Re: [uknof] IPv6 usage explosion

[Neil J. McRae]

On 25 May 2016, at 09:52, Pete Stevens  wrote:

>> Quite possibly! we have IPV6 on business products for five years and the 
>> take up has been almost zero with some customers asking for it to be turned 
>> off - I think this is mostly down to lack of knowledge but I expect that 
>> will start to change soon. Btw I found your presentation really helpful and 
>> informative!
> 
> It's now mission critical in the Raspberry Pi office who have a BT
> leased line. Was most pleased to discover that the v6 configuration was
> supplied as standard with the order - didn't have to ask - and
> everything worked out of the box. Over 50% of their traffic now flows
> over v6.
> 
> So the other reason your customers might not be asking, is because
> your service does it as standard and it works perfectly.

Unfortunately the traffic split says otherwise but glad to see it working  
well! Thanks for the feedback!

Neil 




> 
> Regards,
> 
> Pete
> 
> --
> Pete Stevens
> p...@ex-parrot.com
> http://www.ex-parrot.com/~pete/
> 
> The last time humans crossed space to a destination was the Apollo 17 mission
> in 1972. In the 32 years since, no man has seen, with his own eyes, Earth as
> that beautiful, solitary blue sphere, and - reality check - no woman has ever
>seen it at all.
>   -- James Cameron

Re: [uknof] IPv6 usage explosion

[Mark Tinka]

On 25/May/16 10:47, Neil J. McRae wrote:

>
> Quite possibly! we have IPV6 on business products for five years and the take 
> up has been almost zero with some customers asking for it to be turned off - 
> I think this is mostly down to lack of knowledge but I expect that will start 
> to change soon. Btw I found your presentation really helpful and informative! 

We've seen the same reluctance from our service provider customers. But
we've stayed the course in educating them, and encouraging them to
turn-up IPv6 even when they don't really want to. Success-rate has been
about 50%, but we'll soldier on.

For our enterprise/business customers, we assign each of them a /56 by
default, assign the /126 point-to-point address with each turn-up, and
also include the IPv6 DNS resolvers in the welcome pack. Success-rate
there has been about 1%, which is not unexpected.

We'll continue to encourage them.

Mark.

Re: [uknof] IPv6 usage explosion

[Pete Stevens]

Quite possibly! we have IPV6 on business products for five years and the take 
up has been almost zero with some customers asking for it to be turned off - I 
think this is mostly down to lack of knowledge but I expect that will start to 
change soon. Btw I found your presentation really helpful and informative!


It's now mission critical in the Raspberry Pi office who have a BT
leased line. Was most pleased to discover that the v6 configuration was
supplied as standard with the order - didn't have to ask - and
everything worked out of the box. Over 50% of their traffic now flows
over v6.

So the other reason your customers might not be asking, is because
your service does it as standard and it works perfectly.

Regards,

Pete

--
Pete Stevens
p...@ex-parrot.com
http://www.ex-parrot.com/~pete/

The last time humans crossed space to a destination was the Apollo 17 mission
in 1972. In the 32 years since, no man has seen, with his own eyes, Earth as
that beautiful, solitary blue sphere, and - reality check - no woman has ever
seen it at all.
   -- James Cameron

Re: [uknof] IPv6 usage explosion

[Neil J. McRae]

> 
> Maybe if your customers aren't asking for IPv6, it's because they're
> somebody elses customers?
> 

Quite possibly! we have IPV6 on business products for five years and the take 
up has been almost zero with some customers asking for it to be turned off - I 
think this is mostly down to lack of knowledge but I expect that will start to 
change soon. Btw I found your presentation really helpful and informative! 

Re: [uknof] IPv6 usage explosion

[Mark Tinka]

On 25/May/16 09:37, Pete Stevens wrote:

>  
>
> Maybe if your customers aren't asking for IPv6, it's because they're
> somebody elses customers?

:-). Like that...

Mark.

Re: [uknof] IPv6 usage explosion

[Pete Stevens]

So doesn't this mean that if you have an app which talks directly to
your own back-end server (and not through Apple in some way*), then
your server has to be IPv6 enabled, and if you can only do IPv4 then
you'll not get approved.


Your application must work in an IPv6 only network with a NAT64 gateway.
So it's okay to have an IPv4 only back end (i.e. host in AWS) providing
NAT64 clients aren't an issue.


I wonder how many Apple iOS developers have had to switch ISP in order
to be able to carry out dual stack testing?


We've had a few orders from iOS developers that require IPv6. Certainly
it's started driving awareness from people who aren't network engineers
but software developers instead.

Just to finish with a customer quote from someone who's stuck in
firewall + NAT + overlapping RFC1918 space hell.

"Thanks Pete. I like the sound of ipv6/ssl , sounds like a cleaner
solution"

Maybe if your customers aren't asking for IPv6, it's because they're
somebody elses customers?

Pete

--
Pete Stevens
p...@ex-parrot.com
http://www.ex-parrot.com/~pete/

The last time humans crossed space to a destination was the Apollo 17 mission
in 1972. In the 32 years since, no man has seen, with his own eyes, Earth as
that beautiful, solitary blue sphere, and - reality check - no woman has ever
seen it at all.
   -- James Cameron

Re: [uknof] IPv6 usage explosion

[Paul Mansfield]

On 24 May 2016 at 15:51, Brandon Butterworth  wrote:
>> Apple have mandated for some time that all apps must be fully
>> dual-stack
>
> They've moved on, now all apps have to be capable of ipv6
> only too -
>
> http://www.internetsociety.org/deploy360/blog/2016/05/starting-june-1-apple-requires-all-ios-apps-to-work-in-ipv6-only-networks/

So doesn't this mean that if you have an app which talks directly to
your own back-end server (and not through Apple in some way*), then
your server has to be IPv6 enabled, and if you can only do IPv4 then
you'll not get approved.

I wonder how many Apple iOS developers have had to switch ISP in order
to be able to carry out dual stack testing?


* I don't know whether Apple's App Store ecosystem mandates you use
them as a proxy or even to host your backend

Re: [uknof] IPv6 usage explosion

[Brandon Butterworth]

On Tue May 24, 2016 at 06:23:41PM +, a.l.m.bu...@lboro.ac.uk wrote:
> Hi,
> 
> > They've moved on, now all apps have to be capable of ipv6
> > only too -
> 
>  no. not 'now all apps'. not yet anyway. in about a week that 
> statement is true ;-)

Comiscally insignificant.

As Apple reduced the approval time to a day or so (from 8ish) it is
possible you could write test and deliver your app before the deadline
so OK change that to

 "before people get around to implementing IPv6 instead
  of arguing about it on uknof"

Yes that could be wrong too. I don't mind being wrong you
have a week to roll out IPv6 if you want to gloat

brandon

Re: [uknof] IPv6 usage explosion

[A . L . M . Buxey]

Hi,

> have no IPv6 connectivity. So, sure, in UK and mainland Europe, you
> can probably ignore it, but ultimately you can jump on and ride the
> steamroller or get squashed.

this is why its better to have been doing some IPv6 deployment of at least 
limited scope to
get used to it, get your engineers skillsets up and be ready to deal with it 
when you are forced
to rather than looking at the IPv6 headlights coming at you full speed and 
still denying it.

alan

Re: [uknof] IPv6 usage explosion

[A . L . M . Buxey]

Hi,

> They've moved on, now all apps have to be capable of ipv6
> only too -

 no. not 'now all apps'. not yet anyway. in about a week that 
statement is true ;-)

alan

Re: [uknof] IPv6 usage explosion

[Neil J. McRae]

> I wouldn't expect the VoLTE packets to be using the same "network"
> (VLAN, whatever) as the internet connection network for user
> applications.
> 

Only where ambition is lacking perhaps! Amusingly; in a previous life when 
setting up a new mobile network there was a debate about it being IPV6 ready, 
which was quite funny given launching VoLTE was a requirement! 

Neil.

Re: [uknof] IPv6 usage explosion

[Neil J. McRae]

>> There are solutions to that problem also. Are they nice? - not particularly 
>> but they 
>> do exist and in many ways more customer friendly.
> 
> No; the solutions are meant to be operator-friendly.

They need to be customer friendly - I'm doing nothing that sacrifices customer 
experience over operator lad of ambition!


> 
> Customers don't care.

They do when it doesn't work!

> 
>> VoLTE only really works on IPV6 so nearly all mobile networks are IPV6
> 
> IPv6 traceroutes only work on an IPv6 network. Nearly all ISP's are
> IPv6-ready (not).

Are you trying to be funny ? ;) life is good in a transit ISP ;) VoLTE has lots 
of benefits for customers especially where LTE is extended further than UMTS - 
and is probably the only IPV6 app that is driving true adoption (if it wasn't 
for that we'd never have moved to IPV6 in mobile!)

> 
> Not all mobile networks support VoLTE. So this is a bad assumption to make.
> 

But with VoLTE they have a real driver to move. It would be an interesting view 
to know which networks don't have it. Imagine if sip was IPv6 only and how that 
would have impacted deployment.

Re: [uknof] IPv6 usage explosion

[Paul Mansfield]

On 24 May 2016 at 16:18, Mark Tinka  wrote:
>> VoLTE only really works on IPV6 so nearly all mobile networks are IPV6 
>> already.
>> (surely you knew that? ;))

yes, but...

> Not all mobile networks support VoLTE. So this is a bad assumption to make.


I wouldn't expect the VoLTE packets to be using the same "network"
(VLAN, whatever) as the internet connection network for user
applications.

Re: [uknof] IPv6 usage explosion

[Mark Tinka]

On 24/May/16 15:03, Neil J. McRae wrote:

>
> There are solutions to that problem also. Are they nice? - not particularly 
> but they 
> do exist and in many ways more customer friendly.

No; the solutions are meant to be operator-friendly.

Customers don't care.

> VoLTE only really works on IPV6 so nearly all mobile networks are IPV6 
> already. 
> (surely you knew that? ;))

IPv6 traceroutes only work on an IPv6 network. Nearly all ISP's are
IPv6-ready (not).

Not all mobile networks support VoLTE. So this is a bad assumption to make.

>
> BTW for clarity I’m not saying you shouldn’t roll out IPV6 (I did my first 
> IPV6 roll out almost 
> 10 years ago now) but I do get irked when people have the view that the world 
> will end without V6.
>
> The approach taken for V6 was flawed in my view, and many folks called that 
> out during the standardisation, but 
> here we are anyway.  In my view a better approach would have been some sort 
> of extension to V4, and it would have been done much sooner with a lot less 
> cost and agro

No point in discussing what should have been.

As you said, here we are now...

Mark.

Re: [uknof] IPv6 usage explosion

[Mark Tinka]

On 24/May/16 12:29, Brandon Butterworth wrote:

> I would have expected more v6 = less pressure on CGNAT box so can get
> away with higher user to real IP ratio

That's the intention, but there is more CG-NAT44 out there than there is
native IPv6, with the former growing faster than the latter.

Native IPv6 will also reduce the need for CG-NAT64, but CG-NAT64's fate
is tightly intertwined with how much more IPv4 is still around.

>
> I checked with one such carrier and though they run a fixed ratio for
> planning purposes once more traffic moves to v6 they will revise it.
>
> I presume others do the same?

Yep.

But we do NAT64 instead of NAT44.

Mark.

Re: [uknof] IPv6 usage explosion

[Mark Tinka]

On 24/May/16 11:13, Neil J. McRae wrote:

> There are several solutions that work well - see previous UKNOF presentations.

And that's my point.

Aside from buying IPv4 off the transfer market (a strategy with a finite
life span), all other options to prolong the use of IPv4 are
fundamentally unscalable. Of course, your network, your rules, but if
you can begin forging a plan to make IPv6 a more viable solution in the
long run, you might have a business after the dust settles.

Mark.

Re: [uknof] IPv6 usage explosion

[Neil J. McRae]

On 24/05/2016, 13:37, "uknof on behalf of Paul Mansfield" 
 
wrote:


>On 24 May 2016 at 09:41, Neil J. McRae  wrote:
>> And I don't agree that no IPV6 means no business.
>
>I'm sure if you want to do business with parts of AsiaPAC, where
>IPv6-only services are more common, you'll run into problems if you
>have no IPv6 connectivity. So, sure, in UK and mainland Europe, you
>can probably ignore it, but ultimately you can jump on and ride the
>steamroller or get squashed.


There are solutions to that problem also. Are they nice? - not particularly but 
they 
do exist and in many ways more customer friendly.

>
>Apple have mandated for some time that all apps must be fully
>dual-stack. I wonder whether Apple will use their leverage to insist
>that the mobile networks must offer dual stack connectivity?

VoLTE only really works on IPV6 so nearly all mobile networks are IPV6 already. 
(surely you knew that? ;))

BTW for clarity I’m not saying you shouldn’t roll out IPV6 (I did my first IPV6 
roll out almost 
10 years ago now) but I do get irked when people have the view that the world 
will end without V6.

The approach taken for V6 was flawed in my view, and many folks called that out 
during the standardisation, but 
here we are anyway.  In my view a better approach would have been some sort of 
extension to V4, and it would have been done much sooner with a lot less cost 
and agro

Regards,
Neil.

Re: [uknof] IPv6 usage explosion

[Paul Mansfield]

On 24 May 2016 at 09:41, Neil J. McRae  wrote:
> And I don't agree that no IPV6 means no business.

I'm sure if you want to do business with parts of AsiaPAC, where
IPv6-only services are more common, you'll run into problems if you
have no IPv6 connectivity. So, sure, in UK and mainland Europe, you
can probably ignore it, but ultimately you can jump on and ride the
steamroller or get squashed.

Apple have mandated for some time that all apps must be fully
dual-stack. I wonder whether Apple will use their leverage to insist
that the mobile networks must offer dual stack connectivity?

Re: [uknof] IPv6 usage explosion

[Brian Candler]

On 24/05/2016 11:12, Brandon Butterworth wrote:

That is one of the fundamental problems: deploying IPv6 does*not*  in
>the slightest reduce your need for IPv4 addresses!

I would have expected more v6 = less pressure on CGNAT box so can get
away with higher user to real IP ratio


Sure - if you're a carrier who does NAT on behalf of your customers. 
That means basically just the mobile networks. They force a load of 
stuff through proxies anyway.


For everyone else:

- If you're an edge network, then you're probably happy NATing to a 
single IPv4 address already


(Of course, there are some broken networks who try to NAT far too much 
onto a single IP, e.g. hotels with 1000 rooms. But their networks are 
built with an astounding lack of clue at the best of times)


- If you're a fixed-line ISP, you had better give real IPv4 addresses to 
your customers, or you die in the marketplace


In some markets, you might get away with NAT444 as a consumer ISP. But 
try telling a business customer they can't accept inbound VPN 
connections over IPv4.

Re: [uknof] IPv6 usage explosion

[Tom Hill]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256

On 24/05/16 12:29, Brandon Butterworth wrote:
> On Tue May 24, 2016 at 10:13:00AM +0100, Brian Candler wrote:
>>> That is one of the fundamental problems: deploying IPv6 does
>>> *not* in the slightest reduce your need for IPv4 addresses!
> I would have expected more v6 = less pressure on CGNAT box so can
> get away with higher user to real IP ratio

Yeah, it certainly helps that sites that spawn a lot of concurrent
connections (Google Maps, Facebook) are all v6-enabled. Comcast have
just been showing some figures that indicate far more traffic by
volume to these sites - and with IPv4 in decline.

It also speaks to what Neil's said elsewhere in the thread; use of
IPv6 doesn't yet remove your requirement for IPv4, and thus extending
v4's life (eg. CGN) and implementing IPv6 are concurrent problems to
solve.

- -- 
Tom Hill
Network Engineer

Bytemark Hosting
http://www.bytemark.co.uk/
tel. +44 1904 890 890
-BEGIN PGP SIGNATURE-
Version: GnuPG v2

iQEcBAEBCAAGBQJXRDCFAAoJEH2fKbrp2sQ62IUIAJZMNZrm9bxOmpNdOr+qTWLf
cuBULw2e+t9meizyf36qkN7PjK6K6IgpWETXjN0gyqNCuduJU+qJMJj4A9v+F/uV
KO9f1KePAHVNxHO1/gXSWGC0tWNpCeOQDsJdNcxTp0/hU6PNgQgS0ETytlbzAJmh
sgfnARQyTeAKie2R4PEyhGCuXaXdMm3iJoRAbNVElPD6tQSfNO6zyTNU906DMrk4
xZDmd5Svp8v0aD/+6/+uZa+buYLmw0uYvm45DwYmNKFhvUEUJo1cMThZFPhyCR7c
/+RQRC89rGJPIYfGocOS/8pJ1ckWy/w/6l6gjokw5FP24zLc8PMz9d6wsbu/52c=
=B3+m
-END PGP SIGNATURE-

Re: [uknof] IPv6 usage explosion

[Brandon Butterworth]

On Tue May 24, 2016 at 10:13:00AM +0100, Brian Candler wrote:
> That is one of the fundamental problems: deploying IPv6 does *not* in 
> the slightest reduce your need for IPv4 addresses!

I would have expected more v6 = less pressure on CGNAT box so can get
away with higher user to real IP ratio

I checked with one such carrier and though they run a fixed ratio for
planning purposes once more traffic moves to v6 they will revise it.

I presume others do the same?

brandon

Re: [uknof] IPv6 usage explosion

[Neil J. McRae]

> I'm not asking how much IPv4 is still available. I am asking what you
> will do when you can't get anymore.

There are several solutions that work well - see previous UKNOF presentations.

Cheers,
Neil



> 
> Mark.

Re: [uknof] IPv6 usage explosion

[Brian Candler]

> I've had internal discussions about pushing out IPv6 internally 
everywhere to save on v4


That is one of the fundamental problems: deploying IPv6 does *not* in 
the slightest reduce your need for IPv4 addresses!


You're still going to need as much IPv4 space as you ever did:
- for outbound access to the majority of the Internet (whether it be 
NAT44 or NAT64)

- to allow inbound access from the majority of the Internet

Today, deploying v6 will reduce the processor load on your NAT box, and 
nothing more. This will always be the case until the time it becomes 
feasible to deploy IPv6-only networks. This might happen when:


1. there's so little IPv4-only Internet that you don't care about not 
being fully connected (a very long way away)


2. NAT64 is as easy for a customer to deploy as NAT44 is today (possible 
- but even then ISPs will still need to deliver v4 to customers)


3. ISPs all provide robust, well-managed NAT64 services; and/or there's 
a public NAT64 service which the IPv4-only Internet is reachable through.



> I am utterly of the opinion that the opinions of your customers are 
really irrelevant here.


It's nice to see the customer service ethos still holds strong :-)

But there's a more important point: you can deliver IPv6 to a customer, 
but you can't make them drink.


As an ISP, your business is selling IP packets. It is perfectly natural 
that you should build your network to carry both flavours of IP packets; 
indeed, the work required to achieve this is relatively small.


That's not true for end users, who only care about selling chocolate 
bunnies or whatever.


Dual-stack is a great strategy for switching over a network from IPv4 to 
IPv6. But for most users, dual-stack is a rubbish strategy for permanent 
deployment. There are both initial and ongoing costs, and very little to 
show in the way of benefit.


And before anyone says it: no, there will not be any IPv6-only websites. 
Nobody's business plan involves putting up content which is only 
reachable over IPv6.


And no, the IoT does not need IPv6. The IoT runs today, on IPv4. And 
even with IPv6 they would still use meet-in-the-middle servers. Or do 
you really expect everyone to disable their firewall and allow inbound 
connections from the whole Internet?


Regards,

Brian.

Re: [uknof] IPv6 usage explosion

[Nick Hilliard]

Mark Tinka wrote:
> On 24/May/16 11:01, Nick Hilliard wrote:
>> double-triple-nat?
>>
>> You know I'm not joking.
> 
> You're a braver man than I am...

I'm certainly not advocating it, but it is what people will do in order
to save themselves from getting over them hump of moving to ipv6.

Nick

Re: [uknof] IPv6 usage explosion

[Mark Tinka]

On 24/May/16 11:01, Nick Hilliard wrote:

> double-triple-nat?
>
> You know I'm not joking.

You're a braver man than I am...

Mark.

Re: [uknof] IPv6 usage explosion

[Mark Tinka]

On 24/May/16 11:02, Neil J. McRae wrote:

>
> There is still a huge amount of IPV4 available and there are lots of other 
> solutions to this connectivity requirement 

I'm not asking how much IPv4 is still available. I am asking what you
will do when you can't get anymore.

Mark.

Re: [uknof] IPv6 usage explosion

[Neil J. McRae]

> On 24 May 2016, at 09:52, Mark Tinka  wrote:
> 
> 
> 
>> On 24/May/16 10:41, Neil J. McRae wrote:
>> 
>> The big 4 have done those talks before in my view. 
>> 
>> And I don't agree that no IPV6 means no business.
> 
> Well, when you can't assign anymore IPv4 addresses to your new
> customers, what will you do?


There is still a huge amount of IPV4 available and there are lots of other 
solutions to this connectivity requirement 

IPV6 for the next five (10?) years does nothing to help with the demand for IPv4

Neil.

Re: [uknof] IPv6 usage explosion

[Nick Hilliard]

Mark Tinka wrote:
> Well, when you can't assign anymore IPv4 addresses to your new
> customers, what will you do?

double-triple-nat?

You know I'm not joking.

Nick

Re: [uknof] IPv6 usage explosion

[Alexander Harrowell]

On that chart showing a huge surge in Aussie IPv6, I notice Telstra keeps
having huge multi-service national (ADSL/FTTX/2G/3G/4G, i.e. either their
core IP routers or some centralised provisioning/AAA system) outages. Does
anyone wonder if some device they use has a dodgy IPv6 implementation?

On Tue, May 24, 2016 at 9:52 AM, Chris Russell  wrote:

>
>>> I suppose that could be an interesting talk, particularly if it was
>>> laborious (and assuming the presenter still has enough energy to keep on
>>> living).
>>>
>>> But the basic story to management would be something along the lines of,
>>> "No IPv6, no business".
>>>
>>
> There is a proposed talk for UKNOF Glasgow along these lines…..
>
>  With my UKNOF PC hat on, it would be nice to see more infrastructure
> companies coming forward with war stories - not necessarily ISP's - people
> deploying Cloud, OpenStack, Customer WANS + Networks.
>
> .. And btw agree with Neil, "no IPv6, no business" is rubbish …  of 2
> large people I’m aware of (one us) - we’ve had a combined 5 requests for
> IPv6 - amongst a large customer base.
>
> Customers have no interest unless they are a very techie business with
> interested engineers.
>
> likewise, whilst itemising v6 can be done (and I applaud you Pete for
> doing this) - but if you added a v4 cost for an infrastructure project IMO
> you'd be shooting yourself in the foot.
>
> Chris
>
>
>
>

Re: [uknof] IPv6 usage explosion

[Chris Russell]

I suppose that could be an interesting talk, particularly if it was
laborious (and assuming the presenter still has enough energy to 
keep on

living).

But the basic story to management would be something along the lines 
of,

"No IPv6, no business".


There is a proposed talk for UKNOF Glasgow along these lines…..

 With my UKNOF PC hat on, it would be nice to see more infrastructure 
companies coming forward with war stories - not necessarily ISP's - 
people deploying Cloud, OpenStack, Customer WANS + Networks.


.. And btw agree with Neil, "no IPv6, no business" is rubbish …  of 2 
large people I’m aware of (one us) - we’ve had a combined 5 requests for 
IPv6 - amongst a large customer base.


Customers have no interest unless they are a very techie business with 
interested engineers.


likewise, whilst itemising v6 can be done (and I applaud you Pete for 
doing this) - but if you added a v4 cost for an infrastructure project 
IMO you'd be shooting yourself in the foot.


Chris

Re: [uknof] IPv6 usage explosion

[Mark Tinka]

On 24/May/16 10:41, Neil J. McRae wrote:

> The big 4 have done those talks before in my view. 
>
> And I don't agree that no IPV6 means no business. 

Well, when you can't assign anymore IPv4 addresses to your new
customers, what will you do?

Mark.

Re: [uknof] IPv6 usage explosion

[Neil J. McRae]

The big 4 have done those talks before in my view. 

And I don't agree that no IPV6 means no business. 

Neil.

Sent from my iPhone

> On 24 May 2016, at 09:12, Mark Tinka  wrote:
> 
> 
> 
>> On 24/May/16 10:01, James Bensley wrote:
>> 
>> I wonder if there is scope at a UKNOF for someone to give an IPv6
>> story but from the other end of the spectrum. We've had some great
>> "We're rolling it out, this is how far we have got and how long to
>> finish" and also some "We rolled it out and this is how we did it" - I
>> wonder if anyone wants to give a talk on the business side of things,
>> how did they make it appeal to upper management, how did they get
>> their customers on board, how did they get their internal engineers on
>> board (I encounter no shortage of enigneering colleagues that don't
>> care / don't think its time yet), how did you sell it to customers
>> commercially?
> 
> I suppose that could be an interesting talk, particularly if it was
> laborious (and assuming the presenter still has enough energy to keep on
> living).
> 
> But the basic story to management would be something along the lines of,
> "No IPv6, no business".
> 
> Mark.
> 

Re: [uknof] IPv6 usage explosion

[Pete Stevens]

Hi,


how did you sell it to customers commercially?


On the hosting side I gave this talk at the last UKnof about selling v6
only hosting services.

https://www.youtube.com/watch?v=91hN3PvYUrc

In short, itemise IPv4 addresses. The accountants ask you to stop 
using them to save money. Techies hate talking to accountants even more

than they hate fixing configurations to work with IPv6.

Pete

--
Pete Stevens
p...@ex-parrot.com
http://www.ex-parrot.com/~pete/

The last time humans crossed space to a destination was the Apollo 17 mission
in 1972. In the 32 years since, no man has seen, with his own eyes, Earth as
that beautiful, solitary blue sphere, and - reality check - no woman has ever
seen it at all.
   -- James Cameron

Re: [uknof] IPv6 usage explosion

[Tim Bray]

On 24/05/16 09:01, James Bensley wrote:
> Lots of technical hurdles like buggy routers/switches/firewalls, buggy
> applications or applications that simply don't support IPv6 or have
> some IPv4 hard coded parts etc, that all adds up to management saying
> "you see, it will take too much time/money/whatever, get back to
> writing that report on how many reports you've been writing."

I would urge to start working on these now.

If you use this excuse forever, you won't move on.  (ok, if you do have
a big IPv6 lab setup, keep going)

If you start working through the problems one at a time, you will gain
IPv6 experience.

At some time in the future, a customer will say `We need IPv6`.  Or you
will be looking at a problem, and just wishing you could use IPv6 to do it.

Start skilling up now.  And business is more fun if you keep thinking of
little improvements, new ideas, new mini ways to impress a customer.

Anyway, it is fun to do http://uid0.com/ip6dc/provu.co.uk


[Maybe in 3 years, a CEO of a customer will ring you and say `Everytime
I try to watch the footie on Sky player, it stops working.  It works
fine at home, and it works in Bob's office.  And I rang Sky, and they
said IPv6 is better for watching footie, they have more servers.  At
that stage, a competitor might say `We do IPv6 as standard` and you will
be back of queue]

Tim

Re: [uknof] IPv6 usage explosion

[Marek Isalski]

> On 24 May 2016, at 09:01, James Bensley  wrote:
> how did they make it appeal to upper management, how did they get
> their customers on board, how did they get their internal engineers on
> board (I encounter no shortage of enigneering colleagues that don't
> care / don't think its time yet), how did you sell it to customers
> commercially?

I think it'd be quite interesting to just run a poll along the lines of the 
above questions.  I wouldn't be surprised if it were to find e.g. half the 
networks teams "just did it" without getting senior/executive management 
involved.  Something along the lines of: we know that IPv6 has *got* to be done 
and we're going to do it as part of our day-to-day activities.  We won't get 
management or financial support, because beancounters, but we can make sure 
that over the next three years our kit replacement programme only buys 
IPv6-supporting tin.  We can lab things up, and gradually roll it out as a sort 
of "stealth project" that doesn't need approval - beyond network and security 
teams - and can just become part of those teams' normal business activities.

Of course, the stealth approach would only work up to a certain size, and 
certainly not when there has been a publicly announced target of IPv6 adoption 
(e.g. Sky's, BT's).  I imagine a lot of smaller networks either stealthily 
rolled it out years ago, or had it "big bang" from day one, so any such poll 
would have to factor in organisation size and market.

Hey, maybe AS5089 can come to UKNOF and do a talk about their IPv6-by-stealth 
project? ;)

Marek Isalski

Re: [uknof] IPv6 usage explosion

[Brian Nisbet]

James,

James Bensley wrote on 24/05/2016 10:01:

On 23 May 2016 at 11:14, Paul Mansfield  wrote:

On 23 May 2016 at 09:59, James Bensley  wrote:

as IPv4. On lots of our Cisco edge devices there are bugs present that
relate to IPv6 traffic processing problems, or just the fact that IPv6
is enabled. The Junos kit is more mature and seems pretty bug free. In



we need early adopters to find the bugs. If people wait until v6 is
really mature and solid, they'll be lagging in the skills and
experience to successfully roll it out.
now is a good time for ISPs to be rolling it out at least in trials,
to test their equipment and train their own staff, whilst the
customers adopting it tend to be the clueful ones who are aware it's
imperfect and so likely to be a little more helpful and forgiving.


I definitely agree with you there. However with all these problems it
makes it difficult to make progress. I've had internal discussions
about pushing out IPv6 internally everywhere to save on v4, however I
get met with mostly resistance. $dayjob is more of a "managed service
provider" than a more traditional telco/ISP, which basically means "we
only do what's good for revenue, based on what customers say" - and of
corse a tiny fraction of customers jump up and say "I demand IPv6".


On this, I am utterly of the opinion that the opinions of your customers 
are really irrelevant here. It's not like companies waited for their 
customers to say they wanted CGNs, nor do they ask if they want IPv4.


Customers want connectivity, it's up to the provider to decide how 
they're going to deliver that connectivity and it's up to the customer 
to decide if that connectivity is worth paying for.


People don't want IPv6, they want connectivity. So the notion of 
"waiting for customer demand" is, to my mind, a trap and a great way of 
making sure that nothing will ever happen. (Please note, I'm not saying 
you're saying this!)


So, yes, customers want services at a good price. The vast majority of 
them simply don't care via what colour pipe those services are 
delivered, so don't worry about that part of the equation, worry about 
all the others! :)


Brian

Re: [uknof] IPv6 usage explosion

[Mark Tinka]

On 24/May/16 10:01, James Bensley wrote:

> I wonder if there is scope at a UKNOF for someone to give an IPv6
> story but from the other end of the spectrum. We've had some great
> "We're rolling it out, this is how far we have got and how long to
> finish" and also some "We rolled it out and this is how we did it" - I
> wonder if anyone wants to give a talk on the business side of things,
> how did they make it appeal to upper management, how did they get
> their customers on board, how did they get their internal engineers on
> board (I encounter no shortage of enigneering colleagues that don't
> care / don't think its time yet), how did you sell it to customers
> commercially?

I suppose that could be an interesting talk, particularly if it was
laborious (and assuming the presenter still has enough energy to keep on
living).

But the basic story to management would be something along the lines of,
"No IPv6, no business".

Mark.

Re: [uknof] IPv6 usage explosion

[James Bensley]

On 23 May 2016 at 11:14, Paul Mansfield  wrote:
> On 23 May 2016 at 09:59, James Bensley  wrote:
>> as IPv4. On lots of our Cisco edge devices there are bugs present that
>> relate to IPv6 traffic processing problems, or just the fact that IPv6
>> is enabled. The Junos kit is more mature and seems pretty bug free. In
>
>
> we need early adopters to find the bugs. If people wait until v6 is
> really mature and solid, they'll be lagging in the skills and
> experience to successfully roll it out.
> now is a good time for ISPs to be rolling it out at least in trials,
> to test their equipment and train their own staff, whilst the
> customers adopting it tend to be the clueful ones who are aware it's
> imperfect and so likely to be a little more helpful and forgiving.

I definitely agree with you there. However with all these problems it
makes it difficult to make progress. I've had internal discussions
about pushing out IPv6 internally everywhere to save on v4, however I
get met with mostly resistance. $dayjob is more of a "managed service
provider" than a more traditional telco/ISP, which basically means "we
only do what's good for revenue, based on what customers say" - and of
corse a tiny fraction of customers jump up and say "I demand IPv6".

Lots of technical hurdles like buggy routers/switches/firewalls, buggy
applications or applications that simply don't support IPv6 or have
some IPv4 hard coded parts etc, that all adds up to management saying
"you see, it will take too much time/money/whatever, get back to
writing that report on how many reports you've been writing."

I wonder if there is scope at a UKNOF for someone to give an IPv6
story but from the other end of the spectrum. We've had some great
"We're rolling it out, this is how far we have got and how long to
finish" and also some "We rolled it out and this is how we did it" - I
wonder if anyone wants to give a talk on the business side of things,
how did they make it appeal to upper management, how did they get
their customers on board, how did they get their internal engineers on
board (I encounter no shortage of enigneering colleagues that don't
care / don't think its time yet), how did you sell it to customers
commercially?


Cheers,
James.

Re: [uknof] IPv6 usage explosion

[Richard Patterson]

On Mon, May 23, 2016 at 11:14 AM, Paul Mansfield  wrote:

>
> we need early adopters to find the bugs. If people wait until v6 is
> really mature and solid, they'll be lagging in the skills and
> experience to successfully roll it out.
> now is a good time for ISPs to be rolling it out at least in trials,
> to test their equipment and train their own staff, whilst the
> customers adopting it tend to be the clueful ones who are aware it's
> imperfect and so likely to be a little more helpful and forgiving.
>
>
Sure but http://ipv6excuses.com/

Re: [uknof] IPv6 usage explosion

[Paul Mansfield]

On 23 May 2016 at 09:59, James Bensley  wrote:
> as IPv4. On lots of our Cisco edge devices there are bugs present that
> relate to IPv6 traffic processing problems, or just the fact that IPv6
> is enabled. The Junos kit is more mature and seems pretty bug free. In


we need early adopters to find the bugs. If people wait until v6 is
really mature and solid, they'll be lagging in the skills and
experience to successfully roll it out.
now is a good time for ISPs to be rolling it out at least in trials,
to test their equipment and train their own staff, whilst the
customers adopting it tend to be the clueful ones who are aware it's
imperfect and so likely to be a little more helpful and forgiving.

Re: [uknof] IPv6 usage explosion

[James Bensley]

> On 20 May 2016, at 10:16, Tim Chown  wrote:
>
>>> On 19 May 2016, at 21:57, Neil J. McRae  wrote:
>>>
>>> Sky have done - we should have launched also but we (BT) hit a minor bug 
>>> but we want to patch it before we turn it on but the work is done and we 
>>> have a pile of customer using it.
>>>
>>> the stability of most V6 implementations still leaves a hell of a lot to be 
>>> desired.
>>
>> In what area? OSes, applications, CPEs? Can you give specific examples?
>>
>> Tim


On 20 May 2016 at 14:50, Neil J. McRae  wrote:
> I am talking home equipment but even in edge my view is that it's not at the 
> same level as it will need to be.
>
> Neil
>
> Sent from my iPhone


Certainly for u IPv6 in the edge is not at the same level of maturity
as IPv4. On lots of our Cisco edge devices there are bugs present that
relate to IPv6 traffic processing problems, or just the fact that IPv6
is enabled. The Junos kit is more mature and seems pretty bug free. In
either case these are "vendor" specific problems in some sense,
however they are two massive vendors that many ISPs all over the world
will be using so it's fair to say that on some global level, IPv6 is
not the same at the service provider edge as it is for IPv4 for
stability, security, reliability etc (Cisco in particular, bug
central).

Cheers,
James.

Re: [uknof] IPv6 usage explosion

[Stephen Strowes]

On 19 May 2016 at 04:38, Tom Storey  wrote:

> Other countries have also experienced recent explosions in IPv6 usage,
> e.g. Australia:
>
> https://www.vyncke.org/ipv6status/compare.php?metric=p=ww,gb,au
>
> Is anyone aware of any networks that have flipped some switches (and,
> perhaps, what kind of switches?) recently that would account for this
> kind of explosive growth?
>

AU looks like Telstra (AS 1221). Their v6 traffic tripled about a week ago.

S.

Re: [uknof] IPv6 usage explosion

[Stephen Strowes]

On 20 May 2016 at 02:59, Matthew Ford  wrote:

>
> > On 19 May 2016, at 18:41, Tom Hill  wrote:
> >
> > See: http://www.worldipv6launch.org/measurements/ for more figures. :)
>
> And specifically,
> http://www.worldipv6launch.org/apps/ipv6week/measurement/images/graphs/BritishSkyBroadcasting.png
>


And still going up. Currently seeing ~50% of requests from Sky over v6.

S.

Re: [uknof] IPv6 usage explosion

[Chris Russell]

On 20/05/2016 16:37, Tom Hill wrote:

There were definitely better options back in 2012 - even OpenBSD
would've done a better job, where feature parity is concerned anyway.



There may be a UKNOF lightning in the works about the first campus v6 
network we ($dayjob at the time) did, so don’t want to jump in too much.


Ultimately its about budgets, customer requirements and commercial 
realities, it takes a brave engineer & strong management inside a 
technically ambitious company working alongside a trusting customer to 
get v6 infrastructure projects out in the wild IMO … and sometimes you 
have to sacrifice technical perfection to get there.


With that said, just because something isn’t ‘technically perfect’ 
doesn’t mean its insecure, or designed badly - but if we wait till 
everything is absolutely perfect in v6, we may never get there (for some 
that IS a requirement - see Xbox / v6 from the v6 Council Meeting - but 
not all).


ISP is one thing, projects / managed services another … given Sky's 
(thanks Iain and co) rollout, BT’s soon to be (thanks Neil and co)  
rollout, and Vir… ok leave that there, once they’re done - it’ll be the 
infrastructure / projects companies dragging us behind.


Final thought - Cisco ASA-X for the Firewall v6/dual stack/etc win btw 
…  they are epic level good in my experience...



Chris

Re: [uknof] IPv6 usage explosion

[Pete Stevens]

I think it's the lack of external IPv6 that makes their decision to
use IPv6-only internally a pretty weird idea in the first place.


The network admin had just come off a three month project writing NAT44
rules from RFC1918 space to other RFC1918 space as a result of a merge
of multiple other companies in his previous job.

But yes, I think had they added full v6 connectivity from day 1 it would
have worked much better, although their firewall experience suggests
that many v6 implementations are missing either reliability or
performance or both.


Whilst I'm glad someone's tried it, it's not overly representative of
the dual-stack environment that Sky is rolling to their customers.


Sure, We've had considerable success rolling dual stack through various
providers (BT leased line included, thank you Neil) and providing
internal services on v6 only addresses. Not having to deal with NAT and
a VPN and just using SSL//SSH can make some of your problems significantly
eaiser and more reliable, not withstanding the advantages of full end to
end encryption rather than insecure -> secure -> insecure.

Pete

--
Pete Stevens
https://twitter.com/Mythic_Beasts

Re: [uknof] IPv6 usage explosion

[Tom Hill]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256

On 20/05/16 16:24, Pete Stevens wrote:
> But yes, I think had they added full v6 connectivity from day 1 it
> would have worked much better, although their firewall experience
> suggests that many v6 implementations are missing either
> reliability or performance or both.

I could say a lot about both Vyatta and Fortigates, but none of it
would be positive; both deserve to be cast unto the great blackhole of
suck. Oddly enough my experiences of these were both 2012 and before,
too. It doesn't seem like they've improved much in this area.

There were definitely better options back in 2012 - even OpenBSD
would've done a better job, where feature parity is concerned anyway.

- -- 
Tom Hill
Network Engineer

Bytemark Hosting
http://www.bytemark.co.uk/
tel. +44 1904 890 890
-BEGIN PGP SIGNATURE-
Version: GnuPG v2

iQEcBAEBCAAGBQJXPy9RAAoJEH2fKbrp2sQ6TL0H/2jRdCWN5CmusynHAn2Hx0zT
dhxvUCAHXxsx+fc0ZbrFew1O0HB79DkchjKXgI9j4K9payaCQ0vUvuxuN9ebuYRw
QpRZfvRLF7Xau8rbI384RlgbQu/n4nGFfFdusV0B82VkFNUPTTpoXU/COFxm4cgZ
M9lf/SahgWTvrw+nwXDBaKdbLm/F8VguPVFADXWnRbnogDG2yo7d+ghdBF1fTmEj
MhnBCMrjUwOhxTuK5Ra/dSk1EjT445w98mE6+988LsNiSdiov3J76dhN3c/JifOB
+gEMd7S35S1XSPzZjlen4szRKXoAMGMLXRjRpKYoC+tK8ht6IWhx07VLDawYy7c=
=YFn2
-END PGP SIGNATURE-

Re: [uknof] IPv6 usage explosion

[Alan Buxey]

Yep Sky.  Not Virgin Media :(

alan

Re: [uknof] IPv6 usage explosion

[Bjoern A. Zeeb]

> On 20 May 2016, at 13:44 , Pete Stevens  wrote:
> 
>> OK, I see.   What of this is still true for them today in 2016 (apart from 
>> AWS)?
> 
> I think the summary is
> 
> ..
> Telling the customer they're stupid and have to replace their kit is not
> really an acceptable response.

No, but that wasn’t my intention for asking either.

I bought my “IPv6 ready” labelled (and certified) printer in 2010 as well,
and found IPv6 bugs.  I reported them but people on the remote end did not
understand so they never got fixed (was in a 3rd party software component
they had bought and included).

The point is:  2012 is four years ago.  Was the effort made to make the
vendors aware of their problems back then?

I am asking as the original email asserted that today so many problems
still exist (and probably do but the question is are they any different
to the problems are used to from the legacy world)?

Re: [uknof] IPv6 usage explosion

[Bjoern A. Zeeb]

> On 20 May 2016, at 14:12 , Neil J. McRae  wrote:
> 
> How is it a decade late? We have had plenty of IPV4 (and we still do!) 
> 
> Does anyone truly think that current implementations in IPV6 are an equal to 
> V4 in terms of reliability for mass market deployment? (And even not mass 
> market lots of things still suck!)

Name them please?  And tell me what you did to get this changed?   Asserting 
there are problems isn’t helpful.

I have way less problems on IPv6 today than I used to have on (DFZ) IPv4 15 
years ago.

Re: [uknof] IPv6 usage explosion

[Tom Hill]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256

On 20/05/16 15:22, Pete Stevens wrote:
> Within AWS it's v4 single stack within their VPC, which means they
> have to have IPv4 across their whole private LAN at which point
> IPv6 becomes entirely optional and as their experiences go, easier
> without than with.

I think it's the lack of external IPv6 that makes their decision to
use IPv6-only internally a pretty weird idea in the first place.

Whilst I'm glad someone's tried it, it's not overly representative of
the dual-stack environment that Sky is rolling to their customers.

- -- 
Tom Hill
Network Engineer

Bytemark Hosting
http://www.bytemark.co.uk/
tel. +44 1904 890 890
-BEGIN PGP SIGNATURE-
Version: GnuPG v2

iQEcBAEBCAAGBQJXPyA3AAoJEH2fKbrp2sQ6gd8IAJ2f+q5hdet5OFRguQcbAVZO
8bOkNzOiZF7FpWN36SGsc23HKvSGZOpD+/62t0iNHF/Okxz2vQZ5CLvLTaQYSWPn
FE45s8HQssjWvKrr2l9egD8Ao9fOqT3gSVrmBjHa6UoL6B2ghjTz25s7bTEw6/1+
F7FsYxuTzF7sYQIXPKnmPc+FFXJISpMFSJIQ0xsp0Jsztodp6zWM8Yfwe9XPCV2t
r3VF9mImDJoIg1pGjGSNpavM2/kkORd76SJsjMCh4CpUBdVvHlQPQqmQUkd9H9bt
TrTzmqx87AzkIj38cyR1UbgxHs818bG0BscbCgDlwLL5yOEpiYaG2u3sscpmen4=
=PNZ7
-END PGP SIGNATURE-

Re: [uknof] IPv6 usage explosion

[Pete Stevens]

On Fri, 20 May 2016, Tom Hill wrote:

Hang on. Where they running IPv6-only here on this LAN? Surely
dual-stack wouldn't have been an issue with regards to AWS?


Within AWS it's v4 single stack within their VPC, which means they have
to have IPv4 across their whole private LAN at which point IPv6 becomes
entirely optional and as their experiences go, easier without than with.

Pete

--
Pete Stevens
p...@ex-parrot.com
http://www.ex-parrot.com/~pete/

The last time humans crossed space to a destination was the Apollo 17 mission
in 1972. In the 32 years since, no man has seen, with his own eyes, Earth as
that beautiful, solitary blue sphere, and - reality check - no woman has ever
seen it at all.
   -- James Cameron

Re: [uknof] IPv6 usage explosion

[Neil J. McRae]

How is it a decade late? We have had plenty of IPV4 (and we still do!) 

Does anyone truly think that current implementations in IPV6 are an equal to V4 
in terms of reliability for mass market deployment? (And even not mass market 
lots of things still suck!)

> On 20 May 2016, at 15:04, Bjoern A. Zeeb  
> wrote:
> 
> 
>> On 20 May 2016, at 13:20 , Pete Stevens  wrote:
>> 
>> Hi all,
>> 
>> I've just quizzed one of our customers for their experiences, they
>> adopted an IPv6 internal LAN early on and have had a fairly rough ride.
>> 
>> Note - they chose not to have external IPv6, only on their internal LAN.
>> So you join the VPN and get a v6 only internal route which gives you
>> access to internal services, reserving v4 only for the public internet.
>> They started this in 2012 which was brave.
> 
> Or a decade late, but OK.
> 
> 
 the stability of most V6 implementations still leaves a hell of a lot to 
 be desired.
>>> -v ?
>> 
>> Biggest issues :
>> 
>> Vyatta couldn't do IPv6 cluster failover Fortinet firewalls have perfomrance 
>> issues with high volumes of IPv6
>> traffic.
>> AWS doesn't support EC2 at all, this is a major issue for them.
>> Dynamic DNS filled up with  records
>> They couldn't give Android devices fixed addresses.
>> Some devices can't do IPv6 at all or badly (projector with no support,
>> Cisco switches that seemed to have ssh servers that would eventually
>> stop).
>> Happy Eyeballs having a very high hit rate of picking exactly the wrong
>> protocol and route to use.
>> 
>> 
>> So for them it's largely failed to work out, the lack of v6 in AWS
>> coupled with their Fortinet firewalls collapsing under high traffic load
>> resulted in this comment from them.
>> 
>> [tom] 0/10 would not recommend.
> 
> OK, I see.   What of this is still true for them today in 2016 (apart from 
> AWS)?
> 
> Did they report this in 2012?
> Have they made more careful planning decisions since when upgrading gear?
> 
> 
>> Within Mythic Beasts on the server side it's been much less painful, but
>> largely we have Linux servers, Linux clients and if it's not Linux it's
>> almost certainly somebody elses problem.
> 
> ;-)

Re: [uknof] IPv6 usage explosion

[Tom Hill]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256

On 20/05/16 14:44, Pete Stevens wrote:
> AWS + Fortinet is still a complete show-stopper for them, if you
> know a provider with a global footprint including Brazil and large
> GPU instances available by the hour let me know.

Hang on. Where they running IPv6-only here on this LAN? Surely
dual-stack wouldn't have been an issue with regards to AWS?

- -- 
Tom Hill
Network Engineer

Bytemark Hosting
http://www.bytemark.co.uk/
tel. +44 1904 890 890
-BEGIN PGP SIGNATURE-
Version: GnuPG v2

iQEcBAEBCAAGBQJXPxnnAAoJEH2fKbrp2sQ6LtIIAIuEd0XNBQOdZWuAAglp2ZQK
E/pdU9O6KjkM6qqj9w0b0Hkn27Qlr72aveq4xhFU/wseEL4u6vIs70xv9K8sIwHJ
q0Q89b5In+JqN+LHqfI9GF67OOo2vdD8R2FjKcbWJtU4kqOGjSfd6l3CyD1ZKoyo
MhoWQHdcZiLmlaCBggKObay061tXI+chUL4Zy8f/VMHFKsPKHnJtbaCTsPC6T7C0
3TaHEPLodaD5UFQ3nLl1smTtXWFYBHkk35AweFXDBrB3Xuk62SE23vOEOtH3ohOC
hbAF0soj2l4bb8WHwVl8RNHWQN1WimGVcsTfRUEI+e6GufCBhAPJpUOpJC1taF8=
=j2V4
-END PGP SIGNATURE-

Re: [uknof] IPv6 usage explosion

[Paul Mansfield]

On 20 May 2016 at 14:20, Pete Stevens  wrote:
> adopted an IPv6 internal LAN early on and have had a fairly rough ride.
>
> Note - they chose not to have external IPv6, only on their internal LAN.


I can imagine that would have been a major headache, and don't know
why anyone would even want to do that other than for a short while as
an experiment.

Re: [uknof] IPv6 usage explosion

[Bjoern A. Zeeb]

> On 20 May 2016, at 13:20 , Pete Stevens  wrote:
> 
> Hi all,
> 
> I've just quizzed one of our customers for their experiences, they
> adopted an IPv6 internal LAN early on and have had a fairly rough ride.
> 
> Note - they chose not to have external IPv6, only on their internal LAN.
> So you join the VPN and get a v6 only internal route which gives you
> access to internal services, reserving v4 only for the public internet.
> They started this in 2012 which was brave.

Or a decade late, but OK.


>>> the stability of most V6 implementations still leaves a hell of a lot to be 
>>> desired.
>> -v ?
> 
> Biggest issues :
> 
> Vyatta couldn't do IPv6 cluster failover Fortinet firewalls have perfomrance 
> issues with high volumes of IPv6
> traffic.
> AWS doesn't support EC2 at all, this is a major issue for them.
> Dynamic DNS filled up with  records
> They couldn't give Android devices fixed addresses.
> Some devices can't do IPv6 at all or badly (projector with no support,
> Cisco switches that seemed to have ssh servers that would eventually
> stop).
> Happy Eyeballs having a very high hit rate of picking exactly the wrong
> protocol and route to use.
> 
> 
> So for them it's largely failed to work out, the lack of v6 in AWS
> coupled with their Fortinet firewalls collapsing under high traffic load
> resulted in this comment from them.
> 
> [tom] 0/10 would not recommend.

OK, I see.   What of this is still true for them today in 2016 (apart from AWS)?

Did they report this in 2012?
Have they made more careful planning decisions since when upgrading gear?


> Within Mythic Beasts on the server side it's been much less painful, but
> largely we have Linux servers, Linux clients and if it's not Linux it's
> almost certainly somebody elses problem.

;-)

Re: [uknof] IPv6 usage explosion

[Neil J. McRae]

I am talking home equipment but even in edge my view is that it's not at the 
same level as it will need to be.

Neil 

Sent from my iPhone

On 20 May 2016, at 10:16, Tim Chown  wrote:

>> On 19 May 2016, at 21:57, Neil J. McRae  wrote:
>> 
>> Sky have done - we should have launched also but we (BT) hit a minor bug but 
>> we want to patch it before we turn it on but the work is done and we have a 
>> pile of customer using it. 
>> 
>> the stability of most V6 implementations still leaves a hell of a lot to be 
>> desired.
> 
> In what area? OSes, applications, CPEs? Can you give specific examples?
> 
> Tim
> 
>> Neil 
>> 
>> Sent from my iPhone
>> 
>>> On 19 May 2016, at 18:38, Tom Storey  wrote:
>>> 
>>> See this graph here:
>>> 
>>> https://www.vyncke.org/ipv6status/compare.php?metric=p=ww,gb
>>> 
>>> Other countries have also experienced recent explosions in IPv6 usage,
>>> e.g. Australia:
>>> 
>>> https://www.vyncke.org/ipv6status/compare.php?metric=p=ww,gb,au
>>> 
>>> Is anyone aware of any networks that have flipped some switches (and,
>>> perhaps, what kind of switches?) recently that would account for this
>>> kind of explosive growth?
> 

Re: [uknof] IPv6 usage explosion

[Pete Stevens]

Hi all,

I've just quizzed one of our customers for their experiences, they
adopted an IPv6 internal LAN early on and have had a fairly rough ride.

Note - they chose not to have external IPv6, only on their internal LAN.
So you join the VPN and get a v6 only internal route which gives you
access to internal services, reserving v4 only for the public internet.
They started this in 2012 which was brave.


the stability of most V6 implementations still leaves a hell of a lot to be 
desired.

-v ?


Biggest issues :

Vyatta couldn't do IPv6 cluster failover 
Fortinet firewalls have perfomrance issues with high volumes of IPv6

traffic.
AWS doesn't support EC2 at all, this is a major issue for them.
Dynamic DNS filled up with  records
They couldn't give Android devices fixed addresses.
Some devices can't do IPv6 at all or badly (projector with no support,
Cisco switches that seemed to have ssh servers that would eventually
stop).
Happy Eyeballs having a very high hit rate of picking exactly the wrong
protocol and route to use.


So for them it's largely failed to work out, the lack of v6 in AWS
coupled with their Fortinet firewalls collapsing under high traffic load
resulted in this comment from them.

[tom] 0/10 would not recommend.


Within Mythic Beasts on the server side it's been much less painful, but
largely we have Linux servers, Linux clients and if it's not Linux it's
almost certainly somebody elses problem.

Pete

--
Pete Stevens
p...@mythic-beasts.com

https://twitter.com/Mythic_Beasts

Re: [uknof] IPv6 usage explosion

[Matthew Ford]

> On 19 May 2016, at 18:41, Tom Hill  wrote:
> 
> See: http://www.worldipv6launch.org/measurements/ for more figures. :)

And specifically, 
http://www.worldipv6launch.org/apps/ipv6week/measurement/images/graphs/BritishSkyBroadcasting.png

Some other interesting recent developments:

http://www.worldipv6launch.org/apps/ipv6week/measurement/images/graphs/OrangeBusinessServices.png

http://www.worldipv6launch.org/apps/ipv6week/measurement/images/graphs/CoxCommunications.png

Mat

Re: [uknof] IPv6 usage explosion

[David Reader]

On 19 May 2016, at 21:57, Neil J. McRae wrote:

the stability of most V6 implementations still leaves a hell of a lot 
to be desired.


whose?

are you referring to service providers or consumer equipment?

mine is rock solid, but I would say that :)

d.

Re: [uknof] IPv6 usage explosion

[Bjoern A. Zeeb]

> On 19 May 2016, at 20:57 , Neil J. McRae  wrote:
> 
> the stability of most V6 implementations still leaves a hell of a lot to be 
> desired. 

-v ?


/bz

Re: [uknof] IPv6 usage explosion

[Tim Chown]

> On 19 May 2016, at 21:57, Neil J. McRae  wrote:
> 
> Sky have done - we should have launched also but we (BT) hit a minor bug but 
> we want to patch it before we turn it on but the work is done and we have a 
> pile of customer using it. 
> 
> the stability of most V6 implementations still leaves a hell of a lot to be 
> desired. 

In what area? OSes, applications, CPEs? Can you give specific examples?

Tim

> Neil 
> 
> Sent from my iPhone
> 
>> On 19 May 2016, at 18:38, Tom Storey  wrote:
>> 
>> See this graph here:
>> 
>> https://www.vyncke.org/ipv6status/compare.php?metric=p=ww,gb
>> 
>> Other countries have also experienced recent explosions in IPv6 usage,
>> e.g. Australia:
>> 
>> https://www.vyncke.org/ipv6status/compare.php?metric=p=ww,gb,au
>> 
>> Is anyone aware of any networks that have flipped some switches (and,
>> perhaps, what kind of switches?) recently that would account for this
>> kind of explosive growth?
>> 
> 

Re: [uknof] IPv6 usage explosion

[Dariush Marsh-Mossadeghi]

Sky are on a big ipv6 push.
Ian Dickinson gave an update at LINX93 earlier this week, iirc they’ve got 3m+ 
subscribers running dual-stack now and aggressively rolling out across the rest 
of the network \o/
https://www.linx.net/communications/events/linx93 


> On 19 May 2016, at 12:38, Tom Storey  wrote:
> 
> See this graph here:
> 
> https://www.vyncke.org/ipv6status/compare.php?metric=p=ww,gb
> 
> Other countries have also experienced recent explosions in IPv6 usage,
> e.g. Australia:
> 
> https://www.vyncke.org/ipv6status/compare.php?metric=p=ww,gb,au
> 
> Is anyone aware of any networks that have flipped some switches (and,
> perhaps, what kind of switches?) recently that would account for this
> kind of explosive growth?
> 



signature.asc
Description: Message signed with OpenPGP using GPGMail

Re: [uknof] IPv6 usage explosion

[Tim Chown]

> On 19 May 2016, at 12:38, Tom Storey  wrote:
> 
> See this graph here:
> 
> https://www.vyncke.org/ipv6status/compare.php?metric=p=ww,gb
> 
> Other countries have also experienced recent explosions in IPv6 usage,
> e.g. Australia:
> 
> https://www.vyncke.org/ipv6status/compare.php?metric=p=ww,gb,au
> 
> Is anyone aware of any networks that have flipped some switches (and,
> perhaps, what kind of switches?) recently that would account for this
> kind of explosive growth?

Sky.

Tim

Re: [uknof] IPv6 usage explosion

[Tom Hill]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256

On 19/05/16 12:38, Tom Storey wrote:
> Is anyone aware of any networks that have flipped some switches
> (and, perhaps, what kind of switches?) recently that would account
> for this kind of explosive growth?

Sky? Lots of talks about it over the last 12 months, most recently at
LINX, and prior to that UKNOF in London if I recall correctly.

See: http://www.worldipv6launch.org/measurements/ for more figures. :)

- -- 
Tom Hill
Network Engineer

Bytemark Hosting
http://www.bytemark.co.uk/
tel. +44 1904 890 890
-BEGIN PGP SIGNATURE-
Version: GnuPG v2

iQEcBAEBCAAGBQJXPfq0AAoJEH2fKbrp2sQ60lkIALaYpeakBA51s3NbVxDfQlpA
k2VrbEGpI/nUGR6l8o1bweTOpqeD+cddWgqD0HqaKx8/14SSzQHgb8mIQb98HZU1
9uaLi8B/bkSVGwQOetwSSqxwYlTqQdDbJEgdjbYtQ/8kculZHpaPAXJd5MGe82UL
cE/SF2HaYlQKlzeEhU1Cbnv5eZ9N62TcaiyxP3SS9PfTBljG/DyKmx9g+yR2kS3W
w+q4OSuT+45yDLQEvXz1siejve3ZFo/SS4AZ/5UNJIEFNL4pLpGo4ycd0EGCQouK
CRQhrvtGQJ2x6R7stIdm3iNI9RTDpKzT0wrKzY3sQvjn4UAK+Q8cpoH3YKSZ5hU=
=X9wR
-END PGP SIGNATURE-

Re: [uknof] IPv6 usage explosion

[Russell Heilling]

Those bumps seem to line up pretty well with the graphs Ian Dickinson
presented at the LINX meeting regarding Sky's IPv6 rollout...

On 19 May 2016 at 12:38, Tom Storey  wrote:

> See this graph here:
>
> https://www.vyncke.org/ipv6status/compare.php?metric=p=ww,gb
>
> Other countries have also experienced recent explosions in IPv6 usage,
> e.g. Australia:
>
> https://www.vyncke.org/ipv6status/compare.php?metric=p=ww,gb,au
>
> Is anyone aware of any networks that have flipped some switches (and,
> perhaps, what kind of switches?) recently that would account for this
> kind of explosive growth?
>
>


-- 
Russell Heillinghttp://perlmonkey.blogspot.com
"The amazing ability of the bee to adapt herself often helps the
 beekeeper to overcome the results of his ignorance." - Brother Adam

[uknof] IPv6 usage explosion

[Tom Storey]

See this graph here:

https://www.vyncke.org/ipv6status/compare.php?metric=p=ww,gb

Other countries have also experienced recent explosions in IPv6 usage,
e.g. Australia:

https://www.vyncke.org/ipv6status/compare.php?metric=p=ww,gb,au

Is anyone aware of any networks that have flipped some switches (and,
perhaps, what kind of switches?) recently that would account for this
kind of explosive growth?