Mailing List Archive

yum IPv6 repos
I thought I'd try to setup an IPv6 only network this weekend and noticed
many/most yum repos and mirrors are not IPv6-enabled. Just wondering what
are folks who are trying to wean themselves off of IPv4 doing nowadays to
handle upgrades/patches if you have only IPv6 connectivity?

Antonio Querubin
e-mail: tony@lavanauts.org
xmpp: antonioquerubin@gmail.com
Re: yum IPv6 repos [ In reply to ]
On Mon, Dec 26, 2011 at 3:32 PM, Antonio Querubin <tony@lavanauts.org> wrote:
> I thought I'd try to setup an IPv6 only network this weekend and noticed
> many/most yum repos and mirrors are not IPv6-enabled.  Just wondering what
> are folks who are trying to wean themselves off of IPv4 doing nowadays to
> handle upgrades/patches if you have only IPv6 connectivity?
>

Yum as in Fedora? Have you looked at http://mirror.uoregon.edu/

mirror.uoregon.edu has address 128.223.157.9
mirror.uoregon.edu has IPv6 address 2001:468:d01:103::80df:9d09

And if all else fails, you can setup a NAT64/DNS64 gateway for your
ipv6-only hosts

http://ecdysis.viagenie.ca/
> Antonio Querubin
> e-mail:  tony@lavanauts.org
> xmpp:  antonioquerubin@gmail.com
Re: yum IPv6 repos [ In reply to ]
On Mon, 26 Dec 2011, Cameron Byrne wrote:

> On Mon, Dec 26, 2011 at 3:32 PM, Antonio Querubin <tony@lavanauts.org> wrote:
>> I thought I'd try to setup an IPv6 only network this weekend and noticed
>> many/most yum repos and mirrors are not IPv6-enabled.  Just wondering what
>> are folks who are trying to wean themselves off of IPv4 doing nowadays to
>> handle upgrades/patches if you have only IPv6 connectivity?
>>
>
> Yum as in Fedora? Have you looked at http://mirror.uoregon.edu/

As in Scientific Linux, EPEL, and RPMforge.

> mirror.uoregon.edu has address 128.223.157.9
> mirror.uoregon.edu has IPv6 address 2001:468:d01:103::80df:9d09

Unfortunately not listed on the SL mirror list but there are
others IPv6-enabled. I think I'll need to create custom IPv6-only mirror
lists.

> And if all else fails, you can setup a NAT64/DNS64 gateway for your
> ipv6-only hosts
>
> http://ecdysis.viagenie.ca/

Yeah researching that too - I read there's a potential performance hit.

Antonio Querubin
e-mail: tony@lavanauts.org
xmpp: antonioquerubin@gmail.com
Re: yum IPv6 repos [ In reply to ]
On Mon, 2011-12-26 at 15:58 -0800, Cameron Byrne wrote:
> Yum as in Fedora? Have you looked at http://mirror.uoregon.edu/
>
> mirror.uoregon.edu has address 128.223.157.9
> mirror.uoregon.edu has IPv6 address 2001:468:d01:103::80df:9d09

Also, for those not in the states:

$ host mirror.bytemark.co.uk
mirror.bytemark.co.uk has address 212.110.161.69
mirror.bytemark.co.uk has IPv6 address 2001:41c8:20:5e6::10

> And if all else fails, you can setup a NAT64/DNS64 gateway for your
> ipv6-only hosts

Eee.. Please don't.

I realise that it's an exercise in "IPv6-only", but adding address
translation would be clouding the reality that surrounds the
currently-available Internet access via IPv6.

NAT64 would defeat the point of the exercise entirely.

TOm
Re: yum IPv6 repos [ In reply to ]
On Mon, Dec 26, 2011 at 5:21 PM, Tom Hill <tom@ninjabadger.net> wrote:
> On Mon, 2011-12-26 at 15:58 -0800, Cameron Byrne wrote:
>> Yum as in Fedora?  Have you looked at http://mirror.uoregon.edu/
>>
>> mirror.uoregon.edu has address 128.223.157.9
>> mirror.uoregon.edu has IPv6 address 2001:468:d01:103::80df:9d09
>
> Also, for those not in the states:
>
> $ host mirror.bytemark.co.uk
> mirror.bytemark.co.uk has address 212.110.161.69
> mirror.bytemark.co.uk has IPv6 address 2001:41c8:20:5e6::10
>
>> And if all else fails, you can setup a NAT64/DNS64 gateway for your
>> ipv6-only hosts
>
> Eee.. Please don't.
>
> I realise that it's an exercise in "IPv6-only", but adding address
> translation would be clouding the reality that surrounds the
> currently-available Internet access via IPv6.
>

If you are concerned about it being cloudy, i am sure you can figure
it the relative native vs translated traffic pretty easy.

> NAT64 would defeat the point of the exercise entirely.


I guess that's for the OP to decide what the purpose of his exercise
is...but he did sound like he was on that path to NAT64/DNS64

>

The OP said IPv6-only. And, i assume he wants it to be functional for
something and perhaps approaching production quality (we are beyond
the science experiment phase i hope). That said, NAT64/DNS64 is what
that solution looks like. And, NAT64/DNS64 has become quite common in
shops that are serious about IPv6 and having a strategic answer to the
business challenge of IPv4 exhaustion and multi-protocol complexity
(YMMV).

http://www.slideshare.net/IPv6no/20111122-i-pv6forumnorway2singlestackipv6onlydatacenterdeployments

http://www.slideshare.net/IPv6no/cameron-tmo-ipv6-norway-meeting

The fact is, IPv4 is effectively exhausted from the perspective of
planning any large scale strategic initiative. And, the internet will
soon be composed of IPv4-only, IPv6-only, and DS nodes. I suggest
folks get used to the idea that IPv4-only nodes and IPv6-only nodes
will need to communicate with each other... Like the IPv6-only mobile
phone talking to the "buy-now button" on the IPv4-only e-commerce
site.

cb

> TOm
>
Re: yum IPv6 repos [ In reply to ]
On 12/26/2011 17:43, Cameron Byrne wrote:
> The fact is,

Facts? You mean your speculation. Calling your speculation a fact
doesn't make it any more so.

> IPv4 is effectively exhausted from the perspective of
> planning any large scale strategic initiative.

The carriers that are pushing CGN disagree with you. Not to mention that
"typical" NAT setups today at the end-user network level only require 1
IPv4 address. We're going to see IPv4 continue to exist for a very long
time.

That's not to say that IPv6 isn't a better choice, only that we need to
be realistic about the landscape.

--

[^L]

Breadth of IT experience, and depth of knowledge in the DNS.
Yours for the right price. :) http://SupersetSolutions.com/
Re: yum IPv6 repos [ In reply to ]
On Mon, Dec 26, 2011 at 5:48 PM, Doug Barton <dougb@dougbarton.us> wrote:
> On 12/26/2011 17:43, Cameron Byrne wrote:
>> The fact is,
>
> Facts? You mean your speculation. Calling your speculation a fact
> doesn't make it any more so.
>

Fair enough.

>> IPv4 is effectively exhausted from the perspective of
>> planning any large scale strategic initiative.
>
> The carriers that are pushing CGN disagree with you. Not to mention that
> "typical" NAT setups today at the end-user network level only require 1
> IPv4 address. We're going to see IPv4 continue to exist for a very long
> time.
>

Agreed, that IPv4-only will be around for a long long time, hence the
need to understand NAT64/DNS64.


> That's not to say that IPv6 isn't a better choice, only that we need to
> be realistic about the landscape.
>
> --
>
>                [^L]
>
>        Breadth of IT experience, and depth of knowledge in the DNS.
>        Yours for the right price.  :)  http://SupersetSolutions.com/
>
RE: yum IPv6 repos [ In reply to ]
Long list here:
http://www.sixxs.net/wiki/IPv6_Enabled_Mirrors

Frank

-----Original Message-----
From: ipv6-ops-bounces+frnkblk=iname.com@lists.cluenet.de
[mailto:ipv6-ops-bounces+frnkblk=iname.com@lists.cluenet.de] On Behalf Of
Antonio Querubin
Sent: Monday, December 26, 2011 5:32 PM
To: ipv6-ops@lists.cluenet.de
Subject: yum IPv6 repos

I thought I'd try to setup an IPv6 only network this weekend and noticed
many/most yum repos and mirrors are not IPv6-enabled. Just wondering what
are folks who are trying to wean themselves off of IPv4 doing nowadays to
handle upgrades/patches if you have only IPv6 connectivity?

Antonio Querubin
e-mail: tony@lavanauts.org
xmpp: antonioquerubin@gmail.com
RE: yum IPv6 repos [ In reply to ]
Fedora ones here:
http://mirrors.fedoraproject.org/publiclist/

Frank

-----Original Message-----
From: ipv6-ops-bounces+frnkblk=iname.com@lists.cluenet.de
[mailto:ipv6-ops-bounces+frnkblk=iname.com@lists.cluenet.de] On Behalf Of
Antonio Querubin
Sent: Monday, December 26, 2011 5:32 PM
To: ipv6-ops@lists.cluenet.de
Subject: yum IPv6 repos

I thought I'd try to setup an IPv6 only network this weekend and noticed
many/most yum repos and mirrors are not IPv6-enabled. Just wondering what
are folks who are trying to wean themselves off of IPv4 doing nowadays to
handle upgrades/patches if you have only IPv6 connectivity?

Antonio Querubin
e-mail: tony@lavanauts.org
xmpp: antonioquerubin@gmail.com
RE: yum IPv6 repos [ In reply to ]
On Mon, 26 Dec 2011, Frank Bulk wrote:

> Long list here:
> http://www.sixxs.net/wiki/IPv6_Enabled_Mirrors

Perfect! I think I'll add a Scientific Linux section. Thanks!

Antonio Querubin
e-mail: tony@lavanauts.org
xmpp: antonioquerubin@gmail.com
Re: yum IPv6 repos [ In reply to ]
On Tuesday, December 27, 2011 10:10:41 AM Cameron Byrne
wrote:

> Agreed, that IPv4-only will be around for a long long
> time, hence the need to understand NAT64/DNS64.

Some folks will continue to push NAT44, others NAT444, and
some of us will push NAT64. It's all choice, trade-offs,
yadda yadda...

While there is bound to be a lot more breakage in NAT64
today due to lack of IPv6 support in common applications
such as Skype, Yahoo Messenger, e.t.c., I believe these can
be fixed by software vendors adding IPv6 support into their
code.

Hopefully, it will get to a stage where NAT64 isn't anymore
broken than NAT44 or NAT444 when it comes to translation.

For us, NAT64 makes sense because we don't have to
transition customers in the future when traffic ratios are
in favour of IPv6 (whenever that is).

Mark.
Re: yum IPv6 repos [ In reply to ]
On 27. Dec 2011, at 02:27 , Antonio Querubin wrote:

> On Mon, 26 Dec 2011, Frank Bulk wrote:
>
>> Long list here:
>> http://www.sixxs.net/wiki/IPv6_Enabled_Mirrors
>
> Perfect! I think I'll add a Scientific Linux section. Thanks!

Might it make more sense to just link the project's own list rather
than duplicate the information and possibly let it get stale in two places?

FreeBSD has shipped releases and things from IPv6 enabled mirrors since
May 2003 and there has been quite a bit of fluctuation in mirror sites in
these 8+ years;-)

If http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/mirrors-ftp.html
is unhelpful I am sure we can also add a dedicated page listing only mirrors
with v6.

/bz

--
Bjoern A. Zeeb You have to have visions!
Stop bit received. Insert coin for new address family.
Re: yum IPv6 repos [ In reply to ]
On Mon, Dec 26, 2011 at 01:32:28PM -1000, Antonio Querubin wrote:
> I thought I'd try to setup an IPv6 only network this weekend and noticed
> many/most yum repos and mirrors are not IPv6-enabled. Just wondering what
> are folks who are trying to wean themselves off of IPv4 doing nowadays to
> handle upgrades/patches if you have only IPv6 connectivity?

FWIW, we've been maintaining a v6-only Ubuntu desktop machine for a few months and most things work fine - v6-accessible mirrors, system tools all work, etc. However, there's some interaction with the Canonical mothership related to release jumps - 11.04 to 11.10, for example - and that does *not* work. The machine appears incapable of realizing that there's a new version available.

I don't know whether any of the distros you're using have a similar mechanism, but if so, you may have to enable v4 every now and again in order to get things to work normally.

Bill.
Re: yum IPv6 repos [ In reply to ]
On Tue, 27 Dec 2011, Bjoern A. Zeeb wrote:

> Might it make more sense to just link the project's own list rather
> than duplicate the information and possibly let it get stale in two places?

I haven't found a yum-formatted mirrorlist for Scientific Linux so one
will have to be generated anyway (the yum-conf RPM installed files show a
commented out mirrorlist URI and those apparently don't work any more).
But if there is a current yum mirrorlist somewhere it makes more sense to
place it on a dual-stack server somewhere.

> FreeBSD has shipped releases and things from IPv6 enabled mirrors since
> May 2003 and there has been quite a bit of fluctuation in mirror sites in
> these 8+ years;-)

> If http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/mirrors-ftp.html
> is unhelpful I am sure we can also add a dedicated page listing only mirrors
> with v6.

For freebsd, it would also be good to get freshports.org dual-stacked.

Antonio Querubin
e-mail: tony@lavanauts.org
xmpp: antonioquerubin@gmail.com
Re: yum IPv6 repos [ In reply to ]
On Dec 27, 2011, at 11:43 AM, Antonio Querubin wrote:

> For freebsd, it would also be good to get freshports.org dual-stacked.

Beware FreeBSD w/ IPv6 as it may return misleading errors in some cases for IPv6:

http://www.freebsd.org/cgi/query-pr.cgi?pr=156684

- Jared
Re: yum IPv6 repos [ In reply to ]
I just had a look at the Ubuntu archives mirror list but there is no
information on IPv6 listed there:

https://launchpad.net/ubuntu/+archivemirrors

I created a small script that parses this page, checks if the mirror
has an IPv6 address, and then outputs the 'enhanced' page:

http://cajones.org/~visser/ubuntu-mirrors/

Interesting results, I didn't do a count but it looks like roughly one
third of them can do IPv6.
I'll mail the Ubuntu mirrors team to see if this information can be
added to the original mirrors page.

Dick


On 27 December 2011 18:09, Jared Mauch <jared@puck.nether.net> wrote:
>
> On Dec 27, 2011, at 11:43 AM, Antonio Querubin wrote:
>
>> For freebsd, it would also be good to get freshports.org dual-stacked.
>
> Beware FreeBSD w/ IPv6 as it may return misleading errors in some cases for IPv6:
>
> http://www.freebsd.org/cgi/query-pr.cgi?pr=156684
>
> - Jared



--
Dick Visser
System & Networking Engineer
TERENA Secretariat
Singel 468 D, 1017 AW Amsterdam
The Netherlands
T +31 20 530 44 88 F +31 20 530 44 99
visser@terena.org | www.terena.org
Re: yum IPv6 repos [ In reply to ]
On Tue, Dec 27, 2011 at 10:54 AM, Dick Visser <visser@terena.org> wrote:
> I just had a look at the Ubuntu archives mirror list but there is no
> information on IPv6 listed there:
>
> https://launchpad.net/ubuntu/+archivemirrors
>
> I created a small script that parses this page, checks if the mirror
> has an IPv6 address, and then outputs the 'enhanced' page:
>
> http://cajones.org/~visser/ubuntu-mirrors/
>
> Interesting results, I didn't do a count but it looks like roughly one
> third of them can do IPv6.
> I'll mail the Ubuntu mirrors team to see if this information can be
> added to the original mirrors page.
>

Thanks! This is helpful info.

I am pretty shocked that more mirrors in the USA are not IPv6 enabled,
especially given that so many are at USA universities, which in theory
should have easy access to IPv6 via Internet2, and likely their
primary ISP as well.

It seems like these "non revenue generating" "for the public good"
mirroring services would be prime candidates for those various hosting
organizations to get their feet wet with IPv6 and start generating
some traffic.

Anyone have clue on how best to improve this situation with the
existing mirrors?

CB

> Dick
>
>
> On 27 December 2011 18:09, Jared Mauch <jared@puck.nether.net> wrote:
>>
>> On Dec 27, 2011, at 11:43 AM, Antonio Querubin wrote:
>>
>>> For freebsd, it would also be good to get freshports.org dual-stacked.
>>
>> Beware FreeBSD w/ IPv6 as it may return misleading errors in some cases for IPv6:
>>
>> http://www.freebsd.org/cgi/query-pr.cgi?pr=156684
>>
>> - Jared
>
>
>
> --
> Dick Visser
> System & Networking Engineer
> TERENA Secretariat
> Singel 468 D, 1017 AW Amsterdam
> The Netherlands
> T +31 20 530 44 88 F +31 20 530 44 99
> visser@terena.org | www.terena.org
Re: yum IPv6 repos [ In reply to ]
On Tue, Dec 27, 2011 at 11:14:03AM -0800, Cameron Byrne wrote:
> On Tue, Dec 27, 2011 at 10:54 AM, Dick Visser <visser@terena.org> wrote:
> > I created a small script that parses this page, checks if the mirror
> > has an IPv6 address, and then outputs the 'enhanced' page:
> >
> > http://cajones.org/~visser/ubuntu-mirrors/
> >
>
> I am pretty shocked that more mirrors in the USA are not IPv6 enabled,
> especially given that so many are at USA universities, which in theory
> should have easy access to IPv6 via Internet2, and likely their
> primary ISP as well.

IPv6 uptake amongst US universities isn't great - they have lots of v4 space, are busy with other things and have had their IT budgets squeezed quite a bit in the last couple of years. Of course that doesn't stop some of us from trying to convince them that they still need to do v6, we just don't have much traction.

Incidentally, here I am on a regional network that has v6 transit from Internet2 and NLR, trying to reach cajones.org:

[cookiemonster:~] owens% sudo tcptraceroute6 cajones.org
traceroute to cajones.org (2001:610:148:dead::666) from 2620:f:1:1201:21b:63ff:fea4:4d92, port 80, from port 49823, 30 hops max, 60 bytes packets
1 c7609-32c.nysernet.org (2620:f:1:1201::1) 0.000 ms 0.000 ms 0.000 ms
2 syr-7600-nnsyr.nysernet.net (2620:f:0:901::2) 0.000 ms 0.000 ms 0.000 ms
3 nyc-7600-syr-7600.nysernet.net (2620:f:0:734::3) 10.000 ms 0.000 ms 10.000 ms
4 nlr-nyc-7600.nysernet.net (2620:f:0:305::3) 0.000 ms 10.000 ms 0.000 ms
5 ge1-0-0.1113.jnr02.asd001a.surf.net (2001:610:f16:6016::17) 90.000 ms 90.000 ms 80.000 ms
6 ae2.500.jnr01.asd001a.surf.net (2001:610:e08:76::78) 90.000 ms 90.000 ms 100.000 ms
7 terena-router.customer.surf.net (2001:610:f01:8168::170) 80.000 ms 90.000 ms 80.000 ms
8 * * *
9 * * *

versus v4:
[cookiemonster:~] owens% sudo tcptraceroute cajones.org
Selected device en0, address 199.109.32.135, port 56079 for outgoing packets
Tracing the path to cajones.org (192.87.30.17) on TCP port 80 (http), 30 hops max
1 c7609.nysernet.net (199.109.32.254) 0.657 ms 0.236 ms 0.212 ms
2 syr-7600-nnsyr.nysernet.net (199.109.9.1) 1.983 ms 0.254 ms 0.262 ms
3 nyc-7600-syr-7600.nysernet.net (199.109.7.78) 6.100 ms 6.079 ms 6.072 ms
4 nlr-nyc-7600.nysernet.net (199.109.4.158) 6.195 ms 6.157 ms 6.123 ms
5 ge1-0-0.1113.jnr02.asd001.surf.net (145.145.166.17) 94.152 ms 92.918 ms 94.137 ms
6 ae2.500.jnr01.asd001a.surf.net (145.145.80.78) 106.063 ms 93.239 ms 92.851 ms
7 terena-router.customer.surf.net (145.145.18.170) 93.964 ms 94.909 ms 97.566 ms
8 cajones.org (192.87.30.17) [open] 95.652 ms 94.183 ms 94.033 ms

There's something not right between us, on the v6 side. I don't know whether your script was actually testing connectivity or just pulling AAAA records, but if a test was involved it might be giving false negative answers.

Bill.
Re: yum IPv6 repos [ In reply to ]
On 27 December 2011 19:54, Dick Visser <visser@terena.org> wrote:
> I just had a look at the Ubuntu archives mirror list

Just did the same for the (English) page of the Debian archives:

http://www.debian.org/mirror/list.en.html

http://cajones.org/~visser/debian-mirrors/




--
Dick Visser
System & Networking Engineer
TERENA Secretariat
Singel 468 D, 1017 AW Amsterdam
The Netherlands
T +31 20 530 44 88 F +31 20 530 44 99
visser@terena.org | www.terena.org
Re: yum IPv6 repos [ In reply to ]
On 27 December 2011 23:51, Dick Visser <visser@terena.org> wrote:

> Just did the same for the (English) page of the Debian archives:

And now for the Fedora list:

http://mirrors.fedoraproject.org/publiclist
http://cajones.org/~visser/fedora-mirrors/

What's interesting here is that the original list has a Comment row,
which mentions IPv6 for some sites.
With the actual checks in the same row, a few inconsistencies now surface:

* 'fedora.mirror.nexicom.net' and 'ftp.ntua.gr' allegedly are dual
stack, but they're really only IPv4.

Likewise, there are a lot of sites in the list that actually have
IPv6, but this is not mentioned in the Comment field.




--
Dick Visser
System & Networking Engineer
TERENA Secretariat
Singel 468 D, 1017 AW Amsterdam
The Netherlands
T +31 20 530 44 88 F +31 20 530 44 99
visser@terena.org | www.terena.org
Re: yum IPv6 repos [ In reply to ]
On 27. Dec 2011, at 16:43 , Antonio Querubin wrote:

> On Tue, 27 Dec 2011, Bjoern A. Zeeb wrote:
>
> For freebsd, it would also be good to get freshports.org dual-stacked.

That's a service run by an individual not by the project but I can ask Dan the next time I'll see him around:)

/bz

--
Bjoern A. Zeeb You have to have visions!
Stop bit received. Insert coin for new address family.
Re: yum IPv6 repos [ In reply to ]
I'm not sure if I entirely buy that... The factors you gave (and others) certainly are all true, but at the same time whenever we press vendors for IPv6 support, we get feedback to the tune that nobody else is asking for it, and amongst networking engineer peer meetings there are discussions about keeping up pressure for IPv6 support and keeping IPv6 support in requirement documents for new purchases in order to keep the momentum going.
It may be that I'm just seeing a self-selected group, but overall I think that universities are doing a pretty decent job with IPv6. I know we've embraced it, and have run a public mirror for several distributions with native IPv6 support for years now. (ftp.usf.edu)

Toivo Voll
Information Technology Communications
University of South Florida



On Dec 27, 2011, at 14:58 , Bill Owens wrote:
>
> IPv6 uptake amongst US universities isn't great - they have lots of v4 space, are busy with other things and have had their IT budgets squeezed quite a bit in the last couple of years. Of course that doesn't stop some of us from trying to convince them that they still need to do v6, we just don't have much traction.
>
> Incidentally, here I am on a regional network that has v6 transit from Internet2 and NLR, trying to reach cajones.org:
>
> [cookiemonster:~] owens% sudo tcptraceroute6 cajones.org
> traceroute to cajones.org (2001:610:148:dead::666) from 2620:f:1:1201:21b:63ff:fea4:4d92, port 80, from port 49823, 30 hops max, 60 bytes packets
> 1 c7609-32c.nysernet.org (2620:f:1:1201::1) 0.000 ms 0.000 ms 0.000 ms
> 2 syr-7600-nnsyr.nysernet.net (2620:f:0:901::2) 0.000 ms 0.000 ms 0.000 ms
> 3 nyc-7600-syr-7600.nysernet.net (2620:f:0:734::3) 10.000 ms 0.000 ms 10.000 ms
> 4 nlr-nyc-7600.nysernet.net (2620:f:0:305::3) 0.000 ms 10.000 ms 0.000 ms
> 5 ge1-0-0.1113.jnr02.asd001a.surf.net (2001:610:f16:6016::17) 90.000 ms 90.000 ms 80.000 ms
> 6 ae2.500.jnr01.asd001a.surf.net (2001:610:e08:76::78) 90.000 ms 90.000 ms 100.000 ms
> 7 terena-router.customer.surf.net (2001:610:f01:8168::170) 80.000 ms 90.000 ms 80.000 ms
> 8 * * *
> 9 * * *
>
> versus v4:
> [cookiemonster:~] owens% sudo tcptraceroute cajones.org
> Selected device en0, address 199.109.32.135, port 56079 for outgoing packets
> Tracing the path to cajones.org (192.87.30.17) on TCP port 80 (http), 30 hops max
> 1 c7609.nysernet.net (199.109.32.254) 0.657 ms 0.236 ms 0.212 ms
> 2 syr-7600-nnsyr.nysernet.net (199.109.9.1) 1.983 ms 0.254 ms 0.262 ms
> 3 nyc-7600-syr-7600.nysernet.net (199.109.7.78) 6.100 ms 6.079 ms 6.072 ms
> 4 nlr-nyc-7600.nysernet.net (199.109.4.158) 6.195 ms 6.157 ms 6.123 ms
> 5 ge1-0-0.1113.jnr02.asd001.surf.net (145.145.166.17) 94.152 ms 92.918 ms 94.137 ms
> 6 ae2.500.jnr01.asd001a.surf.net (145.145.80.78) 106.063 ms 93.239 ms 92.851 ms
> 7 terena-router.customer.surf.net (145.145.18.170) 93.964 ms 94.909 ms 97.566 ms
> 8 cajones.org (192.87.30.17) [open] 95.652 ms 94.183 ms 94.033 ms
>
> There's something not right between us, on the v6 side. I don't know whether your script was actually testing connectivity or just pulling AAAA records, but if a test was involved it might be giving false negative answers.
>
> Bill.
Re: yum IPv6 repos [ In reply to ]
On 02/18/2012 16:31, Voll, Toivo wrote:
> I'm not sure if I entirely buy that... The factors you gave (and
> others) certainly are all true, but at the same time whenever we
> press vendors for IPv6 support, we get feedback to the tune that
> nobody else is asking for it

This has been widely debunked by various people comparing notes and
having the same vendors (often the same salescritters) giving the same
answer to each of them.


Doug

--

It's always a long day; 86400 doesn't fit into a short.

Breadth of IT experience, and depth of knowledge in the DNS.
Yours for the right price. :) http://SupersetSolutions.com/
Re: yum IPv6 repos [ In reply to ]
On 19 February 2012 09:41, Doug Barton <dougb@dougbarton.us> wrote:
> On 02/18/2012 16:31, Voll, Toivo wrote:
>> I'm not sure if I entirely buy that... The factors you gave (and
>> others) certainly are all true, but at the same time whenever we
>> press vendors for IPv6 support, we get feedback to the tune that
>> nobody else is asking for it
>
> This has been widely debunked by various people comparing notes and
> having the same vendors (often the same salescritters) giving the same
> answer to each of them.

Literally, indeed.

But I've come to think that when a sales person says "nobody's asking
for it", they aren't saying that nobody has requested it, even though
that would be the logical English interpretation of that phrase.
Indeed, as you say, this has been debunked.

Instead, I think they're speaking an entirely different language
which, when more properly translated into English, might be something
closer to "nobody has actually made IPv6 a purchasing requirement on
an order that was worth enough money such that losing that sale meant
I wouldn't get to keep my 3 yachts". In other words, they're actually
saying that nobody has taken a steel-toed boot and kicked them where
it hurtsーin the walletーover IPv6.

While I have heard the occasional story of withholding payment, and
I'm sure many on this list have such stories, I'm not yet convinced
that these data points have yet amounted to much in the aggregate.
Re: yum IPv6 repos [ In reply to ]
On 02/18/2012 17:44, Erik Kline wrote:
> In other words, they're actually
> saying that nobody has taken a steel-toed boot and kicked them where
> it hurtsーin the walletーover IPv6.

No, the anecdotes I've heard involve plying said salescritter(s) with
sufficient quantity of $BEVERAGE until they admit that upper management
told them to use "No one else is requesting IPv6" as their standard line
so that they can push implementation as far down the road as possible.


Doug (I like your optimism though)

--

It's always a long day; 86400 doesn't fit into a short.

Breadth of IT experience, and depth of knowledge in the DNS.
Yours for the right price. :) http://SupersetSolutions.com/

1 2  View All