[AusNOG] [Ap-ipv6tf] official shutdown date for IPv4. The date he is pushing for is April 4, 2024. "IPv4 can't go on forever, " Latour said. "

Mark ZZZ Smith markzzzsmith at yahoo.com.au
Fri Nov 7 20:45:49 EST 2014





----- Original Message -----
> From: Jonathan Thorpe <jthorpe at Conexim.com.au>
> To: 
> Cc: "ausnog at lists.ausnog.net" <ausnog at lists.ausnog.net>
> Sent: Friday, 7 November 2014, 18:59
> Subject: Re: [AusNOG] [Ap-ipv6tf] official shutdown date for IPv4. The date he is pushing for is April 4, 2024. "IPv4 can't go on forever, " Latour said. "
> 
> I was just thinking - I've got loads of equipment in my lounge room alone 
> which doesn't support IPv6 - all of which is only a couple of years old:
> 
> * Receiver
> * PVR
> * Samsung Smart TV
> 
> Theoretically all of these *could* support IPv6 (the last two being Linux based) 
> with a firmware update, but there are a lot of embedded electronics like the 
> first that aren't. Manufacturers aren't interested.
> 
> This seems to be a huge chicken/egg problem. The manufacturers know that users 
> will have IPv4 and unlikely to have IPv6.
> 

Some of the manufacturers IPv6 enable their entertainment devices.

Google's Chromecast will use IPv6 for the content if it is available over IPv6, although the discovery of the device on your LAN is only occurring over IPv4 (or it was, when I looked just after I got it).

I'd be pretty sure their upcoming Google Player will also use IPv6 if it is available.

Microsoft's Xbox One uses IPv6 natively, and will also tunnel it over IPv4 if it isn't i.e., IPv6 is the protocol used for gaming traffic, regardless of whether the network supports native IPv6 or it needs to be tunnelled over IPv4.



> The lack of availability of IPv6 over here is astonishing.
> 
> I've always thought that IPv6 (dual-stack) should be mandated on new NBN 
> services and enabled by default (users can always turn it off).

The NBN is a layer 2 network, whether to provide IPv6 or not is the choice the RSPs make.

> That would 
> probably force the vast majority of ISPs lagging behind on deploying it to start 
> offering it.
> 

Unlikely, they would probably either object, or just want money from the government to do it (as the TV stations did to cut over to digital TV only. At least in that case the government could recoup some of the money they gave the TV stations by selling the freed up spectrum. There won't be much of a market for left over IPv4 if the government paid ISPs to free it up.). 

It seems to me that most humans only react to problems when they can foresee experiencing the consequences in the relatively near future. Y2K is the example - the problem was recognised and talked about many years before most of the mitigation work started to take place. It seemed to only ramp up about 2 to 1.5 years before actual Y2K. 

So it seems 1.5 to two years out is when people will start taking actions. The trouble with IPv6 compared to Y2K is that Y2K clearly had a known deadline, where as IPv6 doesn't. I'd have much rather deployed IPv6 too early than be trying to deploy it 'too late' and in a rush.

> Kind Regards,

> Jonathan
> 
> 
> -----Original Message-----
> From: AusNOG [mailto:ausnog-bounces at lists.ausnog.net] On Behalf Of Ross Wheeler
> Sent: Friday, 7 November 2014 5:32 PM
> To: Mark ZZZ Smith
> Cc: ausnog at lists.ausnog.net
> Subject: Re: [AusNOG] [Ap-ipv6tf] official shutdown date for IPv4. The date he 
> is pushing for is April 4, 2024. "IPv4 can't go on forever, " 
> Latour said. "
> 
> 
> 
>>  What sort of equipment? A lot of networking equipment has been able to 
>>  do IPv6 for at least decade, it is just a matter of switching it on.
> 
> I know of a bunch of equipment - much of it no longer supported - with
> IPv4 *ONLY* networking. Quite a few network-enabled printers for example.
> Lots of CURRENTLY MADE equipment - for example, the solar MPPT devices I use 
> here - don't support IPv6 in any form, and are not likely to.
> Lots of PLCs. I know of many PVR/DVR devices without V6 support. Webcams. 
> CBus network interfaces (they may have now, but all the ones I've ever seen 
> don't, and replacing them is far from an "inexpensive upgrade").
> 
> 
>>  Well, if they'll have to place a value on accessing IPv6 only content. 
>>  If that value is below what the replacement cost of their equipment 
>>  is, they'll have to live without the IPv6 only content. At some point 
>>  it won't be possible to make it available over IPv4, so they'll 
> need 
>>  to choose what is more important to them.
> 
> I guess THAT will depend on what IPv6-only content there is.
> I can't imagine any of the "mainstream" sites dropping IPv4 any 
> time soon, because they don't want to lose the existing IPv4-only users. 
> Catch-22.
> 
> 
>>  If you haven't been looking, how can you have seen anything? Absence of 
> 
>>  evidence is not evidence of absence, in particular if you haven't been 
>>  looking for evidence!
> 
> Haven't been looking <implied> in earnest or specifically for 
> </implied>
> 
> And yes, absence of evidence is certainly not evidence of absence!
> 
> 
>>  When I occasionally visit Office works I have a look at the spec's of 
>>  various CPE, and IPv6 support is starting to show up fairly often.
> 
> "IPv6 support" isn't what I was asking though.
> I was asking for a "stepping-stone" for users who may have a 
> significant 
> (to them) outlay in their "home computer network" or "small 
> office 
> network" - perhaps network storage, media-centre, security systems, 
> possibly even VoIP devices - that may not all support IPv6.
> 
> Those in multi-billion-dollar companies who just chalk up a few percent of 
> revenue to upgrading, or who make a "multi-million-dollar commitment" 
> to 
> upgrading are all well and good, but as network operators and/or ISPs, 
> we're "middle-men". The people at the other end of the string - 
> those at 
> home and in small offices particularly (because they're the ones for whom 
> even modest additional expense will hurt most) also need to be considered 
> in this whole "upgrade process".
> 
> 
>>  Here's what appears to be the cheapest Netgear ADSL router, and IPv6 is 
> 
>>  listed, and seems to retail at around $60:
> 
> Yes, IPv6 support, but no indication if it is capable of doing a V6/V4 
> "translation" for legacy devices. I know a change to IPv6 should be a 
> complete change, but some people simply won't be ABLE TO. Or, will have to 
> sacrafice a bunch of capabilities in order to. Yes, I'm sure a modern 
> computer is quite capable of running an IPv6 stack AND an IPv4 stack 
> simultaneously, and to access local devices on the network via whichever 
> protocol is required... but (lets just play devils advocate here for a 
> minute)...
> 
> Take the case of my neighbour. Had major problems with breakins, installed 
> a 20-camera security system with a fancy digital video recorder and 
> internet access. He got a unit. Chinese. No support, company since 
> changed to making whatever. Cost him more than he can afford to replace. 
> It's only IPv4. Sure, he can access it from his computer on his local 
> network, but when he needs it most is when he's away. Right now, he can 
> connect to his home internet connection and see what's going on. His 
> recorder can call out (email) to notify him of an alarm, and send a 
> picture. It needs internet access periodically to set the time etc.
> How's this scenario work with the above modem (or any of the others 
> you've 
> generously listed)?
> He's a builder, not a network guru. All these "IP address" 
> settings he had 
> to fill in to get it to work required assistance, but at least he got it 
> going. I'm not even sure it's POSSIBLE with an "IPv6 router" 
> with IPv4 
> devices on the LAN side, is it? (Please tell me it "just works, and the 
> router does the same NAT function, except the outside address is IPv6!")
> 
>>  Given how often people are willing to 'throw-away' smartphones that 
> cost 
>>  in the many $100s of dollars every few years, if there is a need to buy 
>>  an IPv6 supporting router for home, I think people will fairly easily 
>>  find the money.
> 
> You may be right. But there will still be a non-trivial number who won't.
> 
> RossW
> _______________________________________________
> AusNOG mailing list
> AusNOG at lists.ausnog.net
> http://lists.ausnog.net/mailman/listinfo/ausnog
> _______________________________________________
> AusNOG mailing list
> AusNOG at lists.ausnog.net
> http://lists.ausnog.net/mailman/listinfo/ausnog
> 


More information about the AusNOG mailing list