[Anonymous Coward]

I've seen a few already today!

www.facebook.com has IPv6 address 2620:0:1c18:0:face:b00c::
cisco.v6day.akadns.net has IPv6 address 2001:420:80:1:c:15c0:d06:f00d
www.luns.net.uk has IPv6 address 2a01:8900:0:1::b00b:1e5
www.bbc.net.uk has IPv6 address 2001:4b10:bbc::1

Does v6 kick off 'IP addresses as a marketing tool'? :)

Facebook could have tried getting from ARIN something like 2620:face:booc::/48. Right now, the way they have it set up, they can accommodate only 4 billion users b4 they'll need to change. Okay, they can later make it 2620:0:1c18:1:face:b00c:: for the next batch, and so on, and support another 274,873,712,640 once this runs out.

Still, would have been neat to have their name in the global routing prefix.

Re:Funny/interesting addresses

[Anonymous Coward]

Having said what I did above, I think Google could have tried to have something like 2001:0490:091e::

Only that in the 091e space, the leading 0 would have been dropped, as the v6 notation doesn't require that it be included: it's assumed.

Having such special address ranges (but only one such /48 range) guarantees a company or organization all it needs, while not horning into others' areas.

P.S. For Cisco's dog food, shouldn't dog have been c:15c0:do9:food:: instead of do6? Or does 9 represent q?

Re:Funny/interesting addresses

[ryzvonusef]

well.....

In punjabi, "lun" means "dick", so yeah...

I am *not* going to click that link :P

Re:Funny/interesting addresses

[bbn]

ftp://ftp.ripe.net/pub/stats/ripencc/membership/alloclist.txt

2001:4b10 is allocated to a company named Bogons Ltd, a UK based ISP. BBC is NOT owner of 2001:4b10:bbc::/48. There is no entry for that subnet in the whois database.

Curiously BBC _does_ have an allocation from RIPE. But it is 2001:41c0::/32.

% Information related to '2001:41c0::/32'

inet6num: 2001:41c0::/32
netname: UK-BBC-20041108
descr: British Broadcasting Corporation
country: GB ...
 

Re:Funny/interesting addresses

[petermgreen]

4b10 is allocated to RIPE - see here

Yes it is allocated to RIPE as part of the much larger block 2001:4A00::/23.

So RIPE apparently gave BBC 2001:4b10:bbc::/48

I see no evidence to back up this claim, whois clearly states that 2001:4b10::/32 is allocated to bogons limited. The allocation below that is not registered in whois but it seems most likely that bogons limited gave the BBC 2001:4b10:bbc::/48

But this time, the IETF is pretty conservative about how it's distributed the addresses

I've heard the opposite, for example free.fr got a /26 (64 times larger than the default ISP allocation of a /32) to support the highly address space inefficiant technology (at least in the form free deployed it in) known as 6rd.

http://ripe58.ripe.net/content/presentations/ipv6-free.pdf

only 2001::/16 has been given to the IANA so far [iana.org]

BS that page has no mention of 2001::/16 and indeed your first link already shows allocations to the RIRs outside that range.

since every organization will need only one /48 global routing prefix

/48 may seem like a lot but assuming standard sized subnets (nessacery for stateless autoconfiguration to work) it's only 65536 subnets, I could easilly see a large organisation exceeding that.

Re:Funny/interesting addresses

[Anonymous Coward]

4b10 is allocated to RIPE - see here

Yes it is allocated to RIPE as part of the much larger block 2001:4A00::/23.

So RIPE apparently gave BBC 2001:4b10:bbc::/48

I see no evidence to back up this claim, whois clearly states that 2001:4b10::/32 is allocated to bogons limited. The allocation below that is not registered in whois but it seems most likely that bogons limited gave the BBC 2001:4b10:bbc::/48

In that case, RIPE would have given bogons 2001:4b10::/32, and bogons in turn would have given BBC 2001:4b10:bbc::/48. In other words, it's hierarchical - IANA gives 2nd 2 bytes to RIPE, who then gives the next 2 bytes to BBC. Maybe the whois tables have not been updated?

But this time, the IETF is pretty conservative about how it's distributed the addresses

I've heard the opposite, for example free.fr got a /26 (64 times larger than the default ISP allocation of a /32) to support the highly address space inefficiant technology (at least in the form free deployed it in) known as 6rd.

http://ripe58.ripe.net/content/presentations/ipv6-free.pdf

What I meant above was that only 2000::/3 has been given to the IANA, and everything else has been reserved. In other words, IANA can't touch 4000::, 6000::, 8000::, until the IETF releases it.

only 2001::/16 has been given to the IANA so far [iana.org]

BS that page has no mention of 2001::/16 and indeed your first link already shows allocations to the RIRs outside that range.

Yeah, should have said 2000::/3.

since every organization will need only one /48 global routing prefix

/48 may seem like a lot but assuming standard sized subnets (nessacery for stateless autoconfiguration to work) it's only 65536 subnets, I could easilly see a large organisation exceeding that.

How? I'm thinking of some really large organizations, such as the government of China, or the UN, or the EU or the US government. Do you see them requiring 65536 routable networks, when any single network can house 18 quadrillion members?

I did think that instead of just 2 bytes for the subnets, they could have used 4, and that would have allowed for 4.3 billion networks - about as many unique v4 addresses. But that seems to me like a bit of overkill.

So what

[synapse7]

So all this proves is these sites are capable of running both protocols simultaneously and while there is a DNS record that resolves to an ipv6 address, is anybody able to browse to these sites using ipv6 all the way through?

Re:So what

[Anonymous Coward]

yes, I am

Re:So what

[Anonymous Coward]

One major concern is/was that returning IPv6 addresses during DNS lookup could break functionality for devices that can't handle IPv6.

I personally can't access anything via IPv6 on my school's connection. Will try again at home, see if Verizon turned on IPv6 for the day.

Re:So what

[synapse7]

That is good enough for me, I declare IPv6 day a success!

Re:So what

[squiggleslash]

Why not enable IPv6 and find out?

If you have a modern router, the chances are it already allows you to use IPv6 irrespective of your ISP's support for it. Check the IPv6 settings.

1. Enable "stateless configuration" for the network part (as opposed to DHCPv6 - you don't want that)
2. Enable 6to4 tunneling. If the system asks for a gateway IP address, leave it blank (Linksys routers do for some reason, but they route properly if it's absent.)
3. Enable IPv6 on your operating system. Wait until your computer shows that it has an IP address starting with "2002:", and then try to connect to "ipv6.google.com". If that works, start browsing! You'll use the IPv6 version of the sites you browse today because AAAA is generally routed first, and if there is a problem, you'll notice it with either timeouts/host-cannot-be-reached type errors, or huge delays loading pages.
4. If it didn't work, go to tunnelbroker.net, register for an IPv6 tunnel, and repeat from step 3, substituting using a tunnel instead of 6to4, and waiting until your computer has an IP address beginning with the prefix you get from HE.

Either 6to4, or the tunnel broker, should get you live, real, honest to goodness, IPv6 connectivity. Every computer on your network that has IPv6 enabled will suddenly have a real IP address, and a direct connection to the Internet. For those services that support it, there'll be no stupid tricks involving DMZs and port forwarding you have to do to get SIP, Bittorrent, online video games, or anything else working.

Re:So what

[synapse7]

I did, I can't, that is why I asked.

Re:So what

[Cimexus]

Yes ... no problems at all here (I have a native IPv6 connection, not tunnelled). This isn't really very surprising - these sites have mostly been resolvable and accessible via IPv6 for a long time through alternate domains (e.g. ipv6.google.com, www.v6.facebook.com). All that changed is that they have now published AAAA records for the ~main~ domains as well.

Re:So what

[John Hasler]

is anybody able to browse to these sites using ipv6 all the way through?

Yes. Those of us who have free IPv6 tunnels from sixxs.net or he.net, among others.

Re:So what

[squiggleslash]

Interesting. I have two 6to4 networks, and neither are having problems this morning. Here's a traceroute:

$ traceroute6 www.google.com
traceroute to www.google.com (2001:4860:b009::67), 30 hops max, 40 byte packets
1 2002:XXXX:XXXX::2 (2002:XXXX:XXXX::2) 4.888 ms 4.860 ms 4.849 ms
2 2002:c058:6301:: (2002:c058:6301::) 37.760 ms 38.721 ms 38.727 ms
3 ge-7-2-ur02.s3ndigital.ga.atlanta.comcast.net (2001:558:fe12:1::1) 38.756 ms * *
4 * * *
5 * * *
6 * * *
7 * pos-0-0-0-0-pe01.56marietta.ga.ibone.comcast.net (2001:558:0:f594::2) 60.528 ms *
8 2001:559::38e (2001:559::38e) 142.054 ms * 136.804 ms
9 * 2001:4860::1:0:489 (2001:4860::1:0:489) 124.818 ms *
10 2001:4860::1:0:613 (2001:4860::1:0:613) 48.391 ms 54.988 ms 55.037 ms
11 2001:4860::2:0:617 (2001:4860::2:0:617) 54.064 ms 55.936 ms *
12 * * *
13 qy-in-x67.1e100.net (2001:4860:b009::67) 47.046 ms 52.048 ms 47.097 ms

Telneting to port 80 works too. I just spoofed an HTTP request without problems.

What kinds of problems are you getting?

Re:So what

[AndrewNeo]

I'm guessing that they were testing the ipv6 on separate domains because if someone were using IPv6 and it were broken on their end, it might be hard to figure out what was wrong.

Re:So what

[tom17]

Yes

Re:So what

[Rising Ape]

Facebook doesn't really do IPv6. Bits of the site do, bits don't (notably the part responsible for delivering images). Try dropping IPv4 and seeing how well it still works. The BBC is the same.

Still, it's a start.

Re:So what

[RabidMonkey]

Another significant hurdle is that the TTL on records in the TLDs is generally 7 days, which makes backing out take a little longer than most people are comfortable with, should something bad go down.

Re:So what

[Chang]

The NS glue records in the .com TLD are 172,800 seconds (2 days) but that has nothing to do with IPv6 day.

The actual IPv6 DNS records being advertised today are things like www.cisco.com (TTL 30 seconds), www.cisco.com (TTL 300 seconds), www.google.com (TTL 300 seconds) so backing them out isn't a big deal or something that needs a lot of lead time.

Re:So what

[davew]

Nope. It's a scheduled, time limited way to identify end users whose browsers work normally when presented with sites that resolve to IPv4 only, but have problems when presented with sites that resolve to both IPv4 and IPv6. This is a fraction of one percent of users, but they're holding up the show for the rest of us. Without a day like today, they would never even be aware that something is wrong.

Re:So what

[Cimexus]

Ah interesting ... thanks for the info.

Yeah I have thought about turning off IPv4 and going IPv6 only just for kicks (i.e. to be amazed at how few sites I can access, and how badly broken the ones I can access are). But I never got around to it. Sounds like I'm not missing out on much.

It works!

[Anonymous Coward]

So far it works gre... [carrier lost]

meh

[Combatso]

4,294,967,296 should be enough addresses for any internet

Re:meh

[Anonymous Coward]

using NAT technology, only 4,294,967,296 ip addresses are used - but we can have more than 72,057,594,037,927,936 devices connected to the internet.

Re:meh

[Anonymous Coward]

IPv4 will probably still be used behind NAT routers, though. Why bother switching a LAN to IPv6?

Google logo

[Laser_47]

With Google pushing this so hard, why didn't they change the logo? They should have had one for the IPv6 crowd different from the IPv4....

Re:Google logo

[simoncpu was here]

I was hoping to see a dancing turtle or something...

Re:Google logo

[Nigel Stepp]

They used to have a dancing google logo (letters bounced every now and then) on ipv6.google.com. I was hoping they'd put that on the main page, but no.

It is EDT not EST

[grink]

Get your timezone right, it is currently EDT.

Re:It is EDT not EST

[Anonymous Coward]

Not all regions of the USA follow Daylight Savings Time.

Re:It is EDT not EST

[Anonymous Coward]

But then the interval should read 7pm-7pm EST, not 8pm-8pm. By the way not all such regions are in the USA. E.g. Coral Harbour.

Most-watched? PSN?

[iSzabo]

I'm pretty sure the PSN outage was way more watched.

hah..what

[Sase]

Im not sure about this because i havent been keeping tab... but i thought widespread nat'ing pushed ipv6 the way ofnthe dodo

I think arin needs to be a little less lax about their assignments btw.. you can get a /20 with two linksys routers and an ipad these days..

Re:hah..what

[jcurran]

I think arin needs to be a little less lax about their assignments btw.. you can get a /20 with two linksys routers and an ipad these days..

Sase - anyone may submit proposals to change ARIN's policies... Go to www.arin.net/policy for details on doing so.

Re:hah..what

[Anonymous Coward]

why would i do that when i could just complain on /. :-)

Re:hah..what

[John Hasler]

Im not sure about this because i havent been keeping tab...

This is clear.

but i thought widespread nat'ing pushed ipv6 the way ofnthe dodo

Fortunately, you are quite wrong. The horrific kludge that is LSNAT is such a PITA for ISPs that they will go to DS-Lite to avoid it, and nothing but IPv6 can handle the coming mobile explosion.

Next Internet Land Grab

[SloWave]

is already started. Look at Facebooks IPV6 address closely...

snark@toluene:~$ host www.facebook.com
www.facebook.com has address 69.171.224.39
www.facebook.com has IPv6 address 2620:0:1c00:0:face:b00c::

Re:Next Internet Land Grab

[squiggleslash]

It's great but not a land grab. It might have been had the address been face:b00c::1.

The first part of the address is the prefix. This is the part that's assigned by your ISP or ARIN. It's a 64 bit number, and is used to route packets over the public (ie ISP/trunked/etc) Internet. In the above, it's 2620:0:1c00:0:.

The other part of the address (the "face:b00c::" bit above), is a 64 bit number that's used to route packets within the part of the Internet owned by the person running the host (ie Facebook's Ethernet network.) These numbers are assigned by the network admin.

And that's what happened here. Usually the latter part of the IP address is based on the host's MAC address, but for some cases, specific services etc, they're often assigned manually, and Facebook has done this to assign an appropriate address to their server cluster. When you get your own network prefix (which you will when you get IPv6 on your DSL connection) you'll be able to do the same thing, even creating something that's {your prefix}:face:b00c if you want.

Re:Next Internet Land Grab

[Cimexus]

The last 64 bits of the address can be whatever you want, as the entire /64 space is allocated to you. So it can't really be a land-grab ... it's only the ~first~ 64 bits of the address that are unique and assignable.

Re:Next Internet Land Grab

[jrumney]

You too can get enough IPv6 addresses to make silly vanity addresses that noone other than a few geeks will see. Hardly a land grab.

Re:Next Internet Land Grab

[portnoy]

Exactly, not a land grab. On the other hand, you also have the BBC...

host -t AAAA www.bbc.co.uk
www.bbc.co.uk is an alias for www.bbc.net.uk.
www.bbc.net.uk has IPv6 address 2001:4b10:bbc::2

Re:Next Internet Land Grab

[Anonymous Coward]

This time, the IANA hasn't exactly allowed for land grabs, except that with everybody in the world owning a country the size of Russia, why exactly would they want it?

If you look @ the IANA assignments in the links above, the first 2 bytes that the IANA has for now is 2001::/16, second 2 bytes in the Global routing prefix has been delegated to the various RIRs, so that wouldn't have been available to Facebook, Google, the US government, or anybody, since it's from the RIRs. ARIN cannot give face::/16 to anybody, since the IETF has reserved it for future use.

However, in many cases, only the first 23 bits have been locked, but after that, anything can be assigned. So ARIN (or any other RIR) can assign something like 2001:04fa:ceb0::/48 to Facebook, and the latter can do what it likes w/ the rest of it.

All that said, if an organization is pretty happy putting its name in the Interface ID space, good for it!

Re:Next Internet Land Grab

[Anonymous Coward]

And that's what happened here. Usually the latter part of the IP address is based on the host's MAC address, but for some cases, specific services etc, they're often assigned manually, and Facebook has done this to assign an appropriate address to their server cluster. When you get your own network prefix (which you will when you get IPv6 on your DSL connection) you'll be able to do the same thing, even creating something that's {your prefix}:face:b00c if you want.

Forgot to add - the MAC address based EUI-64 assignment is one of the SLAAC assignments, but rarely recommended, as it would make the end user's MAC address visible to the rest of the world. Some other SLAAC algorithms would still use the MAC address, but use random interface identifiers instead of EUI-64 to generate it, so that the MAC address ain't traceable by any cracker. If DHCPv6 was needed (to detect the DNS servers), then the MAC address wouldn't be used at all in the Interface ID.

Re:Next Internet Land Grab

[squiggleslash]

Stateless configuration - which uses the MAC address and prefix to determine the IP, is the recommended way of configuring a network. It's simple, it's more reliable than a dynamic stateful protocol, and I don't think anyone can seriously say that a MAC address is a seriously useful tool for hackers.

DHCPv6 is not needed to detect DNS servers. RFC6106, which is implemented by virtually all route advertisement daemons I've come across, describes how to advertise DNS servers.

dnshat participating

[Anonymous Coward]

smcdonald@mcdesktop:~$ host www.dnshat.com
www.dnshat.com is an alias for www.dnsfailover.info.
www.dnsfailover.info has address 173.230.158.14
www.dnsfailover.info has IPv6 address 2600:3c01::f03c:91ff:fe93:b4b3

But can anyone with IPv6 tell me if its actually working lol ?

Re:dnshat participating

[mab]

marvin:~ mbradbury$ host www.ibm.com
www.ibm.com is an alias for www.ibm.com.cs186.net.
www.ibm.com.cs186.net has address 129.42.58.216

Re:dnshat participating

[mab]

marvin:~ mab$ ping6 www.dnsfailover.info
PING6(56=40+8+8 bytes) 2001:44b8:7871:7f0:72cd:60ff:fef2:dd04 --> 2600:3c01::f03c:91ff:fe93:b4b3
16 bytes from 2600:3c01::f03c:91ff:fe93:b4b3, icmp_seq=0 hlim=246 time=279.777 ms
16 bytes from 2600:3c01::f03c:91ff:fe93:b4b3, icmp_seq=1 hlim=246 time=279.627 ms
16 bytes from 2600:3c01::f03c:91ff:fe93:b4b3, icmp_seq=2 hlim=246 time=279.547 ms
16 bytes from 2600:3c01::f03c:91ff:fe93:b4b3, icmp_seq=3 hlim=246 time=278.873 ms
^C
--- www.dnsfailover.info ping6 statistics ---
4 packets transmitted, 4 packets received, 0.0% packet loss
round-trip min/avg/max/std-dev = 278.873/279.456/279.777/0.347 ms

Re:dnshat participating

[tom17]

~$ wget www.dnshat.com
--2011-06-08 11:54:17-- http://www.dnshat.com/
Resolving www.dnshat.com... 2600:3c01::f03c:91ff:fe93:b4b3, 173.230.158.14
Connecting to www.dnshat.com|2600:3c01::f03c:91ff:fe93:b4b3|:80... connected.
HTTP request sent, awaiting response... 200 OK

Nobody cares! Except maybe you.

[GreggBz]

I work at a sort of small ISP and we've done testing, implementation, published our website with an AAAA record and put some information on the site for everyone to see.

We've gotten exactly one call (this morning) on IPv6 that I can remember. We published information and started doing some obvious IPv6 things, but no one cares. The group of dual-stack test accounts is pretty small, but they have not even seemed to care or notice. I'd put anyone that asks on a list for testing so they can use IPv6 at home. No one has asked. I guess I could put a big(er) banner on the page.. but really I don't think it would matter much.. and probably scare people.

All in all I will say the experience has been pretty anti-climatic. It was not that difficult to implement. There were bugs of course, (Fedora 13+14 blocking DHCPv6 client traffic, and other NetworkManager bugs) the Cisco CMTS and it's weird detection of static IPv4 only clients... duplicate address detection madness, incomplete support of DHCPv6 + SLAAC in routers (D-Link DIR-615..) but it was just me working on it and I did not have that difficult a time getting our network to route, connect and answer to IPv6. Most of the problems I dealt with were incomparable hardware. Routers and DOCSIS 2.0 + IPv6 modems which are pretty much non existent with the exception of one EMTA I've tested. You have to shell out the bucks for a DOCSIS 3.0 modem evidentially.

Of the D-Link routers I've tested the DIR-825 is the star. It was dead easy to configure. DD-WRT and Open-WRT are not easy and probably there is no build for your router if it only has 4Mb of flash.

Re:Nobody cares! Except maybe you.

[Anonymous Coward]

I just bought a DOCSIS 2.0 modem. I thought about going 3.0, but realized that wide adoption of IPv6 was a ways a way and felt like saving a 30-40 bucks for the immediate. I figure in a couple of years when it becomes a pressing concern is when I will buy a new modem anyway. I know its not the most technologically conscious thing to do, but I only need the internet for slashdot and email.

People like me are part of the problem.

Re:Nobody cares! Except maybe you.

[Anonymous Coward]

All in all I will say the experience has been pretty anti-climatic.

I agree. It really contributes to global warming.

Re:Nobody cares! Except maybe you.

[foksoft]

It is not necessary to have all users switched to IPv6. What we need is to have websites available over IPv6 so users who don't have access to IPv4 can access them. Then we can slowly start migrating end-users to IPv6 without worrying about loosing functionality. And this is where today's experiment is aimed. It is test if transition can be done without hacks like naming website ipv6.example.com instead of www.example.com as we are used to with IPv4.

What you tell to your users to be interested in IPv6?
Do you tell them here it is and if you want, you can run it? Or you tell them that when using IPv6, they will get public IP and their skype connection will be better as it will not need to use public relays?
Yes it also depends on the fact whether you actualy give your users public IP address or not. Those who are already behind NAT know what I am talking about.

Re:Nobody cares! Except maybe you.

[Anonymous Coward]

Lack of complaints means today has been a success. The whole point of today is to shake out those people who have badly configured systems that try IPv6 first but do not have connectivity. The worry was that these people would hold back the adoption of IPv6. If you get no complaints that means that IPv6 can easily co-exist with IPv6 and people can work on spreading it.

Re:Nobody cares! Except maybe you.

[Anonymous Coward]

All in all I will say the experience has been pretty anti-climatic.

Exactly as it was intended to be. Yay!

Re:Nobody cares! Except maybe you.

[Hobart]

But - that's fine!

Remember, you're in the equivalent position to a traffic controller who just put in a new feature controlling the flow of morning traffic during everyone's drive to work. Your work, if done, right, doesn't show up to anyone at all.

What surprised me is that slashdot, hacker news, and reddit are all not participating, but FARK.com is running fine if you have ipv6-only.

Re:Nobody cares! Except maybe you.

[Anonymous Coward]

My DIR-655 has been doing a decent job as well. There's been a little bit of quirkiness, but the last firmware update added an IPv6 firewall setting and smoothed some things out. Can't wait til routers support IPv6 as well as IPv4 though.

Re:Nobody cares! Except maybe you.

[tokul]

they will get public IP and their skype connection will be better as it will not need to use public relays

I already have public IP address with IPv4.

Skype won't be any better if all incoming traffic is blocked by firewall.

Re:Nobody cares! Except maybe you.

[JanneM]

"I work at a sort of small ISP and we've done testing, implementation, published our website with an AAAA record and put some information on the site for everyone to see."

Not to disparage you or anything, but does _anybody_ ever go look at their ISPs website? I have never seen the site for our current ISP, nor did I ever visit the site for my previous one; that's a span of about ten years in total. In fact, it didn't even occur to me that they might have some information for me there until just now when I saw your post.

This is all to say that while I agree the end-user interest is (and should be) low, basing it on the reaction to info on your site might not gauge the interest that exists very well.

Re:Nobody cares! Except maybe you.

[thegarbz]

And why should they?

No really this is what I hate about the ISP's attitude towards this. I as the end user should not need to care that you as an industry is incapable of implementing workarounds to a problem you saw coming for 10 years. I do not care about your IPv6, or your BGP, or your how your WAN works. I care about being able to type www.slashdot.org in my browser and it appearing. I care about my computer accessing the internet.

I as a user care only about things that don't work. If NAT is the answer to my problems and you provide it then by all means. If it slightly degrades my service or affects the way I can use the internet THEN AND ONLY THEN will you hear from me. Then you will hear from me daily until the issue is fixed. Then you will lose customers if another ISP is capable of offering to me what you can't (assuming enough competition). It is then that your foresight may make or break your company.

4.3B IP's should be enough...

[jcurran]

7 Billion people on the planet... While many today do not have Internet connectivity, that's changing rapidly where some regions are skipping the copper deployment for end users and going directly to deployment of wireless infrastructure. In more established economies, it is not uncommon to have 1 IP address in use at home for broadband, one in the office, one on your mobile device, etc. 4 or 5 IP's per person, 7 Billion people = 40 or 50 Bill IP addresses would be helpful, and this doesn't even count servers in data centers, virtual machines, clouds, etc. 4.3 billion is looking very tight even with just today's applications.

Re:4.3B IP's should be enough...

[Combatso]

yeah... that was a joke, but thanks for playing... http://gates256.ytmnd.com/

Re:4.3B IP's should be enough...

[Anonymous Coward]

When IPv4 was put together, the world's population was already 4.5billion, so what were they thinking? To make things worse, companies were allocated Class A address ranges, allowing them to have 16 million addresses/networks. I'd argue that even for its time, IPv4 was shoddily done!

while Verizon and Comcast

[Burz]

...still have no IPv6 addresses on their main websites.

Re:while Verizon and Comcast

[MetalliQaZ]

As much as I like to bash Comcast, they actually have been on the ball when it comes to IPv6. They have been testing tunneling for their customers (I'm using their 6to4 to participate in IP6 Day), they have a IPv6 test suite for customers, and they are planning to go Dual-stack.

Slashdot?

[Anonymous Coward]

Slashdot is more interested in adding javascript for the sake of adding javascript, working towards the final goal of making the site completely useless without javascript, as well as completely useless WITH javascript since it now requires a small cluster to render within the hour.

(You may be laughing, but I'm certainly not. Slashdot jumped the shark somewhere around 2005 IMHO.)

Poor business reported at hotels in the area...

[cardpuncher]

... owing to the test site being accessible only by means of a long and winding tunnel housing an 8ft gauge railway on which massively long engines haul tiny carriages.

Eastern Standard Time?

[xquercus]

Sponsored by the Internet Society, World IPv6 Day runs from 8 p.m. EST Tuesday until 7:59 p.m. EST Wednesday.

Hrm. Why note the start and end times in Eastern Standard Time when the entirety of the eastern time zone is using Eastern Daylight time?

Re:Eastern Standard Time?

[jrumney]

And the entire world outside of East Coast USA would probably be more comfortable working out what it is in local time from 1:00am GMT Wednesday until 0:59am Thursday.

Re:Eastern Standard Time?

[John Hasler]

It's actually from 0:00:00 UTC Wednesday through 23:59:59 Wednesday.

The biggest problem with IPV6 ...

[Anonymous Coward]

Is that my ISP doesn't even provide it. Maybe late this fall thy say :-(

time range looks a little less arbitrary in GMT

[doubleyou]

Hey NetworkWorld, way to advertise that you're on the east coast of the US. Want to re-state that time-range in GMT now, so it actually makes sense?

dyndns.org

[tom17]

I have a host on dyndns.org and up until yesterday, the AAAA record came back, but today it says there is no AAAA record, even though it is still configured in my account on their site.

Anyone else found this problem today?

How to get wide IPV6 adoption in months not years.

[IpSo_]

Have Google modify their page rank algorithm to give any website accessible through IPV6 a slight boost. The power they hold over website revenues is so huge the SEO industry would go nuts over this and you'll see adoption rates explode.

IPv6 hall of shame (Please add more)

[WaffleMonster]

Some participants need to grow a clue by not activly working to turn IPv6 day into disaster day... Please add more...

1. Microsoft has a patch that demotes IPv6 access for one day only. Not only does this throw a wrench in the worlds ability to gauge problems but it does nothing to solve the end users issue. Paradoxically simply disabling IPv6 is much better at this point as not breaking IPv4 is much more important to the forward progress of IPv6 deployment than a few end-users who can enable IPv6 later when they can get their issues fixed.

2. NIST advertises an AAAA record for www.nist.gov but only the home page is accessable. All other content on the site presents a page not found error. It turns out this was not a mistake... Quoting via cut and paste... "Note: This top level web page has been setup to test IPv6 capabilities and to participate in World IPv6 Day on June 8, 2011. This IPv6 web page will be disabled after the end of World IPv6 Day. Links on this page do not work. This is a copy of the NIST website, www.nist.gov, and is only reachable using the IPv6 network protocol. To access the entire NIST website, you must use the IPv4 network protocol"

So you want to participate in IPv6 day in order to insure its failure. If you want the first page to be IPv6 reachable fine d00d...but don't break your site.. a global search and replace for hyperlinks to the IPv4 URL or simply including a fricking hyperlink to the IPv4 version... A lot of people will not even know they are using IPv6 or how to disable it or what you are even fricking talking about. How a webmaster can be so fricking clueless is beyond anything I'm capable of comprehending. It is the government so there is that.

3. For about half of IPv6 day level 3 was also advertising an AAAA record. Going to www.level3.com resulted in 404 not found. The entire site was down for anyone with IPv6. I can't believe a huge telecom could be so clueless.

Re:IPv6 hall of shame (Please add more)

[davew]

1. Microsoft has a patch that demotes IPv6 access for one day only. Not only does this throw a wrench in the worlds ability to gauge problems but it does nothing to solve the end users issue. Paradoxically simply disabling IPv6 is much better at this point as not breaking IPv4 is much more important to the forward progress of IPv6 deployment than a few end-users who can enable IPv6 later when they can get their issues fixed.

I think this is the support item you're referring to.

I did at first think the same way, but then I realised - that doesn't appear to be an automatically-pushed patch. It looks like a support article to which an admin can refer a user who is screaming "I don't care, make my internet work NOW." It's something that can be applied in a hurry to temporarily resolve the problem, but doesn't sweep it under the carpet because the underlying problem will still need to be dealt with in time. In that context, I think that this is a more responsible approach than telling users to disable IPv6 permanently.

Re:IPv6 hall of shame (Please add more)

[Anonymous Coward]

I can't believe a huge telecom could be so clueless.

You clearly have no clue about telecom.

Re:IPv6 hall of shame (Please add more)

[WaffleMonster]

I did at first think the same way, but then I realised - that doesn't appear to be an automatically-pushed patch. It looks like a support article to which an admin can refer a user who is screaming "I don't care, make my internet work NOW." It's something that can be applied in a hurry to temporarily resolve the problem, but doesn't sweep it under the carpet because the underlying problem will still need to be dealt with in time. In that context, I think that this is a more responsible approach than telling users to disable IPv6 permanently

I strongly disagree. There was never any indication this would have been pushed out automatically and every reason to assume it would not as such a change would stand a good chance of being disruptive to existing deployments.

The main problem holding up widespread deployment on the content side are broken clients. This is a P1 issue which far outweighs a small subset of end users ability to use IPv6.

Either fixing the problem or turning off IPv6 are the only correct responses in my narrow short-sighted view.

Re:IPv6 hall of shame (Please add more)

[Anonymous Coward]

Some participants need to grow a clue by not activly working to turn IPv6 day into disaster day... Please add more...

2. NIST advertises an AAAA record for www.nist.gov but only the home page is accessable. All other content on the site presents a page not found error. It turns out this was not a mistake... Quoting via cut and paste... "Note: This top level web page has been setup to test IPv6 capabilities and to participate in World IPv6 Day on June 8, 2011. This IPv6 web page will be disabled after the end of World IPv6 Day. Links on this page do not work. This is a copy of the NIST website, www.nist.gov, and is only reachable using the IPv6 network protocol. To access the entire NIST website, you must use the IPv4 network protocol"

So you want to participate in IPv6 day in order to insure its failure. If you want the first page to be IPv6 reachable fine d00d...but don't break your site.. a global search and replace for hyperlinks to the IPv4 URL or simply including a fricking hyperlink to the IPv4 version... A lot of people will not even know they are using IPv6 or how to disable it or what you are even fricking talking about. How a webmaster can be so fricking clueless is beyond anything I'm capable of comprehending. It is the government so there is that.

That's in a way funny, given that the government is the one that had some aggressive mandates about IPv6 deployment.

Yesterday's limited goals were fine, but they should next have a world IPv4less day, where this time, for a day, they turn off IPv4, and then see how much of Internet traffic, runs, or can run. Repeat this exercise after, say, a quarter, and note improvement levels, if any. Since March of next year is the deadline for a lot of entities to be IPv6 enables, such tests should be carried out this 2-3 times this year, so that everybody in the internet food chain does what it takes to make all their networking equipment v6 compatible.

Those who have to have NAT for reasons like abstracting their network from the public domain should just stay w/ IPv4: once the bulk of users have moved, them staying on should not be a problem on dual stack equipment. Although I don't see justifications for future router models to support IPv4, just as i don't see justifications for applications and equipment today to support Windows 95 based OSs. But I certainly am against creating a NAT666 or a NAT66 when it's more likely to be a problem rather than a solution.

Happlily enjoying IPv6 on my network

[Fez]

[Disclaimer: I am a pfSense developer, so I'm a bit biased. For those of you who don't know what pfSense is, it's a BSD-based firewall distribution.]

pfSense 2.0 won't officially support IPv6, but there is a branch available that does IPv6 which will later become 2.1. I'm running it on my home router with a GIF tunnel to Hurricane Electric ( http://he.net/ http://tunnelbroker.net/) to get IPv6 even though my ISPs do not have any native IPv6 support yet. The IPv6 support is a work in progress but is complete enough that it will do what most people want/need.

Instructions for the setup and more info can be found on the pfSense IPv6 board here: http://forum.pfsense.org/index.php/board,52.0.html

I get a 10/10 on the IPv6 tests from http://test-ipv6.com/ on all my PCs as well as my Droid X running 2.3.3. If you're already using pfSense 2.0, give the IPv6 code a try, setup a tunnel to he.net, and enjoy. Doesn't take too long at all to setup.

Why just one day?

[h890231398021]

Why not leave the IPv6 support on? Is there some reason the IPv6 support developed and enabled for World IPv6 Day needs to be disabled tomorrow?

Re:Why just one day?

[John Hasler]

All of these sites have IPv6 support. However, they have their nameservers configured to return only A records except when queried with special names such as ipv6.google.com (in the case of Google it is slightly more complicated). The reason for this is that there are some buggy end-user systems out there that are on IPv4-only networks and will hang if they get a DNS reponse that includes both A and AAAA records. During this test these sites have configured their nameservers to return both A and AAAA records, in order to determine how many such systems there are (and to smoke out any other problems). If there turn out to be very few (1 in 1,000,000, say) they will soon make the change permanent and many other sites will follow. Otherwise, we have work to do.

Mod parent up.

[John Hasler]

n/t

Why only HTTP servers?

[belphegore]

$ host -t mx gmail.com
gmail.com mail is handled by 5 gmail-smtp-in.l.google.com.
gmail.com mail is handled by 10 alt1.gmail-smtp-in.l.google.com.
gmail.com mail is handled by 20 alt2.gmail-smtp-in.l.google.com.
gmail.com mail is handled by 30 alt3.gmail-smtp-in.l.google.com.
gmail.com mail is handled by 40 alt4.gmail-smtp-in.l.google.com.
$ host gmail-smtp-in.l.google.com.
gmail-smtp-in.l.google.com has address 72.14.213.27$ host -t mx cisco.com
cisco.com mail is handled by 10 sj-inbound-a.cisco.com.
cisco.com mail is handled by 10 sj-inbound-b.cisco.com.
cisco.com mail is handled by 10 sj-inbound-c.cisco.com.
cisco.com mail is handled by 10 sj-inbound-d.cisco.com.
cisco.com mail is handled by 10 sj-inbound-e.cisco.com.
cisco.com mail is handled by 10 sj-inbound-f.cisco.com.
cisco.com mail is handled by 15 rtp-mx-01.cisco.com.
cisco.com mail is handled by 20 ams-inbound-a.cisco.com.
$ host sj-inbound-a.cisco.com
sj-inbound-a.cisco.com has address 128.107.234.204
$ ...etc

Re:Why only HTTP servers?

[atomic-penguin]

I'd say for most organizations, the public-facing corporate website breaking is less of a big deal than if all e-mail routing ground to a halt for a day. Not that this was likely to happen today.

HTTP is also an easier IP application to troubleshoot than say SMTP, DNS, or even and routing at layer 3. When troubleshooting effectively, you make small changes then observe the effects of the change. Which is really the next reason not to test SMTP on IPv6 today...

Participants aren't just testing HTTP servers today. Someone also had to keep a close eye on DNS infrastructure, and network layer 3. I have seen Cyrus IMAP Murder clusters failing to replicate with link-local IPv6 turned on in DNS and on the IMAP servers, while IPv6 was disabled at network layer 3. That tends to back up mail routing real quick, when something like IMAP services do not function as expected. Its nearly impossible to troubleshoot a situation like that, mail routing is backed up, DNS queries all seem to work, and then you have Cyrus spewing weird incomprehensible error messages which might lead you on a red herring troubleshooting hunt. I am sure most participants do not want a shit-storm to deal with today, just by throwing mail services into the ring.

So our own organization turned on IPv6 at layer 3 on a few isolated VLANs a couple months ago to test everything out-of-band in a lab environment. We learned a few lessons like what exactly NDP (Network Discovery Protocol) does in IPv6, and how to firewall an IPv6 Linux server. The ICMPv6 rules are drastically different than the equivalent on IPv4, because you're supplanting ARP with NDP for the most part. Gradually we turned on IPv6 in the production DMZ VLAN. Then turned IPv6 on for one external DNS server and the corporate website, observed the effects for a day and made adjustments. Then finally turned IPv6 on for the remaining external DNS servers. At which time, it was discovered our TLD doesn't fully support IPv6 DNS glue yet, despite them being a fairly early adopter of technologies like IPv6 and DNSSEC.

Today was about testing the waters by sticking a toe in, not diving in head first to a pool with only 3 inches of water. Events like today's puts pressure on hardware vendors, major ISPs, and application vendors. It would be great to be able to dump some network stats for my IPv6 interfaces on the DNS boxes, although our network monitoring systems don't quite fully support IPv6 yet. There really isn't a good way to differentiate DNS or interface stats between IPv4 and IPv6, yet.

I kept some pretty thorough notes on IPv6 Linux configuration for anyone who hasn't had a chance to play with IPv6, yet, link here.

Re:How to get wide IPV6 adoption in months not yea

[davew]

Google will not make changes to pagerank that are not specifically about improving the quality of search results.

This has been thought of before. :-)

ASCII art too

[tepples]

Anonymous Coward wrote:

There are relatively few control characters in Unicode. It'd make far more sense to use a blacklist.

For one thing, a new version of Unicode may add more control characters. For another, a lot of characters are more useful for sneaking "ASCII art" (pardon the misnomer) pass the existing lameness filters than for conveying meaning that most readers of Slashdot comments would understand.

Re:How to get wide IPV6 adoption in months not yea

[IpSo_]

If I was an IPV6 only user, I would want IPV6 pages to rank higher than IPV4 pages which I wouldn't be able to access anyway.

Though, Google may already be doing that...

Re:How to get wide IPV6 adoption in months not yea

[davew]

I reckon if you were an IPv6 only user, what you'd want to see is a list of pages you can access, and not ones you can't. That's a matter of filtering for the user, not sorting for relevance to the search query. And that assumes the existence of an IPv6 only user with *no* access of any kind to the IPv4 internet. We've a long distance to go before we start seeing those in the wild, outside of labs.

I think we need to face it that we can't expect Google to damage their core product by introducing changes like this for even the best of technical intentions. There isn't any "How to get IPv6 adoption in months not years." There's a lot of work to be done in crafting proper plans with realistic costs and benefits that can be understood by the people who are going to approve the money. We can do little things here and there, but we can't short-circut that process on an industry-wide basis.

It sounds daunting, but it's doable if we chew one bite's worth at a time. What's happening today is going to contribute to that for the content providers, by quantifying something that was previously uncertain: just how big is the impact on existing users if you dual stack. If the day turns out to be so successful that some big sites dual stack permanently - as such experiments in the past have done - then that contributes to the case for the rest of us, because finally there will be some real content out there that will use the stuff we're paying for.

Linux just works

[thsths]

The funny thing is that Linux just works - it picks up IPv6 from the DHCP server, and Firefox and Google Chrome just chose the right protocol (although I still think that IPv4 should be the default - at least for now). At home I need to use a tunnel - neither the router nor the ISP provide native support for IPv6, but again it just works.

Windows XP (we skipped Vista, and 7 is still being rolled out) however is unable to deal with IPv6 correctly. In a year that problem should be history for us, but I am sure many many companies have the same issue with Windows. Until Windows XP is truly dead, IPv6 will remain a niche protocol.

IPv6 test

[rtyhurst]

There's a fairly useful test at:

http://test-ipv6.comcast.net/

to see if your system is IPv6 compatible.

Its bigger than that

[Anonymous Coward]

In spite of all the lies told by IPv6 promoters, One of the reasons that World IPv6 Day was such a widely ignored non-event is that the IPv4 address space is much bigger than 4,294,967,296 addresses.

Given that you can NAT a routable IPv4 address to a whole class B non-routable internal network, the REAL size of the IPv4 space is about 2^32 * 2^16, or 281,474,976,710,656 addresses. Which is enough to give everyone on earth about 32000 IPv4 addresses.

IPv6 doesn't include any provision for isochronous delivery of voice or video, it breaks IDS with layer 3 encryption, and it is deliberately lacking backwards compatibility with IPv4.

What sort of uber-nerd would design such a HORRIBLE protocol, then actually expect people to use it. Talk about out of touch.

IPv6 is unwanted , unneeded, and just plain bad. Stop flogging the dead horse and Let it die.

Re:Its bigger than that

[Anonymous Coward]

NAT is one of the major reasons to move away from v4 to v6. What's worse is that there are several layers of NAT that are used, which just exacerbates the problem.

It would take at least 4 layers of NAT to deliver what IPv6 is capable of, and resolving them would be a nightmare. Also, what is the above calculation based on? Behind NAT, you can only have 10.x.x.x (Class A), 172.(16-31).x.x. (Class B) and 192.168.x.x(Class C). Those are the only ones you can have behind NAT layers.

Re:Its bigger than that

[Anonymous Coward]

Given that you can NAT a routable IPv4 address to a whole class B non-routable internal network, the REAL size of the IPv4 space is about 2^32 * 2^16, or 281,474,976,710,656 addresses. Which is enough to give everyone on earth about 32000 IPv4 addresses.

Uh, no!!! You're assuming that all IPv4 addresses are equitably distributed across the world. However, there are theoretically ~5 addresses for every American, ~2for every Canadian and Australian, ~1 for every Japanese, Korean, Russian, Brit, 1 for every 6 Chinese, 1 for every 50 Indians, and so on.

And even that 1 for every ~5 Americans assumes that a good number of Americans are employees of the Pentagon, GE, AT&T, Xerox, Level 3, HP, IBM, Google, and a large number of organizations that had Class A routable addresses. Which is not even true!

So essentially, for a lot of your cases, networks would have to implement >2 levels of NAT to accommodate all their requirements. The look-up tables for such clients after so many layers of NAT routing would be unwieldy.

Not to mention that a lot of new devices need peer to peer networking, where NAT is an unacceptable proposal. While some things, like having your garage or car w/ its own IP address may seem far-fetched today, having unique IP addresses for every handset ain't!

This is kind of a dumb response

[sean.peters]

No, not all of them do. But all of the Eastern Time portion of the US does (since 2005, even including Indiana). It's sort of irrelevant to this discussion what Arizona does with respect to daylight time.

Re:How to get wide IPV6 adoption in months not yea

[JanneM]

Well, Google already alters the sort order depending on where you are. No principal difference to sorting it according to accessibility on your part.

Re:How to get wide IPV6 adoption in months not yea

[Anonymous Coward]

Comes down to two interesting questions:

1. How much is the adoption of IPv6 in months not years worth to Google's business?
2. How much is the market share Google would lose, due to loss of trust in search results because of reduced quality, either actual or perceived, worth?

I'm willing to consider that the first is actually quite a lot in absolute terms, as they are well placed to take advantage of the online platforms that are clogging up the IPv4 space, and the faster these roll out the less time weaker contenders have a chance to respond. Maybe it would get into the tens of millions or even hundreds of millions a year.

But I'd bet that the effect on market share would be at least of the order of a percent, which is knocking on the billions of dollars per year. I don't think it comes close to being a feasible move for Google to make.

So what did they conclude from IPv6 day?

[Anonymous Coward]

Did the dual stack equipment work as expected? Could all IPv6 sites be smoothly accessed from IPv6 clients? Did net traffic increase, or stay the same? Were there too many 'broken' implementations of IPv6? And all that?

Also, would this result in an accelerated migration to IPv6 if everything turned out okay?

Re:So what did they conclude from IPv6 day?

[jcurran]

( Here's an example of lessons learned - this message from Facebook Engineering http://www.facebook.com/note.php?note_id=10150198443513920 )

World IPv6 Day came to an end earlier today. We successfully enabled IPv6 on our site for 24 hours, with great results. We saw over 1 million users reach us over IPv6.

We&rsquo;re pleased that we did not see any increase in the number of users seeking help from our Help Center. The estimated 0.03% of users who may have been affected would have experienced slow page loads during the test.

Based on the encouraging results, we&rsquo;ve decided to leave our Developer site dual-stacked, supporting both IPv4 and IPv6. And we will continue to adapt our entire code base and tools to support IPv6.

We are glad to have joined with the Internet Society, major Web companies, and other industry players to enable IPv6 for this test day. It was a great opportunity to test our infrastructure and IPv6 readiness.

IPv6 is vital to the continued growth of the Internet, and World IPv6 Day was a great step in the advancement of the protocol.  We hope the overall success of the 24 hour test will encourage others in the industry to establish reliable IPv6 connectivity and develop robust IPv6 products.

use v4 for NAT, v6 to escape NAT

[Anonymous Coward]

Since things will be dual-stack anyway for a while, I don't see why anybody should bother trying to retrofit NAT into IPv6 as some have suggested, given that the main selling point of IPv6 (other than addresses) is no NAT. Which eliminates a whole slew of problems.

For those whose main issue is address space, getting rid of NAT should not be an issue. However, if people are using NAT to isolate a LAN from the public internet, just stay w/ IPv4, instead of trying to retrofit NAT into IPv6.

In the end, the only people who should use IPv4 are those who must have NAT. For others, just make the move to IPv6. Also, at some point, make all routable IPv4 addresses Class C (irrespective of ranges), so that nobody can have >65,536 public addresses for their network. Should not be an issue, since most people who NAT use the 192.168 addresses anyway.

But if NAT is not needed, or even worse, unwanted, just use IPv6. If it's a LAN, use local IPv6 addresses (fc00::/7)

Dial...

[Chuby007]

at least now I won't dial my IP addy instead of my phone number...

Re:How to get wide IPV6 adoption in months not yea

[Anonymous Coward]

This cannot be Google only - all the search engines - Google, Bing, Yahoo, Facebook, Twitter, et al would need to co-operate here to deliberately promote their IPv6 based services at the expense of IPv4. The last thing you want here is Microsoft or Facebook indulging in one-upsmanship games w/ each other.

A few questions re: IPv6 implementations

[Anonymous Coward]

There are a few things in IPv6 that i haven't been able to figure out. Part of it is the role that local addresses in IPv4 plays in this, and how that gets translated when we go to IPv6.

• Wireless routers: In v4, the wireless router has a gateway address which is a local address, like 192.168.2.1 (in case of Belkin), and so any PC or laptop connected to it - be it through an ethernet port, or wireless, gets assigned a local address like 192.168.2.12 and is off to the races

  How is this done in v6? Is it by using local IPv6 addresses, like fc00::/7 Or is it treated as a router, and given an address like ff0_::2? And if it's the latter, how are networks created? Or are the nodes part of the parent network as assigned by the ISP, and to have public 2001::/64 addresses?

• Wireless access points: In an office, there may be several WAPs evenly distributed. What sort of addresses do they have?
• VPNs: Since a VPN is the connection of 2 networks into one private network, how is it implemented? In particular, how would a remote client be recognized wrt the network s/he is connecting to?
• Assignment of IPs: Let's say I got the 2^80 addresses (including the subnets) that one gets from my ISP, say Comcast, after the 48bit routable prefix is assigned. I assign addresses to my laptop, my IPAD, my home wireless router and so on.

  Then we get to my cellphone, whose service is from Verizon, who'd probably assign one of theirs to that phone. Question: how would i make all my devices belong to a single network?

• Rollover of IPs while changing ISPs: Each ISP gets a range of routable addresses, be it v4 or v6, and assigns one to a customer. Up to now, w/ v4, rollover hasn't been an issue, since the addresses have more often than not been dynamically assigned local addresses behind NAT. But for IPv6, lets say an ISP gives a customer his address range, and the customer assigns a number of them to various devices, or a network, or whatever.

  At some point, the customer is unhappy w/ the ISP and terminates the service - does the ISP or the customer have to part w/ that address? If it's the ISP, then how does the new ISP add that IP family to its network? If it's the customer, wouldn't that be a major inconvenience, particularly if a lot of devices are mapped, and doesn't it make consumers captive to their ISPs?

• Association of IPs w/ phone# or IMEA: Would an address be tied to a phone#, like (650)555-2468, or to the IMEI# of the handset (not applicable for CDMA handsets) and its CDMA equivalent, if any. Could the latter enable a transfer b/w 2 handsets even if one is inactive as a phone? Or would such a number only be tied to the phone#? Also, will it be architecturally allowed for any device to have >1 address, if the phone has to have an IP of the carrier, but I'd also want it to be a part of 'my' network?
• Equivalent of network address and gateway addresses: In IPv4, it's something like 200.43.59.0 for a network address, and 200.43.59.255 for a gateway. What is it in IPv6? Would 2001:0457:aced::0 be a valid node address, or not? Is the /64 a mandatory part of the address, so as to state the unalterable bits?

Okay, that was quite a few questions that i'd like to see answered, since I've not seen any discussion on that anywhere.

Re:A few questions re: IPv6 implementations

[Anonymous Coward]

One more I forgot above -Virtualization. We know that in VMs, resources, including network interfaces are replicated. Assuming that IPv6 addresses are assigned to multiple instances of a connection, which ones would be used? Public ones? Private ones? Which ranges?

Re:A few questions re: IPv6 implementations

[EsbenMoseHansen]

Note, I am no expert, but..

Re Wireless routers: They would be treated as any other routers. I.e, either they are brigding, in the which case they just forward to/from some other router, or they are a router unto itself and broadcast a proper (probably /64) network.

Access points: There is no change here, except that they have real, globally routable addresses

Assignment of IPs. I'm not sure I understand the question. Perhaps the answer is that you can have several IP addresses on one network device.

Re rollover: You don't get to keep your address. Just change your DNS entries already -- this will be the big pain. You don't have to remap your internal network, as they will just pick up the new network prefix from the router advertisments.

Phones: That is outside what I know about. Perhaps there is a mobile IPv6 answer somewhere.

There is, I believe, no broadcast address in IPv6, and I never did figure out what the network addresses were about in IPv4. So I guess I am not helping.

Re:A few questions re: IPv6 implementations

[EsbenMoseHansen]

I'd guess public ones, probably one per VM.

You seem to be very fond of private IP addresses. Really, with IPv6, there is little need for such.

Re:A few questions re: IPv6 implementations

[Anonymous Coward]

Well, there are times when one would want to have a network contained within an intranet or VPN, and not exposed to the outside world, and not routable. So private addresses will be needed, and have nothing to do w/ the scarcity of addresses in v4. The IETF has defined fc00::/7 for such purposes, which should not be publicly routable. Yet, for some bizarre reason, they've included making them unique worldwide, when there's no reason to do it.

Note that there shouldn't be any NAT lilke mapping of public to private addresses: the only time private addresses should be used is when the network is really an isolated LAN, and the sys-adm doesn't need or want to deal w/ their RIRs or local ISPs in getting addresses.

Re:A few questions re: IPv6 implementations

[Anonymous Coward]

By assignment of IPs, I meant do they get the global unicast addresses, like in the 2000 range, or link-local unicast, or multicast addresses? See this table and you'll understand what I'm asking.

For wireless routers, thinking about it some more, I'm guessing that a wireless router would get its global routable address from an ISP, use the subnet bytes to get a network address of its own, and then the interface part of the addresses would just be any clients hooking on to it.

In IPv6, there are no broadcast addresses since that would bring things to a crawl, if you tried broadcasting to 2^64 i.e. 18 quadrillion addresses. In other words, broadcast just doesn't scale when going from v4 to v6, which is why it was dropped. In v4, network addresses were used in routers to help route traffic - you'd use them mainly w/ the routing protocols. So here, I'm guessing that saying something like 2001:0459:de11:0001::/64 would define a network? In other words, are all zeros in the interface ID excluded from usage?

Actually, your post was a great help.

Re:A few questions re: IPv6 implementations

[Anonymous Coward]

Correction: By assignment of IPs, I meant that whose IP does a device accept? Your answer of maybe it can have multiple IPs might work - equivalent to a PC having 2 or more NIC cards, each w/ its own IP address.