Hello NZNog, We (Vocus) are just in the final stages of getting approval to build our Vocus POP into Auckland. One of the questions asked by our board was if there was interest in a native IPv6 service delivered at Sky Tower. Would anyone who would likely be interested in a $500/month few mbps of native IPv6 type product contact me to express such interest :-) Of course anyone interested in IPv4 transit contact also feel free to email, that itself would clinch the business case :-) Cheers, -- James
On 3 Sep 2008, at 02:30, James Spenceley wrote:
Would anyone who would likely be interested in a $500/month few mbps of native IPv6 type product contact me to express such interest :-) Of course anyone interested in IPv4 transit contact also feel free to email, that itself would clinch the business case :-)
This seems interesting. I don't believe I have seen anybody trying to sell a v6 network access product for money before -- the usual approach seems to be to sell v4 and give v6 away for free. Joe
Joe Abley wrote:
On 3 Sep 2008, at 02:30, James Spenceley wrote:
Would anyone who would likely be interested in a $500/month few mbps of native IPv6 type product contact me to express such interest :-) Of course anyone interested in IPv4 transit contact also feel free to email, that itself would clinch the business case :-)
This seems interesting. I don't believe I have seen anybody trying to sell a v6 network access product for money before -- the usual approach seems to be to sell v4 and give v6 away for free.
There are various world-wide transit providers who charge for IPv6, though generally not a lot compared to IPv4. Some simply call it 'traffic' and charge you for the total bits that you actually send, like it is supposed to be. Btw, the ACSData PoP (http://www.sixxs.net/pops/acsdata/) is still chucking along happily, thus if people want to play with IPv6, then don't hesitate to actually use it. Greets, Jeroen
On 03/09/2008, at 10:30 PM, Joe Abley wrote:
Would anyone who would likely be interested in a $500/month few mbps of native IPv6 type product contact me to express such interest :-) Of course anyone interested in IPv4 transit contact also feel free to email, that itself would clinch the business case :-)
This seems interesting. I don't believe I have seen anybody trying to sell a v6 network access product for money before -- the usual approach seems to be to sell v4 and give v6 away for free.
Sorry, that sentence may have come across slightly ambiguous :-) The $500 was for at least a few mbps of IPv6 (not per mbps), the $500 would cover the port and provide a generous amount of IPv6 only transit. The idea would be for those not buying IPv4 (dual stack) transit from us, yet still wanting a native IPv6 connection, a few such customers would help justify the POP.
Joe
-- James
The most interesting thing here is that it's the only thing to break the 'chicken and egg' proposal in years. The "We're not deploying v6 because our upstreams can't deliver it" argument has been one which is touted about for ages. Couple that with "We can't even get a price out of them" and you have a set of reasons which have been used to justify non-action for at least 5 years. Here we are with a provider, willing to not only build v6 infrastructure if it's in demand, but has even given us all a ballpark price for it, AND offered to deliver it to a place where most of us have a POP anyway. Question for you James. If someone buys a v4 transit link off you, do they get v6 for 'free'? In other words the same price they would pay for a similar amount of additional v4 capacity. If you all think the price is too high, then let I'd encourage you to James know that and he can feed it back into his business case. No one however is going to be able to use "I can't get native transit so I'm doing nothing" as an excuse again =) What do people think, is this the real reason nothing has been happening? Or were you all just tricking =) Dean Joe Abley wrote:
On 3 Sep 2008, at 02:30, James Spenceley wrote:
Would anyone who would likely be interested in a $500/month few mbps of native IPv6 type product contact me to express such interest :-) Of course anyone interested in IPv4 transit contact also feel free to email, that itself would clinch the business case :-)
This seems interesting. I don't believe I have seen anybody trying to sell a v6 network access product for money before -- the usual approach seems to be to sell v4 and give v6 away for free.
Joe _______________________________________________ NZNOG mailing list NZNOG(a)list.waikato.ac.nz http://list.waikato.ac.nz/mailman/listinfo/nznog
Massey University have recently shifted our upstream to Verizon Business. One of the considerations in their selection was that they appeared to be the only transit provider that would offer IPv6. It has turned out that that is only partially true. The connectivity is tunneled, which isn't a big deal and the connectivity is provided on an 'this is experimental and provided primarily for testing purposes' basis. IPv6 is simply bits on the wire in a different format and hence carries no additional cost. In our case the bit count is inflated a little as they have a v4 wrapper but at present, who cares. We have been advised that we are receiving all the routes that VzB have in AsiaPac, a total of 875 prefixes, nothing longer than /32. At present the majority of the destinations favour the NREN (over KAREN) path as opposed to the commodity internet but at least its a start. Our prefix doesn't seem to appear in places like the HE looking glasses so VzB's peering would appear not to be great. If a transit provider isn't able to go native, surely providing tunnels to a suitably connected AS either elsewhere in their own network or to another proovider with whom they peer should be an option. As far as I can see they don't miss out on any revenue for the reasons above and maybe we will see some growth in the deployment of v6 at the edges.
Glen, Glen Eustace wrote:
[...]
If a transit provider isn't able to go native, surely providing tunnels to a suitably connected AS either elsewhere in their own network or to another proovider with whom they peer should be an option. As far as I can see they don't miss out on any revenue for the reasons above and maybe we will see some growth in the deployment of v6 at the edges.
TelstraClear have been providing tunnels to some existing IPv4 customers for a number of years. Could you tell us how much native IPv6 Massey has deployed and when that started? - Donald Neal -- Donald Neal | "I have indisputable blind faith." Research Officer | - Lt. Giardello WAND | The University of Waikato |
TelstraClear have been providing tunnels to some existing IPv4 customers for a number of years.
I don't recall them responding so in the RFP but that was quite some time ago now.
Could you tell us how much native IPv6 Massey has deployed and when that started?
Our boundary is native IPv6, we are peered with the KAREN routers at 3 sites with IPv6 and now with AS 18061 as well. Internally, the core of the network is v6 and the distribution layer is v6 but we have been waiting for an IOS release for the Cisco 3750s and 4500s that supports DHCP relaying. This is now available and will be deployed shortly. That will give us v6 to the desktop. At the moment we have ISATAP setup so that any windows platform with v6 turned on, MacOS or Linux can use v6 if they want. Our external name servers and mail MX'es have AAAA records and we see a small amount of v6 mail, mainly from the UK and Europe. At present we have not put AAAAs on any web resources. Until I get 6to4 setup and confirm that our miredo server is behaving as expected I don't want to get complaints about poor response because all external v6 traffic goes via an offshore relay. As to how much is v6 being used, probably not a lot but it was my aim to lead not to follow. Was there a demand ? Probably the same answer. v6 was rolled out after the workshop that Nathan and crowd did in Wellington last year. When more v6 content is available and when we are able to offer v6 content, we will have the infrastructure in place. There are still issues we haven't found answers too, one that is frustrating me is the lack of v6 support in the Bluecoat web cache appliances. With Massey's HTTP policy, one can not easily browse to v6 content if using the cache is enforced, which it is :-(
On 4/09/2008, at 9:16 AM, Glen Eustace wrote:
TelstraClear have been providing tunnels to some existing IPv4 customers for a number of years.
I don't recall them responding so in the RFP but that was quite some time ago now.
They do not do it as a supported product. It is very much a "you need to know who to talk to" thing.
Could you tell us how much native IPv6 Massey has deployed and when that started?
Our boundary is native IPv6, we are peered with the KAREN routers at 3 sites with IPv6 and now with AS 18061 as well. Internally, the core of the network is v6 and the distribution layer is v6 but we have been waiting for an IOS release for the Cisco 3750s and 4500s that supports DHCP relaying. This is now available and will be deployed shortly. That will give us v6 to the desktop. At the moment we have ISATAP setup so that any windows platform with v6 turned on, MacOS or Linux can use v6 if they want.
Our external name servers and mail MX'es have AAAA records and we see a small amount of v6 mail, mainly from the UK and Europe.
At present we have not put AAAAs on any web resources. Until I get 6to4 setup and confirm that our miredo server is behaving as expected I don't want to get complaints about poor response because all external v6 traffic goes via an offshore relay.
There are more problems than this - put a Windows Vista machine on a filtered non-RFC1918 address, and you'll get breakage with 6to4.
As to how much is v6 being used, probably not a lot but it was my aim to lead not to follow. Was there a demand ? Probably the same answer. v6 was rolled out after the workshop that Nathan and crowd did in Wellington last year.
Cool. We just got done doing the same thing in Christchurch, and are also doing it next week in Auckland. Good to see things are happening.
When more v6 content is available and when we are able to offer v6 content, we will have the infrastructure in place.
There are still issues we haven't found answers too, one that is frustrating me is the lack of v6 support in the Bluecoat web cache appliances. With Massey's HTTP policy, one can not easily browse to v6 content if using the cache is enforced, which it is :-(
I hope you've told your vendor you won't upgrade/whatever unless they do v6 :-) (Squid works just fine... Some vendors are putting Squid proxies in front of IPv4-only HTTP stacks in order to check the "IPv6 ready" box). -- Nathan Ward
At present we have not put AAAAs on any web resources. Until I get 6to4 setup and confirm that our miredo server is behaving as expected I don't want to get complaints about poor response because all external v6 traffic goes via an offshore relay.
There are more problems than this - put a Windows Vista machine on a filtered non-RFC1918 address, and you'll get breakage with 6to4.
Please expand on the above and anything else that comes to mind. I need to know about things that are likely to break when I do 'turn it all on' -- =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Glen and Rosanne Eustace GodZone Internet Services, a division of AGRE Enterprises Ltd. P.O. Box 8020, Palmerston North, New Zealand 4446. Ph: +64 6 357 8168, Fax +64 6 357 8165, Mob: +64 21 424 015 http://www.godzone.net.nz "A Ministry specialising in providing low-cost Internet Services to NZ Christian Churches, Ministries and Organisations."
On 4/09/2008, at 6:32 PM, Glen Eustace wrote:
At present we have not put AAAAs on any web resources. Until I get 6to4 setup and confirm that our miredo server is behaving as expected I don't want to get complaints about poor response because all external v6 traffic goes via an offshore relay. There are more problems than this - put a Windows Vista machine on a filtered non-RFC1918 address, and you'll get breakage with 6to4.
Please expand on the above and anything else that comes to mind. I need to know about things that are likely to break when I do 'turn it all on'
6to4 has no qualification mechanism - that is, it cannot test whether it is working or not as the interface comes up. The MS 6to4 stack assumes that if it has an RFC1918 address, it should run ISATAP or Teredo (in order of preference). If it has a non-RFC1918 address, it assumes it should run ISATAP or 6to4 (in order of preference). So, for those networks that have non-RFC1918 addresses on end nodes, but filter or NAT those end nodes, 6to4 will not function. In many cases (I need to test this some more), it means that TCP connection set up will time out (90sec or so). Unfortunately, these sort of networks are common in academia, as academia has long had lots of IPv4 space available, so why bother with RFC1918? Student connects to student wifi network, gets a public address, and is NATed or filtered at the border = broken 6to4. This was a problem at Auckland Uni until some time ago when they put RFC1918 on their student wifi network. This is not a problem for hosts that are part of a Windows domain - only ISATAP is used in that case, which does not have these problems. Note that many other automatic 6to4 boxes (Apple Airport Extreme) behave similarly, I believe. It's not me bashing on Windows IPv6 this time, I save that for Teredo default settings :-) -- Nathan Ward
Nathan Ward wrote:
On 4/09/2008, at 9:16 AM, Glen Eustace wrote:
TelstraClear have been providing tunnels to some existing IPv4 customers for a number of years. I don't recall them responding so in the RFP but that was quite some time ago now.
They do not do it as a supported product. It is very much a "you need to know who to talk to" thing.
Could you tell us how much native IPv6 Massey has deployed and when that started? Our boundary is native IPv6, we are peered with the KAREN routers at 3 sites with IPv6 and now with AS 18061 as well. Internally, the core of the network is v6 and the distribution layer is v6 but we have been waiting for an IOS release for the Cisco 3750s and 4500s that supports DHCP relaying. This is now available and will be deployed shortly. That will give us v6 to the desktop. At the moment we have ISATAP setup so that any windows platform with v6 turned on, MacOS or Linux can use v6 if they want.
Our external name servers and mail MX'es have AAAA records and we see a small amount of v6 mail, mainly from the UK and Europe.
At present we have not put AAAAs on any web resources. Until I get 6to4 setup and confirm that our miredo server is behaving as expected I don't want to get complaints about poor response because all external v6 traffic goes via an offshore relay.
There are more problems than this - put a Windows Vista machine on a filtered non-RFC1918 address, and you'll get breakage with 6to4.
As to how much is v6 being used, probably not a lot but it was my aim to lead not to follow. Was there a demand ? Probably the same answer. v6 was rolled out after the workshop that Nathan and crowd did in Wellington last year.
Cool. We just got done doing the same thing in Christchurch, and are also doing it next week in Auckland. Good to see things are happening.
When more v6 content is available and when we are able to offer v6 content, we will have the infrastructure in place.
There are still issues we haven't found answers too, one that is frustrating me is the lack of v6 support in the Bluecoat web cache appliances. With Massey's HTTP policy, one can not easily browse to v6 content if using the cache is enforced, which it is :-(
I hope you've told your vendor you won't upgrade/whatever unless they do v6 :-)
(Squid works just fine... Some vendors are putting Squid proxies in front of IPv4-only HTTP stacks in order to check the "IPv6 ready" box).
Thanks for the plug guys. :-) Yes, Squid-3 does full v4/v6 multi-plexing in the Beta release code, with a years worth of testing behind it now. I'm your local IPv6 Squid developer and Squid-3 release maintainer. Here for contact at squid3(a)treenet.co.nz, if anyone has issues or wants assistance. Amos Jeffries Treehouse Networks Ltd Hamilton
TreeNet Admin wrote:
(Squid works just fine... Some vendors are putting Squid proxies in front of IPv4-only HTTP stacks in order to check the "IPv6 ready" box).
Thanks for the plug guys. :-)
Yes, Squid-3 does full v4/v6 multi-plexing in the Beta release code, with a years worth of testing behind it now.
I'm your local IPv6 Squid developer and Squid-3 release maintainer. Here for contact at squid3(a)treenet.co.nz, if anyone has issues or wants assistance.
I have deployed Squid 3.1 at Massey Uni as suggested as an interim to getting IPv6 supported in the BlueCoat caches. So far things look pretty good. Where IPv6 connectivity exists Squid retrieves via v6 otherwise it uses v4, pretty much what one would expect. If anyone wants to test their IPv6 connectivity with some New Zealand content, GodZone's servers are all v6 enabled and provided you are peered at the NZ v6 exchanges on APE or WIX deliver pretty quickly. Feel free to try http://www.splice.cc http://www.thread.co.nz http://www.promisekeepers.org.nz -- =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Glen and Rosanne Eustace GodZone Internet Services, a division of AGRE Enterprises Ltd. P.O. Box 8020, Palmerston North, New Zealand 4446. Ph: +64 6 357 8168, Fax +64 6 357 8165, Mob: +64 21 424 015 http://www.godzone.net.nz "A Ministry specialising in providing low-cost Internet Services to NZ Christian Churches, Ministries and Organisations."
On 3 Sep 2008, at 16:06, Glen Eustace wrote:
We have been advised that we are receiving all the routes that VzB have in AsiaPac, a total of 875 prefixes, nothing longer than /32.
So you're not seeing the /48s that are used to number things like anycast root and TLD nameservers? Seems like a shame. I believe all the RIRs have specified ranges within which the minimum assignment is specified as a /48. The size of a full table as viewed by Afilias routers is: 6453 1115 3257 1253 2914 1121 15412 1350 Our 3549 transit seems to be down right now. All the numbers in that right column are reliably bigger than 875. Joe
On 6/09/2008, at 10:06 AM, Joe Abley wrote:
On 3 Sep 2008, at 16:06, Glen Eustace wrote:
We have been advised that we are receiving all the routes that VzB have in AsiaPac, a total of 875 prefixes, nothing longer than /32.
So you're not seeing the /48s that are used to number things like anycast root and TLD nameservers? Seems like a shame. I believe all the RIRs have specified ranges within which the minimum assignment is specified as a /48.
The size of a full table as viewed by Afilias routers is:
6453 1115 3257 1253 2914 1121 15412 1350
Our 3549 transit seems to be down right now. All the numbers in that right column are reliably bigger than 875.
The best way to figure out what's going on it to set up a BGP session with SIXXS' Ghost Route Hunter, and have a look at their web interface. Idea is, you peer with it, and they tell you what you should have that most other tables have, and what you have that most other tables don't, etc. -- Nathan Ward
participants (9)
-
Dean Pemberton -
Donald Neal -
Glen Eustace -
James Spenceley -
Jeroen Massar -
Joe Abley -
Matthew Moyle-Croft -
Nathan Ward -
TreeNet Admin