On Fri, 4 Oct 2002, John Fraizer wrote:
> OK Hugh. Just so the rest of the world doesn't get confused here:
>
> END-TO-END is the key point when using MTUs >1500.
>
> If you have a path that looks like this (MTUs):
>
> [1500]-[4500]-[4500]-[17994]-[17994]-[4500]-[1500]
>
> ...It doesn't matter what is in the middle, the max packet size
> (unfragmented) that is going to get through the chain is 1500.
>
> With v6-in-v4 tunnels, 1500 won't even make it through.
True enough. We're probably looking at MTU 1476 or somesuch
on these tunnels, regardless of whether or not the underlying
mtu on any particular link is 17994. The tunnel is a virtual
link with a low MTU.
I just got concerned that some folks were generalizing this
phenomenon beyond the domain of tunnels and ethernet. People
go to great lengths sometimes to get END-TO-END 9000 byte MTU
because it makes a huge difference in end-end performance
between hosts separated by a large RTT. Actually, it makes
a big difference even if the hosts are locally connected and
on the same subnet, if the hosts are actually running user
programs instead of just running network benchmarks ;-)
From rsamprat@cisco.com Mon Oct 7 19:24:22 2002
From: rsamprat@cisco.com (Ravikanth Samprathi)
Date: Mon, 07 Oct 2002 11:24:22 -0700
Subject: [6bone] tunnel to 6bone?
Message-ID: <4.3.2.7.2.20021007112044.03f50da8@mira-sjc5-7.cisco.com>
--=====================_351420075==_.ALT
Content-Type: text/plain; charset="us-ascii"; format=flowed
Hi
I have a topology with a set of ipv6-only-hosts
and a set of ipv4-only-hosts as follows:
IPv6 Hosts IPv4 Hosts
| |
| |
| |
+--------+-----------+
|
|
IPv4/IPv6 Dual Stack Gateway
||
||
||
||
||
IPv4/IPv6 Dual Stack Relay Router
|
|
|
6Bone ISP
My goal is to provide global-ipv6-connectivity to the IPv6-hosts.
Approach:
----------------
To accomplish this goal, I have configured a dual-stack-relay-router.
From each of the dual-stack-gateways, i have configured 6to4
tunnels to the dual-stack-relay-router. From the dual-stack-relay-
router, i intend to connect to 6Bone.
If the IPv4 address of dual-stack-gateway is 1.2.3.4, the IPv6
address of the gateway is 2002:0102:0304::1/64, and the prefix delegated
to the IPv6-hosts within that home would be 2002:0102:0304:0001 (64 bits).
If the IPv4 address of dual-stack-relay-router is 5.6.7.8, the IPv6
address of the relay-router is 2002:0506:0708::1/64.
My questions to you:
-------------------------------
On this front, can you please let me know if my approach is right?
With this kind of configuration to the IPv6 hosts, will I
be able to provide global connectivity to the ipv6-hosts?
Is 6to4 tunnels the right approach to use to connect the gateways to
relay-router? And if so, how should i configure the ipv6-routing-
tables in the gateways to forward all the ipv6 traffic to the
relay-router?
How should i configure the relay-router to connect to 6bone? Do we
use native-ipv6 or 6to4-tunnel?
Is there a better approach to solve this problem?
Since i am new to 6bone, i would greatly appreciate any help or pointers.
With best rgds.
Ravi
--=====================_351420075==_.ALT
Content-Type: text/html; charset="us-ascii"
Hi
I have a topology with a set of ipv6-only-hosts
and a set of ipv4-only-hosts as follows:
IPv6 Hosts IPv4
Hosts
|
|
|
|
|
|
+--------+-----------+
|
|
IPv4/IPv6 Dual Stack Gateway
||
||
||
||
||
IPv4/IPv6 Dual Stack Relay Router
|
|
|
6Bone ISP
My goal is to provide global-ipv6-connectivity to the
IPv6-hosts.
Approach:
----------------
To accomplish this goal, I have configured a
dual-stack-relay-router.
From each of the dual-stack-gateways, i have configured
6to4
tunnels to the dual-stack-relay-router. From the
dual-stack-relay-
router, i intend to connect to 6Bone.
If the IPv4 address of dual-stack-gateway is 1.2.3.4, the
IPv6
address of the gateway is 2002:0102:0304::1/64, and the prefix
delegated
to the IPv6-hosts within that home would be 2002:0102:0304:0001 (64
bits).
If the IPv4 address of dual-stack-relay-router is 5.6.7.8, the
IPv6
address of the relay-router is 2002:0506:0708::1/64.
My questions to you:
-------------------------------
On this front, can you please let me know if my approach is
right?
With this kind of configuration to the IPv6 hosts, will I
be able to provide global connectivity to the ipv6-hosts?
Is 6to4 tunnels the right approach to use to connect the gateways
to
relay-router? And if so, how should i configure the
ipv6-routing-
tables in the gateways to forward all the ipv6 traffic to
the
relay-router?
How should i configure the relay-router to connect to 6bone?
Do we
use native-ipv6 or 6to4-tunnel?
Is there a better approach to solve this problem?
Since i am new to 6bone, i would greatly appreciate any help or
pointers.
With best rgds.
Ravi
--=====================_351420075==_.ALT--
From morth@morth.org Tue Oct 8 01:27:11 2002
From: morth@morth.org (Pelle Johansson)
Date: Tue, 8 Oct 2002 02:27:11 +0200
Subject: [6bone] tunnel to 6bone?
In-Reply-To: <4.3.2.7.2.20021007112044.03f50da8@mira-sjc5-7.cisco.com>
Message-ID:
--Apple-Mail-2--104943400
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain;
charset=ISO-8859-1;
format=flowed
m=E5ndagen den 7 oktober 2002 kl 20.24 skrev Ravikanth Samprathi:
> To accomplish this goal, I have configured a dual-stack-relay-router.
> =46rom each of the dual-stack-gateways, i have configured 6to4
> tunnels to the dual-stack-relay-router.=A0 =46rom the =
dual-stack-relay-
> router, i intend to connect to 6Bone.
> If the IPv4 address of dual-stack-gateway is 1.2.3.4, the IPv6
> address of the gateway is 2002:0102:0304::1/64, and the prefix=20
> delegated
> to the IPv6-hosts within that home would be 2002:0102:0304:0001 (64=20
> bits).
> If the IPv4 address of dual-stack-relay-router is 5.6.7.8, the IPv6
> address of the relay-router is 2002:0506:0708::1/64.
If you want all ipv6 traffic to pass through the router both in and=20
out, this is not the right approach. You then need to allocate all your=20=
ipv6 addresses under 2002:0506:0708::/48. Typically you'd give each=20
gateway one or more /64 prefixes (2002:0506:0708:1:: 2002:0506:0708:2::=20=
etc) and use gif tunnels (or similar) to move the traffic between the=20
gateways and the router. If, however, it's fine if the incoming packets=20=
go direct to the gateways you can do it like this (depends on how much=20=
you're firewalling). The best approach would of course to use native=20
IPv6 between the router and the gateways, but I assumed this was not=20
possible? (ie there's some network not under your control between them)
>
> My questions to you:
> -------------------------------
> On this front, can you please let me know if my approach is right?
> With this kind of configuration to the IPv6 hosts, will I
> be able to provide global connectivity to the ipv6-hosts?
>
> Is 6to4 tunnels the right approach to use to connect the gateways to
> relay-router?=A0 And if so, how should i configure the ipv6-routing-
> tables in the gateways to forward all the ipv6 traffic to the
> relay-router?
As I mentioned, typically you'd use gif tunnels or similar instead.=20
Ofc, 6to4 is basically implicit gif tunnels, but they have some=20
security issues (spoofing).
> How should i configure the relay-router to connect to 6bone?=A0 Do we
> use native-ipv6 or 6to4-tunnel?
If your upstream ISP does IPv6 that's the best choice. Otherwise you=20
can either use 6to4 or a tunnel broker (freenet6 seems to be the most=20
popular). I'm not sure what the status of getting reverse dns is when=20
using 6to4, and you have to find some 6to4 router who'll accept your=20
packets.
--=20
Pelle Johansson
--Apple-Mail-2--104943400
Content-Transfer-Encoding: quoted-printable
Content-Type: text/enriched;
charset=ISO-8859-1
m=E5ndagen den 7 oktober 2002 kl 20.24 skrev Ravikanth Samprathi:
To accomplish this goal, I have configured a
dual-stack-relay-router.
=46rom each of the dual-stack-gateways, i have configured 6to4
tunnels to the dual-stack-relay-router.=A0 =46rom the dual-stack-relay-
router, i intend to connect to 6Bone.
If the IPv4 address of dual-stack-gateway is 1.2.3.4, the IPv6
address of the gateway is 2002:0102:0304::1/64, and the prefix
delegated
to the IPv6-hosts within that home would be 2002:0102:0304:0001 (64
bits).
If the IPv4 address of dual-stack-relay-router is 5.6.7.8, the IPv6
address of the relay-router is 2002:0506:0708::1/64.
If you want all ipv6 traffic to pass through the router both in and
out, this is not the right approach. You then need to allocate all
your ipv6 addresses under 2002:0506:0708::/48. Typically you'd give
each gateway one or more /64 prefixes (2002:0506:0708:1::
2002:0506:0708:2:: etc) and use gif tunnels (or similar) to move the
traffic between the gateways and the router. If, however, it's fine if
the incoming packets go direct to the gateways you can do it like this
(depends on how much you're firewalling). The best approach would of
course to use native IPv6 between the router and the gateways, but I
assumed this was not possible? (ie there's some network not under your
control between them)
My questions to you:
-------------------------------
On this front, can you please let me know if my approach is right?
With this kind of configuration to the IPv6 hosts, will I
be able to provide global connectivity to the ipv6-hosts?
Is 6to4 tunnels the right approach to use to connect the gateways to
relay-router?=A0 And if so, how should i configure the ipv6-routing-
tables in the gateways to forward all the ipv6 traffic to the
relay-router?
As I mentioned, typically you'd use gif tunnels or similar
instead. Ofc, 6to4 is basically implicit gif tunnels, but they have
some security issues (spoofing).
How should i configure the relay-router to connect to 6bone?=A0
Do we
use native-ipv6 or 6to4-tunnel?
If your upstream ISP does IPv6 that's the best choice.
Otherwise you can either use 6to4 or a tunnel broker (freenet6 seems
to be the most popular). I'm not sure what the status of getting
reverse dns is when using 6to4, and you have to find some 6to4 router
who'll accept your packets.
--=20
Pelle Johansson
<
--Apple-Mail-2--104943400--
From kenpohniman@yahoo.com Tue Oct 8 13:17:37 2002
From: kenpohniman@yahoo.com (Ken Pohniman)
Date: Tue, 8 Oct 2002 20:17:37 +0800
Subject: [6bone] tunnel to 6bone?
In-Reply-To: <4.3.2.7.2.20021007112044.03f50da8@mira-sjc5-7.cisco.com>
Message-ID: <000401c26ec4$b67f8110$0365a8c0@laptop>
To provide ipv6 connectivity to 6bone, you can use the following
methods:
Method1:
Ipv6hosts---6to4router---(internet)---6to4relay---(6bone)
The router will be configured with 6to4, and advertise the 6to4 prefix
(depending on the router's public v4 address), to the v6 hosts. For the
hosts to connect to a 6to4 address (eg. 2002:...) in 6bone, the router
will perform 6to4 tunneling to the destination. However, to access a
global ipv6 address (eg. 2001:...), you'll also need to add a default v6
route on the router, pointing to the 6to4relay router (eg.
::/192.88.99.1).
Method2:
Ipv6hosts---(internet)---6to4relay---(6bone)
You can treat the MS XP host like a router, and configure 6to4 and the
necessary routing similar to method 1. Seems that with XP service pack
1, XP will automatically configure itself whenever it gets a public ipv4
address. When you do a ping6 (eg. Ping6 www.kame.net), you should be
able to get an ipv6 reply.
Regards,
Ken
-----Original Message-----
From: 6bone-admin@mailman.isi.edu [mailto:6bone-admin@mailman.isi.edu]
On Behalf Of Ravikanth Samprathi
Sent: Tuesday, October 08, 2002 2:24 AM
To: 6bone@mailman.isi.edu
Subject: [6bone] tunnel to 6bone?
Hi
I have a topology with a set of ipv6-only-hosts
and a set of ipv4-only-hosts as follows:
IPv6 Hosts IPv4 Hosts
| |
| |
| |
+--------+-----------+
|
|
IPv4/IPv6 Dual Stack Gateway
||
||
||
||
||
IPv4/IPv6 Dual Stack Relay Router
|
|
|
6Bone ISP
My goal is to provide global-ipv6-connectivity to the IPv6-hosts.
Approach:
----------------
To accomplish this goal, I have configured a dual-stack-relay-router.
>From each of the dual-stack-gateways, i have configured 6to4
tunnels to the dual-stack-relay-router. From the dual-stack-relay-
router, i intend to connect to 6Bone.
If the IPv4 address of dual-stack-gateway is 1.2.3.4, the IPv6
address of the gateway is 2002:0102:0304::1/64, and the prefix delegated
to the IPv6-hosts within that home would be 2002:0102:0304:0001 (64
bits).
If the IPv4 address of dual-stack-relay-router is 5.6.7.8, the IPv6
address of the relay-router is 2002:0506:0708::1/64.
My questions to you:
-------------------------------
On this front, can you please let me know if my approach is right?
With this kind of configuration to the IPv6 hosts, will I
be able to provide global connectivity to the ipv6-hosts?
Is 6to4 tunnels the right approach to use to connect the gateways to
relay-router? And if so, how should i configure the ipv6-routing-
tables in the gateways to forward all the ipv6 traffic to the
relay-router?
How should i configure the relay-router to connect to 6bone? Do we
use native-ipv6 or 6to4-tunnel?
Is there a better approach to solve this problem?
Since i am new to 6bone, i would greatly appreciate any help or
pointers.
With best rgds.
Ravi
From bmanning@ISI.EDU Tue Oct 8 18:23:07 2002
From: bmanning@ISI.EDU (Bill Manning)
Date: Tue, 8 Oct 2002 10:23:07 -0700 (PDT)
Subject: AW: [6bone] big packets - redux
In-Reply-To: <73D3E97F639DD5119642000347055F0505BAB612@G9JNS.mgb01.telekom.de> from "Scheffler, Thomas" at "Oct 8, 2 10:57:56 am"
Message-ID: <200210081723.g98HN7Q10549@boreas.isi.edu>
key length (in bits) generates KEY/SIG rrs that are in bytes.
for example:
$dig . ns
...
;; WHEN: Tue Oct 8 10:17:24 2002
;; MSG SIZE rcvd: 196
while:
$dig . ns +dnssec
...
;; WHEN: Tue Oct 8 10:18:51 2002
;; MSG SIZE rcvd: 1822
the change in bytes is nearly an order of magnitude
and the key length was 512 bits.
this triggers UDP fragmentation.
% To me there seems to be a confusion of units in this
% posting.
% Encryption keys are typically measured in bits,
% whereas MTU size is in bytes.
%
% It should therefore be possible to send all, but
% the largest keys unfragmented in one paket.
%
% Regards,
% Thomas
%
%
% > -----Ursprungliche Nachricht-----
% > Von: Bill Manning [mailto:bmanning@ISI.EDU]
% > Gesendet: Samstag, 5. Oktober 2002 01:44
% > An: 6bone@ISI.EDU
% > Betreff: [6bone] big packets - redux
% >
% >
% > once more, with big packets....
% > the topic was discussed on the keydist list. Feedback is
% > appreciated.
% > --------
% > experimenting with DS keys of various lengths shows that
% > when keys are over a certain size, UDP fragmentation sets
% > in. In some cases, it is possible to actually get rollover
% > to TCP (although this seems to be a corner case)
% >
% > now I've been told that UDP fragmentation can be a bad thing,
% > leading to all sorts (well some kinds anyway) of odd
% > operational failures that are hard to debug. UDP failure
% > and rolling over to TCP is also considered a bad thing.
% >
% > so this question, "should key lengths be selected to
% > avoid fragmentation/TCP use?"
% >
% > if so, why?
% > if not, why not?
% >
% > testing was done using single keys. RSA/SHA1 keys of
% > 512 & 1024
% > bits, which generated "reasonable" packets. RSA/SHA1
% > keys of 4096
% > bits, generated UDP fragmentation. Multiple keys will
% > aggrevate the
% > issue.
% >
% > Somewhat sidenote: there has been some discussion
% > that RSA keys
% > over 2048 in length incurred a significant
% > performance impact over
% > smaller keys. This performance impact hits the
% > resolver though,
% > "on the wire" usage wasn't mentioned. Further
% > testing is needed.
% >
% >
% > Thoughtful responses follow:
% >
% > - "no" I don't think operational issues should dictate key
% > lengths, but huge keys don't necessarily mean more secure either :)
% >
% > - some IDS/firewalls toss UDP packets larger than 512 bytes.
% > Maybe the
% > right answer is to tune the EDNS packet size to avoid UDP
% > fragmentation? 4096 is bigger than most MTUs, but 1280
% > probably isn't,
% > and should be enough for most common responses.
% >
% > - this is not just a server/resolver issue. if transit
% > infrastructure is
% > making assumptions on "viable" datagram sizes, we will have
% > to make a
% > tradeoff in recommended key lengths.
% > In an ideal world, security reasons (whatever that means)
% > may be the only
% > vector for selecting length. This being the "real" world,
% > it may be that
% > to be useful, one has to trade of between "enough"
% > crypto-strength and the
% > ability to deliver the key(s) to the intended target.)
% >
% > - one experience is that keeping DNSSEC messages (plus
% > overhead) below a
% > MTU of 1500 can be sort of difficult and too restrictive besides.
% >
% > - a general opinion is that the firewall or router that drops
% > UDP fragments
% > or large UDP packets is broken and will need to be
% > upgraded/replaced.
% > if clients behind such broken devices set their EDNS0 max
% > buffer size to
% > something that will fit in the MTU, everything should work.
% > There will
% > probably be a lot of TCP DNS traffic, but it should work.
% >
% > Acknowledgements:
% >
% > David Blacka Scott Rose Brian Wellington
% > Mark Andrews Ed Lewis Joahn Ihren Olaf M. Kolkman
% >
% > --bill
% > _______________________________________________
% > 6bone mailing list
% > 6bone@mailman.isi.edu
% > http://mailman.isi.edu/mailman/listinfo/6bone
% >
%
--
--bill
From rsamprat@cisco.com Wed Oct 9 01:21:20 2002
From: rsamprat@cisco.com (Ravi Samprathi)
Date: Tue, 8 Oct 2002 17:21:20 -0700 (PDT)
Subject: [6bone] tunnel to 6bone?
In-Reply-To: <000401c26ec4$b67f8110$0365a8c0@laptop>
Message-ID:
Thanks.
I will look into this.
Best rgds.
Ravi
On Tue, 8 Oct 2002, Ken Pohniman wrote:
> To provide ipv6 connectivity to 6bone, you can use the following
> methods:
>
> Method1:
> Ipv6hosts---6to4router---(internet)---6to4relay---(6bone)
> The router will be configured with 6to4, and advertise the 6to4 prefix
> (depending on the router's public v4 address), to the v6 hosts. For the
> hosts to connect to a 6to4 address (eg. 2002:...) in 6bone, the router
> will perform 6to4 tunneling to the destination. However, to access a
> global ipv6 address (eg. 2001:...), you'll also need to add a default v6
> route on the router, pointing to the 6to4relay router (eg.
> ::/192.88.99.1).
>
> Method2:
> Ipv6hosts---(internet)---6to4relay---(6bone)
> You can treat the MS XP host like a router, and configure 6to4 and the
> necessary routing similar to method 1. Seems that with XP service pack
> 1, XP will automatically configure itself whenever it gets a public ipv4
> address. When you do a ping6 (eg. Ping6 www.kame.net), you should be
> able to get an ipv6 reply.
>
> Regards,
> Ken
>
>
> -----Original Message-----
> From: 6bone-admin@mailman.isi.edu [mailto:6bone-admin@mailman.isi.edu]
> On Behalf Of Ravikanth Samprathi
> Sent: Tuesday, October 08, 2002 2:24 AM
> To: 6bone@mailman.isi.edu
> Subject: [6bone] tunnel to 6bone?
>
> Hi
> I have a topology with a set of ipv6-only-hosts
> and a set of ipv4-only-hosts as follows:
>
> IPv6 Hosts IPv4 Hosts
> | |
> | |
> | |
> +--------+-----------+
> |
> |
> IPv4/IPv6 Dual Stack Gateway
> ||
> ||
> ||
> ||
> ||
> IPv4/IPv6 Dual Stack Relay Router
> |
> |
> |
> 6Bone ISP
>
> My goal is to provide global-ipv6-connectivity to the IPv6-hosts.
>
> Approach:
> ----------------
> To accomplish this goal, I have configured a dual-stack-relay-router.
> From each of the dual-stack-gateways, i have configured 6to4
> tunnels to the dual-stack-relay-router. From the dual-stack-relay-
> router, i intend to connect to 6Bone.
> If the IPv4 address of dual-stack-gateway is 1.2.3.4, the IPv6
> address of the gateway is 2002:0102:0304::1/64, and the prefix delegated
> to the IPv6-hosts within that home would be 2002:0102:0304:0001 (64
> bits).
> If the IPv4 address of dual-stack-relay-router is 5.6.7.8, the IPv6
> address of the relay-router is 2002:0506:0708::1/64.
>
> My questions to you:
> -------------------------------
> On this front, can you please let me know if my approach is right?
> With this kind of configuration to the IPv6 hosts, will I
> be able to provide global connectivity to the ipv6-hosts?
>
> Is 6to4 tunnels the right approach to use to connect the gateways to
> relay-router? And if so, how should i configure the ipv6-routing-
> tables in the gateways to forward all the ipv6 traffic to the
> relay-router?
>
> How should i configure the relay-router to connect to 6bone? Do we
> use native-ipv6 or 6to4-tunnel?
>
> Is there a better approach to solve this problem?
>
> Since i am new to 6bone, i would greatly appreciate any help or
> pointers.
> With best rgds.
> Ravi
>
>
> _______________________________________________
> 6bone mailing list
> 6bone@mailman.isi.edu
> http://mailman.isi.edu/mailman/listinfo/6bone
>
From fink@es.net Wed Oct 9 06:46:18 2002
From: fink@es.net (Bob Fink)
Date: Tue, 08 Oct 2002 22:46:18 -0700
Subject: [6bone] 6bone pTLA 3FFE:4012::/32 allocated to EUROVIEW-GROUP
Message-ID: <5.1.0.14.0.20021008224111.025813a8@imap2.es.net>
EUROVIEW-GROUP has been allocated pTLA 3FFE:4012::/32 having finished its
2-week review period.
Note that it will take a short while for their pTLA inet6num entry to
appear in the 6bone registry as they have to create it themselves. However,
their registration is listed on:
[To create a reverse DNS registration for pTLAs, please send the prefix
allocated above, and a list of at least two authoritative nameservers, to
hostmaster@ep.net.]
Thanks,
Bob
From anil.bhaskar@wipro.com Fri Oct 11 05:35:37 2002
From: anil.bhaskar@wipro.com (Anil B)
Date: Fri, 11 Oct 2002 10:05:37 +0530
Subject: [6bone] Want help in setting Mozilla browser for IPv6.
In-Reply-To:
Message-ID: <005701c270df$a53b4ca0$1a14c80a@wipro.com>
This is a multi-part message in MIME format.
------=_NextPartTM-000-255efa5b-7fa9-4d42-b3f3-b7622757790f
Content-Type: text/plain;
charset="us-ascii"
Content-Transfer-Encoding: 7bit
Hi,
Can any body help me in setting Mozilla browser for IPv6; I am still
troubling with this on Linux M/C. I have IPv6 enable IE on Windows.
Thanks and Best Regards,
Anil B.
*******************************************
Manager-Talent Transformation
Mezzanine Floor, Floating Learning Centre,
Wipro Tech.
Electronics City-2
Board No. 8520408-Ext: 5438
Mob: 9844003364
*******************************************
------=_NextPartTM-000-255efa5b-7fa9-4d42-b3f3-b7622757790f
Content-Type: text/plain;
name="Wipro_Disclaimer.txt"
Content-Transfer-Encoding: 7bit
Content-Disposition: attachment;
filename="Wipro_Disclaimer.txt"
**************************Disclaimer************************************
Information contained in this E-MAIL being proprietary to Wipro Limited is
'privileged' and 'confidential' and intended for use only by the individual
or entity to which it is addressed. You are notified that any use, copying
or dissemination of the information contained in the E-MAIL in any manner
whatsoever is strictly prohibited.
***************************************************************************
------=_NextPartTM-000-255efa5b-7fa9-4d42-b3f3-b7622757790f--
From nicolas.deffayet@ndsoftware.net Fri Oct 11 14:20:41 2002
From: nicolas.deffayet@ndsoftware.net (Nicolas DEFFAYET)
Date: 11 Oct 2002 15:20:41 +0200
Subject: [6bone] Want help in setting Mozilla browser for IPv6.
In-Reply-To: <005701c270df$a53b4ca0$1a14c80a@wipro.com>
References: <005701c270df$a53b4ca0$1a14c80a@wipro.com>
Message-ID: <1034342441.667.6725.camel@wks1.fr.corp.ndsoftware.com>
On Fri, 2002-10-11 at 06:35, Anil B wrote:
Hi,
> Can any body help me in setting Mozilla browser for IPv6; I am still
> troubling with this on Linux M/C. I have IPv6 enable IE on Windows.
Mozilla have a native IPv6 support (only in Unix version), no need set
an option.
What's version on Mozilla do you use ?
The Windows version of Mozilla don't support IPv6.
Best Regards,
Nicolas DEFFAYET
http://noc.ndsoftwarenet.com/
FNIX6, the first french IPv6 Internet Exchange: http://www.fnix6.net/
From Ron.Barker@v-pe.de Fri Oct 11 14:20:44 2002
From: Ron.Barker@v-pe.de (Barker, Ron, vpe)
Date: Fri, 11 Oct 2002 15:20:44 +0200
Subject: [6bone] .NET RC1 IPv6 6to4
Message-ID: <55D9F49A60EF2A44BF1AC7F348D9385227CD07@mail.v-pe.de>
This is a multi-part message in MIME format.
------_=_NextPart_001_01C27129.00DBCE67
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Has anyone been able to get connected to the 6bone with .NET RC1 IPv6. =
I have tried in vain for 2 day to no avail. Some really strange things =
happens depending on whether the computer is / is not
attached to the internet. Really strange is the fact that prior to =
plugging in the ethernet cable the 6to4 interface is disable, however, =
whenever I have link to the net, the 6to4 has two strange
looking default routes that are automatically generated:
Publish Type Met Prefix Idx =
Gateway/Interface Name
------- -------- ---- ------------------------ --- =
---------------------
yes Manual 1221 ::/0 3 =
2002:836b:213c:1:e0:8f08:f020:8
yes Manual 2147483648 ::/0 3 =
2002:c058:6301::c058:6301
yes Manual 1001 2002::/16 3 6to4 Tunneling =
Pseudo-Interface
netsh interface ipv6>
1. is a ::/0 route to what is obviously a rely router at MS - =
131.107.33.60 - tried pinging but no answer
2- is a ../0 route to a 192.xxx. route ?????=20
Where do these addresses come from???? magic.
Additionaly, I miss any any all reference to setting up the tunnel.
Any comments appreciated
Ron=20
Dr. Ronald D. Barker
Vodafone Pilotentwicklung GmbH
Chiemgaustr. 116
D-81549 Munich
Germany
Fon +49 (89) 95 410 -0
Fax +49 (89) 95 410 -111
www.v-pe.de
------_=_NextPart_001_01C27129.00DBCE67
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
.NET RC1 IPv6 6to4
Has anyone been able to get connected =
to the 6bone with .NET RC1 IPv6. I have tried in vain for 2 =
day to no avail. Some really strange things happens depending on =
whether the computer is / is not attached to the internet. Really =
strange is the fact that prior to plugging in the ethernet cable the =
6to4 interface is disable, however, whenever I have link to =
the net, the 6to4 has two strange looking default routes that are =
automatically generated:
Publish =
Type Met =
Prefix &=
nbsp; Idx =
Gateway/Interface Name
------- -------- =
---- ------------------------ --- =
---------------------
yes =
Manual 1221 =
=
::/0 &nb=
sp; =
3 2002:836b:213c:1:e0:8f08:f020:8
yes =
Manual 2147483648 =
::/0 &nb=
sp; 3 =
2002:c058:6301::c058:6301
yes =
Manual 1001 =
2002::/16 &nbs=
p; 3 6to4 Tunneling =
Pseudo-Interface
netsh interface ipv6>
1. is a ::/0 route to what is =
obviously a rely router at MS - 131.107.33.60 - tried pinging but =
no answer
2- is a ../0 route to a 192.xxx. =
route ?????
Where do these addresses come from???? =
magic.
Additionaly, I miss any any all =
reference to setting up the tunnel.
Any comments appreciated
Ron
Dr. Ronald D. Barker
Vodafone Pilotentwicklung GmbH
Chiemgaustr. 116
D-81549 Munich
Germany
Fon +49 (89) 95 410 -0
Fax +49 (89) 95 410 -111
www.v-pe.de
------_=_NextPart_001_01C27129.00DBCE67--
From bortzmeyer@gitoyen.net Fri Oct 11 15:13:43 2002
From: bortzmeyer@gitoyen.net (Stephane Bortzmeyer)
Date: Fri, 11 Oct 2002 16:13:43 +0200
Subject: [6bone] Want help in setting Mozilla browser for IPv6.
In-Reply-To: <1034342441.667.6725.camel@wks1.fr.corp.ndsoftware.com>
References: <005701c270df$a53b4ca0$1a14c80a@wipro.com> <1034342441.667.6725.camel@wks1.fr.corp.ndsoftware.com>
Message-ID: <20021011141343.GA18732@nic.fr>
On Fri, Oct 11, 2002 at 03:20:41PM +0200,
Nicolas DEFFAYET wrote
a message of 23 lines which said:
> FNIX6, the first french IPv6 Internet Exchange: http://www.fnix6.net/
Wrong, the first one was Sfinx , several
months ago.
From nicolas.deffayet@ndsoftware.net Fri Oct 11 15:16:25 2002
From: nicolas.deffayet@ndsoftware.net (Nicolas DEFFAYET)
Date: 11 Oct 2002 16:16:25 +0200
Subject: [6bone] .NET RC1 IPv6 6to4
In-Reply-To: <55D9F49A60EF2A44BF1AC7F348D9385227CD07@mail.v-pe.de>
References: <55D9F49A60EF2A44BF1AC7F348D9385227CD07@mail.v-pe.de>
Message-ID: <1034345786.664.6745.camel@wks1.fr.corp.ndsoftware.com>
On Fri, 2002-10-11 at 15:20, Barker, Ron, vpe wrote:
> Has anyone been able to get connected to the 6bone with .NET RC1 IPv6. I have tried in vain for 2 day to no avail. Some really strange things happens depending on whether the computer is / is not
> attached to the internet. Really strange is the fact that prior to plugging in the ethernet cable the 6to4 interface is disable, however, whenever I have link to the net, the 6to4 has two strange
> looking default routes that are automatically generated:
>
> Publish Type Met Prefix Idx Gateway/Interface Name
> ------- -------- ---- ------------------------ --- ---------------------
> yes Manual 1221 ::/0 3 2002:836b:213c:1:e0:8f08:f020:8
> yes Manual 2147483648 ::/0 3 2002:c058:6301::c058:6301
> yes Manual 1001 2002::/16 3 6to4 Tunneling Pseudo-Interface
>
> netsh interface ipv6>
>
>
> 1. is a ::/0 route to what is obviously a rely router at MS - 131.107.33.60 - tried pinging but no answer
> 2- is a ../0 route to a 192.xxx. route ?????
>
> Where do these addresses come from???? magic.
>
> Additionaly, I miss any any all reference to setting up the tunnel.
For setup a IPv6 over IPv4 tunnel on Windows .NET:
In Windows command line:
netsh
netsh> interface
netsh interface> ipv6
netsh interface ipv6> install
netsh interface ipv6> add v6v4tunnel "IPv6"
netsh interface ipv6> add address "IPv6"
netsh interface ipv6> add route ::/0 "IPv6
I didn't have yet time to test IPv6 with Windows .NET, but the
NDSoftware Research team have do it.
Best Regards,
Nicolas DEFFAYET
http://noc.ndsoftwarenet.com/
FNIX6, the first french IPv6 Internet Exchange: http://www.fnix6.net/
From nicolas.deffayet@ndsoftware.net Fri Oct 11 15:30:30 2002
From: nicolas.deffayet@ndsoftware.net (Nicolas DEFFAYET)
Date: 11 Oct 2002 16:30:30 +0200
Subject: [6bone] Want help in setting Mozilla browser for IPv6.
In-Reply-To: <20021011141343.GA18732@nic.fr>
References: <005701c270df$a53b4ca0$1a14c80a@wipro.com>
<1034342441.667.6725.camel@wks1.fr.corp.ndsoftware.com>
<20021011141343.GA18732@nic.fr>
Message-ID: <1034346630.1667.7.camel@wks1.fr.corp.ndsoftware.com>
On Fri, 2002-10-11 at 16:13, Stephane Bortzmeyer wrote:
> On Fri, Oct 11, 2002 at 03:20:41PM +0200,
> Nicolas DEFFAYET wrote
> a message of 23 lines which said:
>
> > FNIX6, the first french IPv6 Internet Exchange: http://www.fnix6.net/
>
> Wrong, the first one was Sfinx , several
> months ago.
SFINX is a IPv4 and IPv6 Internet Exchange, it's not a IPv6 Internet
Exchange.
IPv6 Internet Exchange = IPv6 only
FNIX6 is the first french IPv6 Internet Exchange and the second IPv6
Internet Exchange in Europe (the first is UK6x).
For information, a list of IPv6 Internet Exchange:
- 6IIX (New-York, Los Angeles, Santa Clara, US)
- 6NGIX (Seoul, KR)
- 6TAP (Chicago, US)
- KNIX6 (Soul, KR)
- NSPIXP-6 (Tokyo, JP)
- NY6IX (New-York, US)
- S-IX (San Jose,US)
- UK6X (London, UK)
Best Regards,
Nicolas DEFFAYET
http://noc.ndsoftwarenet.com/
FNIX6, the first french IPv6 Internet Exchange: http://www.fnix6.net/
From pekkas@netcore.fi Fri Oct 11 15:35:29 2002
From: pekkas@netcore.fi (Pekka Savola)
Date: Fri, 11 Oct 2002 17:35:29 +0300 (EEST)
Subject: [6bone] .NET RC1 IPv6 6to4
In-Reply-To: <55D9F49A60EF2A44BF1AC7F348D9385227CD07@mail.v-pe.de>
Message-ID:
Note '2002:c058:6301::c058:6301' below.
That only works with Microsoft relays because there's a completely bogus
and non-interoperable assumption that the relay has address
'2002:c058:6301::c058:6301' and it's pingable.
Not a good way to push IPv6 *at all*...
On Fri, 11 Oct 2002, Barker, Ron, vpe wrote:
> Has anyone been able to get connected to the 6bone with .NET RC1 IPv6. I have tried in vain for 2 day to no avail. Some really strange things happens depending on whether the computer is / is not
> attached to the internet. Really strange is the fact that prior to plugging in the ethernet cable the 6to4 interface is disable, however, whenever I have link to the net, the 6to4 has two strange
> looking default routes that are automatically generated:
>
> Publish Type Met Prefix Idx Gateway/Interface Name
> ------- -------- ---- ------------------------ --- ---------------------
> yes Manual 1221 ::/0 3 2002:836b:213c:1:e0:8f08:f020:8
> yes Manual 2147483648 ::/0 3 2002:c058:6301::c058:6301
> yes Manual 1001 2002::/16 3 6to4 Tunneling Pseudo-Interface
>
> netsh interface ipv6>
>
>
> 1. is a ::/0 route to what is obviously a rely router at MS - 131.107.33.60 - tried pinging but no answer
> 2- is a ../0 route to a 192.xxx. route ?????
>
> Where do these addresses come from???? magic.
>
> Additionaly, I miss any any all reference to setting up the tunnel.
>
>
> Any comments appreciated
>
> Ron
>
>
>
>
>
> Dr. Ronald D. Barker
> Vodafone Pilotentwicklung GmbH
> Chiemgaustr. 116
> D-81549 Munich
> Germany
> Fon +49 (89) 95 410 -0
> Fax +49 (89) 95 410 -111
> www.v-pe.de
>
>
>
--
Pekka Savola "Tell me of difficulties surmounted,
Netcore Oy not those you stumble over and fall"
Systems. Networks. Security. -- Robert Jordan: A Crown of Swords
From owens@nysernet.org Fri Oct 11 16:07:03 2002
From: owens@nysernet.org (Bill Owens)
Date: Fri, 11 Oct 2002 11:07:03 -0400
Subject: [6bone] .NET RC1 IPv6 6to4
In-Reply-To:
References:
Message-ID:
At 17:35 +0300 10/11/02, Pekka Savola wrote:
>Note '2002:c058:6301::c058:6301' below.
>
>That only works with Microsoft relays because there's a completely bogus
>and non-interoperable assumption that the relay has address
>'2002:c058:6301::c058:6301' and it's pingable.
Are you sure? I had it from MS that they ping the IPv4 address of the
relay, not v6.
Bill.
From mlehman@microsoft.com Fri Oct 11 16:16:58 2002
From: mlehman@microsoft.com (Matthew Lehman)
Date: Fri, 11 Oct 2002 08:16:58 -0700
Subject: [6bone] .NET RC1 IPv6 6to4
Message-ID:
It's a holdover from the original research stack. However it doesn't
work quite like you've detailed. When the 6to4 auto-tunneling interface
comes up, it attempts to discover 6to4 relays via a DNS lookup to
6to4.ipv6.microsoft.com. If you lookup that record it resolves to the
MS relay (131.107.33.60) and the 6to4 anycast address. It attempts to
ping the v4 addresses that are returned. If you want to override this
behavior, you can set the relay manually:
netsh interface ipv6 6to4 set relay enabled
My machines at home are setup to go to the MS relay because I know it's
topologically close and the closest advertised relay using the 6to4
anycast address is at SWITCH which is a long way from home.
To setup a configured tunnel, follow the instructions in Nicolas' email.
You can find most of this information in the help section of .Net
server. Just search for IPv6 and/or 6to4 via the help interface.
> -----Original Message-----
> From: Pekka Savola [mailto:pekkas@netcore.fi]
> Sent: Friday, October 11, 2002 7:35 AM
> To: Barker, Ron, vpe
> Cc: 6bone@mailman.isi.edu
> Subject: Re: [6bone] .NET RC1 IPv6 6to4
>
> Note '2002:c058:6301::c058:6301' below.
>
> That only works with Microsoft relays because there's a completely
bogus
> and non-interoperable assumption that the relay has address
> '2002:c058:6301::c058:6301' and it's pingable.
>
> Not a good way to push IPv6 *at all*...
>
> On Fri, 11 Oct 2002, Barker, Ron, vpe wrote:
>
> > Has anyone been able to get connected to the 6bone with .NET RC1
IPv6.
> I have tried in vain for 2 day to no avail. Some really strange
things
> happens depending on whether the computer is / is not
> > attached to the internet. Really strange is the fact that prior to
> plugging in the ethernet cable the 6to4 interface is disable,
however,
> whenever I have link to the net, the 6to4 has two strange
> > looking default routes that are automatically generated:
> >
> > Publish Type Met Prefix Idx
> Gateway/Interface Name
> > ------- -------- ---- ------------------------ ---
----------------
> -----
> > yes Manual 1221 ::/0 3
> 2002:836b:213c:1:e0:8f08:f020:8
> > yes Manual 2147483648 ::/0 3
> 2002:c058:6301::c058:6301
> > yes Manual 1001 2002::/16 3 6to4
Tunneling
> Pseudo-Interface
> >
> > netsh interface ipv6>
> >
> >
> > 1. is a ::/0 route to what is obviously a rely router at MS -
> 131.107.33.60 - tried pinging but no answer
> > 2- is a ../0 route to a 192.xxx. route ?????
> >
> > Where do these addresses come from???? magic.
> >
> > Additionaly, I miss any any all reference to setting up the tunnel.
> >
> >
> > Any comments appreciated
> >
> > Ron
> >
> >
> >
> >
> >
> > Dr. Ronald D. Barker
> > Vodafone Pilotentwicklung GmbH
> > Chiemgaustr. 116
> > D-81549 Munich
> > Germany
> > Fon +49 (89) 95 410 -0
> > Fax +49 (89) 95 410 -111
> > www.v-pe.de
> >
> >
> >
>
> --
> Pekka Savola "Tell me of difficulties surmounted,
> Netcore Oy not those you stumble over and fall"
> Systems. Networks. Security. -- Robert Jordan: A Crown of Swords
>
> _______________________________________________
> 6bone mailing list
> 6bone@mailman.isi.edu
> http://mailman.isi.edu/mailman/listinfo/6bone
From pekkas@netcore.fi Fri Oct 11 16:35:10 2002
From: pekkas@netcore.fi (Pekka Savola)
Date: Fri, 11 Oct 2002 18:35:10 +0300 (EEST)
Subject: [6bone] .NET RC1 IPv6 6to4
In-Reply-To:
Message-ID:
On Fri, 11 Oct 2002, Matthew Lehman wrote:
> It's a holdover from the original research stack. However it doesn't
> work quite like you've detailed. When the 6to4 auto-tunneling interface
> comes up, it attempts to discover 6to4 relays via a DNS lookup to
> 6to4.ipv6.microsoft.com. If you lookup that record it resolves to the
> MS relay (131.107.33.60) and the 6to4 anycast address. It attempts to
> ping the v4 addresses that are returned. If you want to override this
> behavior, you can set the relay manually:
>
> netsh interface ipv6 6to4 set relay enabled
>
Has that behaviour been changed lately? At least Windows 2000 and Windows
XP "operate" in a completely interoperable manner. We have tcpdumps to
prove that.
FWIW, in our environment, we are using 192.88.99.1, and our relay does
*not* have 2002:192.88.99.1::192.168.88.99.1 (ie:
2002:c058:6301::c058:6301) configured.
--
Pekka Savola "Tell me of difficulties surmounted,
Netcore Oy not those you stumble over and fall"
Systems. Networks. Security. -- Robert Jordan: A Crown of Swords
From rsamprat@cisco.com Fri Oct 11 19:02:09 2002
From: rsamprat@cisco.com (Ravi Samprathi)
Date: Fri, 11 Oct 2002 11:02:09 -0700 (PDT)
Subject: [6bone] IPv6 DNS forwarders
In-Reply-To:
Message-ID:
Hi
In my home gateway, I have installed and configured bind-9.2.0
DNS server for both IPv4 and IPv6. I have setup IPv4 forwarders
as follows:
forwarders {
1.2.3.4;
3.4.5.6;
};
>From my home gateway, i also have a 6to4-tunnel to a 6to4
relay-router, which inturn has tunnel to 6bone.
Can anyone please let me know how to setup such DNS forwarders
for IPv6? Are there any IPv6 root servers? Are there IPv6 DNS
servers in 6bone? How can i configure my IPv6 DNS server to
be cognizant of IPv6 DNS forwarders? How can my IPv6-only hosts
resolve the global IPv6 addresses?
Thanks and best rgds.
Ravi
On Fri, 11 Oct 2002, Bill Owens wrote:
> At 17:35 +0300 10/11/02, Pekka Savola wrote:
> >Note '2002:c058:6301::c058:6301' below.
> >
> >That only works with Microsoft relays because there's a completely bogus
> >and non-interoperable assumption that the relay has address
> >'2002:c058:6301::c058:6301' and it's pingable.
>
> Are you sure? I had it from MS that they ping the IPv4 address of the
> relay, not v6.
>
> Bill.
> _______________________________________________
> 6bone mailing list
> 6bone@mailman.isi.edu
> http://mailman.isi.edu/mailman/listinfo/6bone
>
From bmanning@ISI.EDU Fri Oct 11 20:42:51 2002
From: bmanning@ISI.EDU (Bill Manning)
Date: Fri, 11 Oct 2002 12:42:51 -0700 (PDT)
Subject: [6bone] IPv6 DNS forwarders
In-Reply-To: from Ravi Samprathi at "Oct 11, 2 11:02:09 am"
Message-ID: <200210111942.g9BJgpv13510@boreas.isi.edu>
% Can anyone please let me know how to setup such DNS forwarders
% for IPv6? Are there any IPv6 root servers? Are there IPv6 DNS
% servers in 6bone? How can i configure my IPv6 DNS server to
% be cognizant of IPv6 DNS forwarders? How can my IPv6-only hosts
% resolve the global IPv6 addresses?
First, bind 9.2.0 has known problems. You will want to be running
9.2.1.
In order:
) read the ARM that comes with the BIND code for forwarder setup.
) Yes, but only on an experimental basis.
) Lots.
) read the ARM that comes with the BIND code.
) use v6 aware nameservers.
You really should be talking to the v6 support folks in your company.
There is a wealth of support there. They can give you specifics when
answering your questions.
--bill
From Ron.Barker@v-pe.de Mon Oct 14 07:24:23 2002
From: Ron.Barker@v-pe.de (Barker, Ron, vpe)
Date: Mon, 14 Oct 2002 08:24:23 +0200
Subject: [6bone] AW: IPv6 DNS forwarders
Message-ID: <55D9F49A60EF2A44BF1AC7F348D9385227CD08@mail.v-pe.de>
Thanks to all for the comments,
Something strange again happenen on Fri afteranoon. The ping to the MS Relay Router started working. Once I was able to run a few tracesrts it became obvious what the corresponding passages in the
plethora of MS Docs were trying to tell me, e.g. that it is going to look for a relay router etc. In addition to Pekkas comment, to which I subscribe whole heartedly, I should like to add that
notwithstanding the fact that this is a pre-release version, the deviation in operational behaviour between .NET and others, especially the numerous xxxBSDs, ought to occupy a prime position in
introduction / summary sections of the MS IPv6 documentaton, something that Vincent Pice and the Carnige Institute would applaud in catagory of "...winning friends..etc". For Instance:
Is this reference to the MS relay router hard coded?? Perhaps a "persistent" relic of the test enviorment ( which is what I suspect ) that winds up on the CD. Note that whenever
netsh.interface.ipv6.6to4 show relay is run, the very informative result is !default!
Oh well, on the next chore.
ron
-----Ursprüngliche Nachricht-----
Von: Ravi Samprathi [mailto:rsamprat@cisco.com]
Gesendet: Freitag, 11. Oktober 2002 20:02
An: Bill Owens
Cc: Pekka Savola; Barker, Ron, vpe; 6bone@mailman.isi.edu; Ravi
Samprathi
Betreff: IPv6 DNS forwarders
Hi
In my home gateway, I have installed and configured bind-9.2.0
DNS server for both IPv4 and IPv6. I have setup IPv4 forwarders
as follows:
forwarders {
1.2.3.4;
3.4.5.6;
};
>From my home gateway, i also have a 6to4-tunnel to a 6to4
relay-router, which inturn has tunnel to 6bone.
Can anyone please let me know how to setup such DNS forwarders
for IPv6? Are there any IPv6 root servers? Are there IPv6 DNS
servers in 6bone? How can i configure my IPv6 DNS server to
be cognizant of IPv6 DNS forwarders? How can my IPv6-only hosts
resolve the global IPv6 addresses?
Thanks and best rgds.
Ravi
On Fri, 11 Oct 2002, Bill Owens wrote:
> At 17:35 +0300 10/11/02, Pekka Savola wrote:
> >Note '2002:c058:6301::c058:6301' below.
> >
> >That only works with Microsoft relays because there's a completely bogus
> >and non-interoperable assumption that the relay has address
> >'2002:c058:6301::c058:6301' and it's pingable.
>
> Are you sure? I had it from MS that they ping the IPv4 address of the
> relay, not v6.
>
> Bill.
> _______________________________________________
> 6bone mailing list
> 6bone@mailman.isi.edu
> http://mailman.isi.edu/mailman/listinfo/6bone
>
---------------------------------------------------------
This Mail has been checked for Viruses
Attention: Encrypted mails can NOT be checked!
**
Diese Mail wurde auf Viren geprueft
Hinweis: Verschluesselte mails koennen NICHT auf Viren geprueft werden!
---------------------------------------------------------
From owens@nysernet.org Mon Oct 14 14:19:17 2002
From: owens@nysernet.org (Bill Owens)
Date: Mon, 14 Oct 2002 09:19:17 -0400
Subject: [6bone] Microsoft 6to4 configuration
In-Reply-To: <55D9F49A60EF2A44BF1AC7F348D9385227CD08@mail.v-pe.de>
References: <55D9F49A60EF2A44BF1AC7F348D9385227CD08@mail.v-pe.de>
Message-ID:
At 8:24 +0200 10/14/02, Barker, Ron, vpe wrote:
>Is this reference to the MS relay router hard coded?? Perhaps a
>"persistent" relic of the test enviorment ( which is what I suspect
>) that winds up on the CD.
I'm not an expert on the Microsoft implementation, but I've had some
email conversations with the folks there and they seem very willing
to answer questions about their code. I asked about the relay router
name and they gave me this pointer:
At 15:22 -0700 8/30/02, Christian Huitema wrote:
>The details of our configuration options are explained in a white paper,
>"IPv6 Configurations and Test Lab for Windows XP", available on the
>Microsoft web site:
>
>http://www.microsoft.com/WindowsXP/pro/techinfo/administration/ipv6/ipv6
>configs.doc
>
>It includes instruction on how to use "net shell" to change the DNS name
>that documents the "preferred 6to4 relay":
>
> netsh interface ipv6 6to4 set relay
>
>There is a mailing address for feedback on our IPv6 implementation:
>ipv6-fb@microsoft.com. If you send your queries to that address, they
>can be directly addressed by the development team.
The docs are somewhat difficult to dig out, but after some searching
I found the following bits:
Main Page:
http://www.microsoft.com/windows.netserver/technologies/ipv6/default.mspx
Microsoft Research page
http://research.microsoft.com/msripv6/
IPv6 for W2K
http://msdn.microsoft.com/downloads/sdks/platform/tpipv6.asp
XP IPv6 FAQ:
http://www.microsoft.com/windowsxp/pro/techinfo/administration/ipv6/default.asp
6to4 with the MSR stack:
http://research.microsoft.com/msripv6/docs/6to4.htm
Microsoft IPv6 lab paper:
http://www.microsoft.com/technet/treeview/default.asp?url=/technet/itsolutions/network/maintain/security/ipv6cfg.asp
IPv6/IPv4 coexistence and migration (leads to a .doc file):
http://www.microsoft.com/windows.netserver/technologies/ipv6/ipv6coexist.mspx
I think the best doc is probably the help files on the machine, but I
don't run XP so I haven't had the opportunity to look at them. . .
Hope that helps,
Bill.
From Ron.Barker@v-pe.de Mon Oct 14 14:25:17 2002
From: Ron.Barker@v-pe.de (Barker, Ron, vpe)
Date: Mon, 14 Oct 2002 15:25:17 +0200
Subject: [6bone] .NET RC1 Unicast Address
Message-ID: <55D9F49A60EF2A44BF1AC7F348D9385227CD09@mail.v-pe.de>
This is a multi-part message in MIME format.
------_=_NextPart_001_01C27385.22C0DC71
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Hi,
Does anyone know .NET RC1 configured as an ipv6 router is itself =
configured to advertise the 2001:xxxx:xxxx::/48 prefix and provide =
itself with an global unicast address. I know how to assign the
address manually using netsh.
Thanks
Ron
=20
Dr. Ronald D. Barker
Vodafone Pilotentwicklung GmbH
Chiemgaustr. 116
D-81549 Munich
Germany
Fon +49 (89) 95 410 -0
Fax +49 (89) 95 410 -111
www.v-pe.de
------_=_NextPart_001_01C27385.22C0DC71
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
.NET RC1 Unicast Address
Hi,
Does anyone know .NET RC1 =
configured as an ipv6 router is itself configured to advertise the =
2001:xxxx:xxxx::/48 prefix and provide itself with an global unicast =
address. I know how to assign the address manually using =
netsh.
Thanks
Ron
Dr. Ronald D. Barker
Vodafone Pilotentwicklung GmbH
Chiemgaustr. 116
D-81549 Munich
Germany
Fon +49 (89) 95 410 -0
Fax +49 (89) 95 410 -111
www.v-pe.de
------_=_NextPart_001_01C27385.22C0DC71--
From nicolas.deffayet@ndsoftware.net Tue Oct 15 14:35:37 2002
From: nicolas.deffayet@ndsoftware.net (Nicolas DEFFAYET)
Date: 15 Oct 2002 15:35:37 +0200
Subject: [6bone] RIPE and IPv6 ASN
Message-ID: <1034688937.646.1807.camel@wks1.fr.corp.ndsoftware.com>
Hello,
Disclaimer:
I hide sensitive information (ASN, company name,...).
Flames & co > /dev/null
RIPE are welcome to contact me because i don't find contact except the
registration contact for members only.
My company have an operational ipv6-site since 17 january 2001 on 6bone
(http://whois.6bone.net/cgi-bin/whois?NDSOFTWARE).
We made a lot of tests with our IPv6 experimental network and we provide
IPv6 connectivity to many IPv6 projects. We do BGP peering with a
private ASN (we announce only our routes with community no-export if the
peer accept it and use correct filter) because we don't have our ASN.
Now, we will build a real IPv6 network (native peering, peering with
other ISP on Internet Exchange,...). My company created the first french
IPv6 Internet Exchange, FNIX6 (http://www.fnix6.net) but my company
can't peer on FNIX6 becase we don't have public ASN.
Yesterday, i sent my ASN request to the RIPE by my LIR (peer1):
--------------------------------------------------------------------->
X-NCC-RegID:
#[ANNOUNCED ADDRESS RANGES]#
#[PEERING CONTACTS]#
#[AUT-NUM TEMPLATE]#
aut-num: NEW
as-name: NDSOFTWARE-AS
descr: NDSoftware IP Network
import: from AS
action pref=100;
accept ANY
import: from AS
action pref=100;
accept ANY
export: to AS
announce NEW
export: to AS
announce NEW
remarks: Network problems to: noc@ndsoftwarenet.com
remarks: Peering requests to: peering@ndsoftwarenet.com
remarks: Abuse notifications to: abuse@ndsoftwarenet.com
remarks: NDSoftware have an open peering policy.
admin-c: AUTO-1
tech-c: AUTO-1
notify: notify@ndsoftwarenet.com
mnt-by: NDSOFTWARE-MNT
changed: hostmaster@ripe.net
source: RIPE
#[MAINTAINER TEMPLATE]#
mntner: NDSOFTWARE-MNT
descr: NDSoftware IP Network
admin-c: AUTO-1
tech-c: AUTO-1
upd-to: ipmaster@ndsoftwarenet.com
mnt-nfy: notify@ndsoftwarenet.com
auth: MD5-PW $1$uAUWFve7$aPxYj8kqY4sCqr7g7fL6J/
notify: notify@ndsoftwarenet.com
mnt-by: NDSOFTWARE-MNT
referral-by: RIPE-DBM-MNT
changed: hostmaster@ripe.net
source: RIPE
#[PERSON TEMPLATE]#
person: Nicolas DEFFAYET
address: NDSoftware
address: 57 rue du president Wilson
address: 92300 Levallois-Perret
address: France
phone: +33 671887502
e-mail: nicolas.deffayet@ndsoftware.net
nic-hdl: AUTO-1
notify: notify@ndsoftwarenet.com
mnt-by: NDSOFTWARE-MNT
changed: nicolas.deffayet@ndsoftware.net
source: RIPE
#[ADDITIONAL INFORMATION]#.
We request a pTLA to 6bone as soon we get our ASN.
We are the founder of FNIX6 (French National Internet Exchange IPv6).
We provide IPv6 transit to many projects.
#[TEMPLATE END]#
--------------------------------------------------------------------->
Reply of RIPE:
--------------------------------------------------------------------->
Thank you for your request for an AS number, but with IPv6 you do not
need to
use any AS numbers nor route objects.
Please have a look at the IPv6 documentation at the following link:
http://www.ripe.net/ripe/docs/ipv6.html
--------------------------------------------------------------------->
Very stupid reply, for do BGP peering, request a pTLA to 6bone and peer
on FNIX6, we need a public ASN.
There isn't IPv6 documentation for ASN on
http://www.ripe.net/ripe/docs/ipv6.html
=> How get an ASN for our new IPv6 network ?
(we don't want do IPv4 network)
Thanks
Best Regards,
Nicolas DEFFAYET
From A.Weinberger@ebv.com Tue Oct 15 15:26:39 2002
From: A.Weinberger@ebv.com (Weinberger Andreas)
Date: Tue, 15 Oct 2002 16:26:39 +0200
Subject: [6bone] RIPE and IPv6 ASN
Message-ID: <51420C40A341F14499B8C5A488BEDA58026899@EXSRV02.int.ebv.com>
hi nicolas,
hi 6bone ml and ripe stuff,
> Hello,
[snipp]
> Very stupid reply, for do BGP peering, request a pTLA to
> 6bone and peer
> on FNIX6, we need a public ASN.
>
> There isn't IPv6 documentation for ASN on
> http://www.ripe.net/ripe/docs/ipv6.html
>
> => How get an ASN for our new IPv6 network ?
> (we don't want do IPv4 network)
well,
- is ndsoftware a multihomes isp?
- is ndsoftware a ripe member?
- does ndsoftware has a lir status?
- does ndsoftware have production, 2001:: ipv6 space?
i dont know why you are crying around, but it seems that
ndsoftware isnt any kind of an isp.
and for just playing around with ipv6 (3ffe::) there is
no reason for a public asn. use your current private asn.
its enough.
> Thanks
>
> Best Regards,
>
> Nicolas DEFFAYET
my 2 (euro) cents,
bye,
::::
:: andreas 'randy' weinberger
:: networking group
:: ebv elektronik gmbh&co. kg
:: mail: a.weinberger@ebv.com
:: phone: +49 (0)8121 774-508
From bmanning@ISI.EDU Tue Oct 15 15:33:03 2002
From: bmanning@ISI.EDU (Bill Manning)
Date: Tue, 15 Oct 2002 07:33:03 -0700 (PDT)
Subject: [6bone] RIPE and IPv6 ASN
In-Reply-To: <1034688937.646.1807.camel@wks1.fr.corp.ndsoftware.com> from Nicolas DEFFAYET at "Oct 15, 2 03:35:37 pm"
Message-ID: <200210151433.g9FEX3x23083@boreas.isi.edu>
% My company created the first french
% IPv6 Internet Exchange, FNIX6 (http://www.fnix6.net) but my company
% can't peer on FNIX6 becase we don't have public ASN.
this is not the first v6 enabled exchange in France.
there are three other exchanges which are v6 enabled.
--bill
From bortzmeyer@gitoyen.net Tue Oct 15 15:41:47 2002
From: bortzmeyer@gitoyen.net (Stephane Bortzmeyer)
Date: Tue, 15 Oct 2002 16:41:47 +0200
Subject: [6bone] Re: RIPE and IPv6 ASN
In-Reply-To: <1034688937.646.1807.camel@wks1.fr.corp.ndsoftware.com>
References: <1034688937.646.1807.camel@wks1.fr.corp.ndsoftware.com>
Message-ID: <20021015144147.GA26257@nic.fr>
On Tue, Oct 15, 2002 at 03:35:37PM +0200,
Nicolas DEFFAYET wrote
a message of 126 lines which said:
> RIPE are welcome to contact me because i don't find contact except the
> registration contact for members only.
Until very recently (1st July), RIPE-NCC had an incredibly restrictive
policy regarding IPv6 allocation. So, it is quite possible that many
of the hostmasters are not yet used to IPv6. I suggest to reply to
RIPE-NCC with explanations. You can ask lir-help@ripe.net, too.
> Now, we will build a real IPv6 network (native peering, peering with
> other ISP on Internet Exchange,...). My company created the first french
> IPv6 Internet Exchange, FNIX6
Wrong, the first one was Sfinx , several
months before.
> Thank you for your request for an AS number, but with IPv6 you do not
> need to
> use any AS numbers
:-)
From nicolas.deffayet@ndsoftware.net Tue Oct 15 15:57:36 2002
From: nicolas.deffayet@ndsoftware.net (Nicolas DEFFAYET)
Date: 15 Oct 2002 16:57:36 +0200
Subject: [6bone] RIPE and IPv6 ASN
In-Reply-To: <51420C40A341F14499B8C5A488BEDA58026899@EXSRV02.int.ebv.com>
References: <51420C40A341F14499B8C5A488BEDA58026899@EXSRV02.int.ebv.com>
Message-ID: <1034693856.634.1819.camel@wks1.fr.corp.ndsoftware.com>
On Tue, 2002-10-15 at 16:26, Weinberger Andreas wrote:
Hello,
> [snipp]
>
> > Very stupid reply, for do BGP peering, request a pTLA to
> > 6bone and peer
> > on FNIX6, we need a public ASN.
> >
> > There isn't IPv6 documentation for ASN on
> > http://www.ripe.net/ripe/docs/ipv6.html
> >
> > => How get an ASN for our new IPv6 network ?
> > (we don't want do IPv4 network)
>
> well,
>
> - is ndsoftware a multihomes isp?
yes, we have 2 official IPv6 transit
> - is ndsoftware a ripe member?
no, but it's planned when our ISP activity will be stable
> - does ndsoftware has a lir status?
no, but it's not a problem for request an ASN
> - does ndsoftware have production, 2001:: ipv6 space?
no, because we aren't LIR, we can't request a sTLA
> i dont know why you are crying around, but it seems that
> ndsoftware isnt any kind of an isp.
we start an activity of ISP.
> and for just playing around with ipv6 (3ffe::) there is
> no reason for a public asn. use your current private asn.
> its enough.
You can't peer with a private ASN on an Internet Exchange.
Best Regards,
Nicolas DEFFAYET
From nicolas.deffayet@ndsoftware.net Tue Oct 15 16:00:54 2002
From: nicolas.deffayet@ndsoftware.net (Nicolas DEFFAYET)
Date: 15 Oct 2002 17:00:54 +0200
Subject: [6bone] RIPE and IPv6 ASN
In-Reply-To: <200210151433.g9FEX3x23083@boreas.isi.edu>
References: <200210151433.g9FEX3x23083@boreas.isi.edu>
Message-ID: <1034694054.648.1824.camel@wks1.fr.corp.ndsoftware.com>
On Tue, 2002-10-15 at 16:33, Bill Manning wrote:
> % My company created the first french
> % IPv6 Internet Exchange, FNIX6 (http://www.fnix6.net) but my company
> % can't peer on FNIX6 becase we don't have public ASN.
>
> this is not the first v6 enabled exchange in France.
> there are three other exchanges which are v6 enabled.
>
This other Internet Exchange in France are IPv4 and IPv6 Internet
Exchange (dualstack).
IPv6 Internet Exchange = IPv6 only
FNIX6 is the first french IPv6 Internet Exchange and the second IPv6
Internet Exchange in Europe (the first is UK6x).
For information, a list of IPv6 Internet Exchange:
- 6IIX (New-York, Los Angeles, Santa Clara, US)
- 6NGIX (Seoul, KR)
- 6TAP (Chicago, US)
- KNIX6 (Soul, KR)
- NSPIXP-6 (Tokyo, JP)
- NY6IX (New-York, US)
- S-IX (San Jose,US)
- UK6X (London, UK)
Best Regards,
Nicolas DEFFAYET
From pekkas@netcore.fi Tue Oct 15 16:02:51 2002
From: pekkas@netcore.fi (Pekka Savola)
Date: Tue, 15 Oct 2002 18:02:51 +0300 (EEST)
Subject: [6bone] Re: RIPE and IPv6 ASN
In-Reply-To:
Message-ID:
On Tue, 15 Oct 2002, Arien Vijn wrote:
> > Thank you for your request for an AS number, but with IPv6 you do not
> > need to
> > use any AS numbers nor route objects.
> >
> > Please have a look at the IPv6 documentation at the following link:
> > http://www.ripe.net/ripe/docs/ipv6.html
> >
> > --------------------------------------------------------------------->
> >
>
>
> From a technical point of view RIPE's reply is not accurate indeed. How can
> you talk BGP without an AS number? That has nothing to do with IPv6 as such.
Incorrect. There is a private ASN space, see:
http://www.iana.org/assignments/as-numbers
--
Pekka Savola "Tell me of difficulties surmounted,
Netcore Oy not those you stumble over and fall"
Systems. Networks. Security. -- Robert Jordan: A Crown of Swords
From arien+6bone@ams-ix.net Tue Oct 15 16:17:07 2002
From: arien+6bone@ams-ix.net (Arien Vijn)
Date: Tue, 15 Oct 2002 17:17:07 +0200
Subject: [6bone] RIPE and IPv6 ASN
In-Reply-To: <1034694054.648.1824.camel@wks1.fr.corp.ndsoftware.com>
Message-ID:
On 15-10-2002 17:00PM, "Nicolas DEFFAYET"
wrote:
> On Tue, 2002-10-15 at 16:33, Bill Manning wrote:
>> % My company created the first french
>> % IPv6 Internet Exchange, FNIX6 (http://www.fnix6.net) but my company
>> % can't peer on FNIX6 becase we don't have public ASN.
>>
>> this is not the first v6 enabled exchange in France.
>> there are three other exchanges which are v6 enabled.
>>
>
> This other Internet Exchange in France are IPv4 and IPv6 Internet
> Exchange (dualstack).
> IPv6 Internet Exchange = IPv6 only
Why? Is what we are doing not IPv6?
Arien
From A.Weinberger@ebv.com Tue Oct 15 16:29:03 2002
From: A.Weinberger@ebv.com (Weinberger Andreas)
Date: Tue, 15 Oct 2002 17:29:03 +0200
Subject: [6bone] RIPE and IPv6 ASN
Message-ID: <51420C40A341F14499B8C5A488BEDA5802689B@EXSRV02.int.ebv.com>
hi again,
...
> > well,
> >
> > - is ndsoftware a multihomes isp?
> yes, we have 2 official IPv6 transit
2 ipv6 transits? if you dont have any valid asn, so how
do you use bgp4+ for ip transit?
...
> > and for just playing around with ipv6 (3ffe::) there is
> > no reason for a public asn. use your current private asn.
> > its enough.
>
> You can't peer with a private ASN on an Internet Exchange.
sure you can.
peering != transit routing.
> Best Regards,
>
> Nicolas DEFFAYET
ps: completel, with my help, opened as first european commercial
company (in words TWO) internet exchanges in germany with ipv6
native peering _AND_ transitrouting (BCIX and N-IX)... - scnr ;)
bye,
::::
:: andreas 'randy' weinberger
:: networking group
:: ebv elektronik gmbh&co. kg
:: mail: a.weinberger@ebv.com
:: phone: +49 (0)8121 774-508
From arien.vijn@ams-ix.net Tue Oct 15 15:36:14 2002
From: arien.vijn@ams-ix.net (Arien Vijn)
Date: Tue, 15 Oct 2002 16:36:14 +0200
Subject: [6bone] Re: RIPE and IPv6 ASN
In-Reply-To: <1034688937.646.1807.camel@wks1.fr.corp.ndsoftware.com>
Message-ID:
On 15-10-2002 15:35PM, "Nicolas DEFFAYET"
wrote:
[...]
>
> --------------------------------------------------------------------->
>
> Reply of RIPE:
>
> --------------------------------------------------------------------->
>
> Thank you for your request for an AS number, but with IPv6 you do not
> need to
> use any AS numbers nor route objects.
>
> Please have a look at the IPv6 documentation at the following link:
> http://www.ripe.net/ripe/docs/ipv6.html
>
> --------------------------------------------------------------------->
>
>From a technical point of view RIPE's reply is not accurate indeed. How can
you talk BGP without an AS number? That has nothing to do with IPv6 as such.
Hmmm... if there is absolutely something wrong if this is truly in the
policies.
Arien
--
Arien Vijn tel: +31 205 141 718
Amsterdam Internet Exchange mobile: +31 651 836 444
http://www.ams-ix.net e-mail: arien.vijn@ams-ix.net
From leo@ripe.net Tue Oct 15 16:25:38 2002
From: leo@ripe.net (leo vegoda)
Date: Tue, 15 Oct 2002 17:25:38 +0200
Subject: [6bone] Re: RIPE and IPv6 ASN
In-Reply-To: <1034688937.646.1807.camel@wks1.fr.corp.ndsoftware.com>
References: <1034688937.646.1807.camel@wks1.fr.corp.ndsoftware.com>
Message-ID: <20021015152538.GA26892@ripe.net>
Dear Nicolas,
I apologise for the problem you have experienced obtaining an ASN
from the RIPE NCC. The answer you received was obviously a mistake.
We will correct this by the end of this business day.
You mentioned that you found it difficult to find a list of
alternative contact information. There is a list available on our
web site that can be found at:
We will make the link from the front page of the site more obvious.
Best regards,
--
leo vegoda
RIPE NCC
Registration Services
From gert@space.net Tue Oct 15 17:04:28 2002
From: gert@space.net (Gert Doering)
Date: Tue, 15 Oct 2002 18:04:28 +0200
Subject: [6bone] RIPE and IPv6 ASN
In-Reply-To: <51420C40A341F14499B8C5A488BEDA5802689B@EXSRV02.int.ebv.com>; from A.Weinberger@ebv.com on Tue, Oct 15, 2002 at 05:29:03PM +0200
References: <51420C40A341F14499B8C5A488BEDA5802689B@EXSRV02.int.ebv.com>
Message-ID: <20021015180428.H94537@Space.Net>
Hi,
On Tue, Oct 15, 2002 at 05:29:03PM +0200, Weinberger Andreas wrote:
> ps: completel, with my help, opened as first european commercial
> company (in words TWO) internet exchanges in germany with ipv6
> native peering _AND_ transitrouting (BCIX and N-IX)... - scnr ;)
OK, can we please stop this "my IXP is much more IPv6 than yours"
thread?
The two major IXPs in .DE (DECIX, INXS) have native IPv6 as well, and both
since well over a year. Of course both have IPv4, too, as an IPv6-only
exchange sounds more like a reason to waste tax money than anything useful.
Gert Doering
-- NetMaster
--
Total number of prefixes smaller than registry allocations: 48282 (47686)
SpaceNet AG Mail: netmaster@Space.Net
Joseph-Dollinger-Bogen 14 Tel : +49-89-32356-0
80807 Muenchen Fax : +49-89-32356-299
From nicolas.deffayet@ndsoftware.net Tue Oct 15 17:17:56 2002
From: nicolas.deffayet@ndsoftware.net (Nicolas DEFFAYET)
Date: 15 Oct 2002 18:17:56 +0200
Subject: [6bone] RIPE and IPv6 ASN
In-Reply-To: <51420C40A341F14499B8C5A488BEDA5802689B@EXSRV02.int.ebv.com>
References: <51420C40A341F14499B8C5A488BEDA5802689B@EXSRV02.int.ebv.com>
Message-ID: <1034698677.641.1835.camel@wks1.fr.corp.ndsoftware.com>
On Tue, 2002-10-15 at 17:29, Weinberger Andreas wrote:
hi again,
> ...
> > > well,
> > >
> > > - is ndsoftware a multihomes isp?
> > yes, we have 2 official IPv6 transit
>
> 2 ipv6 transits? if you dont have any valid asn, so how
> do you use bgp4+ for ip transit?
I used before a private ASN.
> > > and for just playing around with ipv6 (3ffe::) there is
> > > no reason for a public asn. use your current private asn.
> > > its enough.
> >
> > You can't peer with a private ASN on an Internet Exchange.
>
> sure you can.
No, a lot of ISP don't want peer with private ASN.
> peering != transit routing.
FNIX6 allow transit.
Best Regards,
Nicolas DEFFAYET
From czmok@gatel.net Tue Oct 15 17:22:38 2002
From: czmok@gatel.net (Jan Czmok)
Date: Tue, 15 Oct 2002 18:22:38 +0200
Subject: [6bone] RIPE and IPv6 ASN
In-Reply-To: <51420C40A341F14499B8C5A488BEDA5802689B@EXSRV02.int.ebv.com>
References: <51420C40A341F14499B8C5A488BEDA5802689B@EXSRV02.int.ebv.com>
Message-ID: <20021015162238.GB5797@gollum.gatel.net>
Weinberger Andreas (A.Weinberger@ebv.com) wrote:
> hi again,
>
> ...
> > > well,
> > >
> > > - is ndsoftware a multihomes isp?
> > yes, we have 2 official IPv6 transit
>
> 2 ipv6 transits? if you dont have any valid asn, so how
> do you use bgp4+ for ip transit?
>
> ...
>
> > > and for just playing around with ipv6 (3ffe::) there is
> > > no reason for a public asn. use your current private asn.
> > > its enough.
> >
> > You can't peer with a private ASN on an Internet Exchange.
>
> sure you can.
>
> peering != transit routing.
>
> > Best Regards,
> >
> > Nicolas DEFFAYET
>
>
> ps: completel, with my help, opened as first european commercial
> company (in words TWO) internet exchanges in germany with ipv6
> native peering _AND_ transitrouting (BCIX and N-IX)... - scnr ;)
>
and while we're at this ...
gigabell (former employer) was founding member of ipv6-forum
and also built out the first exchange in berlin with ipv6 quite long
ago (1999 afaik) [SCNR]
--jan
--
Jan Ahrent Czmok - Senior Network Engineer - Access Networks
Global Access Telecommunications, Inc. - Stephanstr. 3 - 60313 Frankfurt
voice: +49 69 299896-35 - fax: +49 69 299896-66 - email: czmok@gatel.de
From tjc@ecs.soton.ac.uk Tue Oct 15 22:12:01 2002
From: tjc@ecs.soton.ac.uk (Tim Chown)
Date: Tue, 15 Oct 2002 22:12:01 +0100
Subject: [6bone] .NET RC1 IPv6 6to4
In-Reply-To:
References:
Message-ID: <20021015211201.GQ20360@login.ecs.soton.ac.uk>
On Fri, Oct 11, 2002 at 08:16:58AM -0700, Matthew Lehman wrote:
>
> My machines at home are setup to go to the MS relay because I know it's
> topologically close and the closest advertised relay using the 6to4
> anycast address is at SWITCH which is a long way from home.
So if you have a relay, advertise it like the Swiss one, and let nature
take its course?
What sort of activity level do you see on your honey pot relay? :)
Tim
From MMESTDAG@ncsbe.jnj.com Wed Oct 16 13:35:27 2002
From: MMESTDAG@ncsbe.jnj.com (MMESTDAG@ncsbe.jnj.com)
Date: Wed, 16 Oct 2002 14:35:27 +0200
Subject: [6bone] Requesting IPv6 addresses
Message-ID:
This message is in MIME format. Since your mail reader does not understand
this format, some or all of this message may not be legible.
------_=_NextPart_001_01C27510.8169B430
Content-Type: text/plain;
charset="iso-8859-1"
Hi
I want to do some tests with IPv6, but in the manuals it says to contact an
ISP. I tried to contact 2 Belgian ISP's but they don't reply. I want do get
some IPv6 addresses and than tunnel with 6to4.
Where do I get started now?
Regs
Mark
------_=_NextPart_001_01C27510.8169B430
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Requesting IPv6 addresses
Hi
I want to do some tests with IPv6, but in the manuals =
it says to contact an ISP. I tried to contact 2 Belgian ISP's but they =
don't reply. I want do get some IPv6 addresses and than tunnel with =
6to4.
Where do I get started now?
Regs
Mark
------_=_NextPart_001_01C27510.8169B430--
From fink@es.net Wed Oct 16 14:25:47 2002
From: fink@es.net (Bob Fink)
Date: Wed, 16 Oct 2002 06:25:47 -0700
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
Message-ID: <5.1.0.14.0.20021016061927.02891008@imap2.es.net>
6bone Folk,
NDSOFTWARE has requested a pTLA allocation and I find their request fully
compliant with RFC2772. The open review period for this will close 23
October 2002. Please send your comments to me or the list.
Thanks,
Bob
=====
>Hello,
>
>On behalf of NDSoftware, I would like to submit our application for a
>pTLA.
>
>Best Regards,
>
>Nicolas DEFFAYET
>
>------------------------------------------------------------------------
>
> From RFC 2772
>
>
>7. Guidelines for 6Bone pTLA sites
>
>
> The following rules apply to qualify for a 6Bone pTLA allocation. It
> should be recognized that holders of 6Bone pTLA allocations are
> expected to provide production quality backbone network services for
> the 6Bone.
>
>
> 1. The pTLA Applicant must have a minimum of three (3) months
> qualifying experience as a 6Bone end-site or pNLA transit.
>During
> the entire qualifying period the Applicant must be operationally
> providing the following:
>
>Our ipv6-site is operational since 17 january 2001 on 6bone.
>
> a. Fully maintained, up to date, 6Bone Registry entries for their
> ipv6-site inet6num, mntner, and person objects, including each
> tunnel that the Applicant has.
>
>http://whois.6bone.net/cgi-bin/whois?NDSOFTWARE
>
>
> b. Fully maintained, and reliable, BGP4+ peering and connectivity
> between the Applicant's boundary router and the appropriate
> connection point into the 6Bone. This router must be IPv6
> pingable. This criteria is judged by members of the 6Bone
> Operations Group at the time of the Applicant's pTLA request.
>
>We have currently 101 BGP4+ sessions.
>
>Our ASN is AS25358:
>aut-num: AS25358
>as-name: NDSOFTWARE-AS
>descr: NDSoftware IP Network
>
>We use 2 routers:
> - parcr1.fr.ndsoftwarenet.net
> - parcr2.fr.ndsoftwarenet.net
>Looking Glass: http://noc.ndsoftwarenet.com/lg/
>
>
> c. Fully maintained DNS forward (AAAA) and reverse (ip6.int)
> entries for the Applicant's router(s) and at least one host
> system.
>
>We have 3 nameservers:
> - ns1.ndsoftwarenet.com
> - ns2.ndsoftwarenet.com
> - ns3.ndsoftwarenet.com
>
> d. A fully maintained, and reliable, IPv6-accessible system
> providing, at a mimimum, one or more web pages, describing the
> Applicant's IPv6 services. This server must be IPv6 pingable.
>
>http://noc.ndsoftwarenet.com/
>
> 2. The pTLA Applicant MUST have the ability and intent to provide
> "production-quality" 6Bone backbone service. Applicants must
> provide a statement and information in support of this claim.
> This MUST include the following:
>
>
> a. A support staff of two persons minimum, three preferable, with
> person attributes registered for each in the ipv6-site object
> for the pTLA applicant.
>
>NDN1-6BONE
>CB2-6BONE
>BN3-6BONE
>MM14-6BONE
>MC7-6BONE
>
> b. A common mailbox for support contact purposes that all support
> staff have acess to, pointed to with a notify attribute in the
> ipv6-site object for the pTLA Applicant.
>
>ipmaster@ndsoftwarenet.com
>
> 3. The pTLA Applicant MUST have a potential "user community" that
> would be served by its becoming a pTLA, e.g., the Applicant is a
> major provider of Internet service in a region, country, or focus
> of interest. Applicant must provide a statement and information
> in support this claim.
>
>NDSoftware operates an IPv6 network and provide a lot of IPv6 services
>to many projects.
>
>We provide to:
>
>IPv6-FR (a non profit organisation for the developement of IPv6 in France
> 200 users, each user have a /48.
>
>NexGenCollective (http://www.nexgencollective.net/)
> 150 users, each user have a /48.
>
>ATI (A tunisian ISP, http://www.ipv6net.tn/)
>
>and a lot of others (see our whois), this services: IPv6 connectivity
>(STATIC or BGP with a IPv6 block), IPv6 newsfeeds/newsread,...
>
>We do many actions in IPv6 research, we created FNIX6 (French
>International Internet Exchange IPv6, http://www.fnix6.net/), we host many
>mirrors
>available in IPv6, we created ftp://ftp.openipv6.com/ (a FTP with a lot
>of IPv6 stuff).
>
> 4. The pTLA Applicant MUST commit to abide by the current 6Bone
> operational rules and policies as they exist at time of its
> application, and agree to abide by future 6Bone backbone
> operational rules and policies as they evolve by consensus of the
> 6Bone backbone and user community.
>
>
>We agree to all current and future rules and policies.
>
>----
-end
From gert@space.net Wed Oct 16 14:31:46 2002
From: gert@space.net (Gert Doering)
Date: Wed, 16 Oct 2002 15:31:46 +0200
Subject: [6bone] Requesting IPv6 addresses
In-Reply-To: ; from MMESTDAG@ncsbe.jnj.com on Wed, Oct 16, 2002 at 02:35:27PM +0200
References:
Message-ID: <20021016153146.W94537@Space.Net>
Hi,
On Wed, Oct 16, 2002 at 02:35:27PM +0200, MMESTDAG@ncsbe.jnj.com wrote:
> I want to do some tests with IPv6, but in the manuals it says to contact an
> ISP. I tried to contact 2 Belgian ISP's but they don't reply. I want do get
> some IPv6 addresses and than tunnel with 6to4.
> Where do I get started now?
For 6to4, the IPv6 addresses are constructed from your IPv4 address,
so there's no need to go to an ISP.
Gert Doering
-- NetMaster
--
Total number of prefixes smaller than registry allocations: 48282 (47686)
SpaceNet AG Mail: netmaster@Space.Net
Joseph-Dollinger-Bogen 14 Tel : +49-89-32356-0
80807 Muenchen Fax : +49-89-32356-299
From tjc@ecs.soton.ac.uk Wed Oct 16 14:56:47 2002
From: tjc@ecs.soton.ac.uk (Tim Chown)
Date: Wed, 16 Oct 2002 14:56:47 +0100
Subject: [6bone] Requesting IPv6 addresses
In-Reply-To:
References:
Message-ID: <20021016135647.GU5132@starling.ecs.soton.ac.uk>
On Wed, Oct 16, 2002 at 02:35:27PM +0200, MMESTDAG@ncsbe.jnj.com wrote:
> Hi
>
> I want to do some tests with IPv6, but in the manuals it says to contact an
> ISP. I tried to contact 2 Belgian ISP's but they don't reply. I want do get
> some IPv6 addresses and than tunnel with 6to4.
> Where do I get started now?
Try the NGN LAB at ULB in Brussels? (Paul van Binst's group)
Tim
From pekkas@netcore.fi Wed Oct 16 16:00:07 2002
From: pekkas@netcore.fi (Pekka Savola)
Date: Wed, 16 Oct 2002 18:00:07 +0300 (EEST)
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <5.1.0.14.0.20021016061927.02891008@imap2.es.net>
Message-ID:
I support this -- there seems to be a real need for the pTLA.
As an unofficial constructive criticism...
On Wed, 16 Oct 2002, Bob Fink wrote:
> > between the Applicant's boundary router and the appropriate
> > connection point into the 6Bone. This router must be IPv6
> > pingable. This criteria is judged by members of the 6Bone
> > Operations Group at the time of the Applicant's pTLA request.
> >
> >We have currently 101 BGP4+ sessions.
.. I'd appreciate it if you decreased the amount of BGP sessions by about
80-90%.
(Hopefully you don't intend to provide transit between those...)
Thanks!
--
Pekka Savola "Tell me of difficulties surmounted,
Netcore Oy not those you stumble over and fall"
Systems. Networks. Security. -- Robert Jordan: A Crown of Swords
From lilesl@corp.earthlink.net Wed Oct 16 16:23:15 2002
From: lilesl@corp.earthlink.net (Lorin Liles)
Date: Wed, 16 Oct 2002 08:23:15 -0700
Subject: [6bone] Unsubscribe
Message-ID: <00bd01c27527$f2c92cc0$7d2811ac@SAC022214>
This is a multi-part message in MIME format.
------=_NextPart_000_00BA_01C274ED.463C66F0
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Please take me off this distro.
THX,
L
------=_NextPart_000_00BA_01C274ED.463C66F0
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Please take me off this =
distro.
THX,
L
------=_NextPart_000_00BA_01C274ED.463C66F0--
From kristoff.bonne@skypro.be Wed Oct 16 20:03:28 2002
From: kristoff.bonne@skypro.be (Kristoff Bonne)
Date: Wed, 16 Oct 2002 21:03:28 +0200 (CEST)
Subject: [6bone] Requesting IPv6 addresses
In-Reply-To:
Message-ID:
Greetings,
On Wed, 16 Oct 2002 MMESTDAG@ncsbe.jnj.com wrote:
> I want to do some tests with IPv6, but in the manuals it says to contact an
> ISP. I tried to contact 2 Belgian ISP's but they don't reply. I want do get
> some IPv6 addresses and than tunnel with 6to4.
> Where do I get started now?
Have you tried belnet? I got my 6bone-addresses from them.
But, for 6to4 you just need to set up a 6to4 gateway and that should be
it. You don't need an ISP for this.
> Mark
Cheerio! Kr. Bonne.
--
KB905-RIPE Belgacom IP networking
(c=be,a=rtt,p=belgacomgroup,s=Bonne,g=Kristoff) Internet, IP and IP/VPN
kristoff.bonne@skypro.be Faxbox : +32 2 2435122
From ipsopi@yahoo.com Wed Oct 16 20:10:23 2002
From: ipsopi@yahoo.com (Shashikanth Sopirala)
Date: Wed, 16 Oct 2002 12:10:23 -0700 (PDT)
Subject: [6bone] How can I connect to a 6 BONE Tunnel
Message-ID: <20021016191023.1080.qmail@web21305.mail.yahoo.com>
Hello there,
I am in a University network, I want to connect a host
to the 6BONE tunnel. I want to make a host as a IPv6
host. How can I do this?
Please help.
Shashi
__________________________________________________
Do you Yahoo!?
Faith Hill - Exclusive Performances, Videos & More
http://faith.yahoo.com
From rsamprat@cisco.com Wed Oct 16 20:41:14 2002
From: rsamprat@cisco.com (Ravi Samprathi)
Date: Wed, 16 Oct 2002 12:41:14 -0700 (PDT)
Subject: [6bone] How can I connect to a 6 BONE Tunnel
In-Reply-To: <20021016191023.1080.qmail@web21305.mail.yahoo.com>
Message-ID:
Hi
You may want to go through www.6bone.net/6bone_hookup.html.
Contact your nearest 6bone ISP, you will be assigned a
6bone prefix, configure your tunnel to 6bone using
their point-to-point v6 address, and it should be straight
forward from there.
The other way is to hookup to your nearest 6to4 relay router,
and configure a 6to4 tunnel to them, from your v6 host.
To make a host IPv6-only, besides compiling IPv6 into the kernel,
v4 stack may still not be completely removable, you should
be fine by assigning 0.0.0.0 ipv4 addresses.
-Ravi
>>>>>>>> Go the IPv6 way >>>>>>>>
On Wed, 16 Oct 2002, Shashikanth Sopirala wrote:
> Hello there,
> I am in a University network, I want to connect a host
> to the 6BONE tunnel. I want to make a host as a IPv6
> host. How can I do this?
> Please help.
> Shashi
>
> __________________________________________________
> Do you Yahoo!?
> Faith Hill - Exclusive Performances, Videos & More
> http://faith.yahoo.com
> _______________________________________________
> 6bone mailing list
> 6bone@mailman.isi.edu
> http://mailman.isi.edu/mailman/listinfo/6bone
>
From michiel@e-concepts.be Wed Oct 16 22:45:30 2002
From: michiel@e-concepts.be (Michiel Van Opstal)
Date: Wed, 16 Oct 2002 23:45:30 +0200
Subject: [6bone] Requesting IPv6 addresses
In-Reply-To:
References:
Message-ID: <20021016214530.GA23318@e-concepts.be>
On Wed, Oct 16, 2002 at 02:35:27PM +0200, MMESTDAG@ncsbe.jnj.com wrote:
> Hi
>
> I want to do some tests with IPv6, but in the manuals it says to contact an
> ISP. I tried to contact 2 Belgian ISP's but they don't reply. I want do get
> some IPv6 addresses and than tunnel with 6to4.
> Where do I get started now?
Chello/UPCBelgium is providing native IPv6 in belgium.
And you can request a tunnel at wanadoo also (http://www.ipv6.wanadoo.be)
>
> Regs
> Mark
From Ronald.vanderPol@rvdp.org Thu Oct 17 12:24:11 2002
From: Ronald.vanderPol@rvdp.org (Ronald van der Pol)
Date: Thu, 17 Oct 2002 13:24:11 +0200
Subject: [6bone] Requesting IPv6 addresses
In-Reply-To: <20021016214530.GA23318@e-concepts.be>
References: <20021016214530.GA23318@e-concepts.be>
Message-ID: <20021017112411.GD4430@rvdp.org>
On Wed, Oct 16, 2002 at 23:45:30 +0200, Michiel Van Opstal wrote:
> Chello/UPCBelgium is providing native IPv6 in belgium.
Interesting, can you elaborate or give some pointers?
rvdp
From michiel@e-concepts.be Thu Oct 17 13:03:40 2002
From: michiel@e-concepts.be (Michiel Van Opstal)
Date: Thu, 17 Oct 2002 14:03:40 +0200
Subject: [6bone] Requesting IPv6 addresses
In-Reply-To: <20021017112411.GD4430@rvdp.org>
References: <20021016214530.GA23318@e-concepts.be> <20021017112411.GD4430@rvdp.org>
Message-ID: <20021017120339.GB23318@e-concepts.be>
On Thu, Oct 17, 2002 at 01:24:11PM +0200, Ronald van der Pol wrote:
> On Wed, Oct 16, 2002 at 23:45:30 +0200, Michiel Van Opstal wrote:
>
> > Chello/UPCBelgium is providing native IPv6 in belgium.
>
> Interesting, can you elaborate or give some pointers?
They are running a router advertisement deamon so you can simply obtain an ipv6 address
and to request a /64 subnet mail ipv6@chello.com
>
> rvdp
From rjorgensen@upctechnology.com Thu Oct 17 13:32:03 2002
From: rjorgensen@upctechnology.com (Roger Jorgensen)
Date: Thu, 17 Oct 2002 14:32:03 +0200
Subject: [6bone] Requesting IPv6 addresses
In-Reply-To: <20021017120339.GB23318@e-concepts.be>
References: <20021017112411.GD4430@rvdp.org>
<20021016214530.GA23318@e-concepts.be>
<20021017112411.GD4430@rvdp.org>
Message-ID: <5.1.0.14.0.20021017143046.00b1dbf0@213.46.233.213>
At 02:03 PM 10/17/2002 +0200, Michiel Van Opstal wrote:
>On Thu, Oct 17, 2002 at 01:24:11PM +0200, Ronald van der Pol wrote:
> > On Wed, Oct 16, 2002 at 23:45:30 +0200, Michiel Van Opstal wrote:
> >
> > > Chello/UPCBelgium is providing native IPv6 in belgium.
> >
> > Interesting, can you elaborate or give some pointers?
>
>They are running a router advertisement deamon so you can simply obtain an
>ipv6 address
>and to request a /64 subnet mail ipv6@chello.com
more correct would be ipv6@aorta.net :)
and yes, UPC.BE are native IPv6, been so for several years (object TVD
@whois.6bone.net), not yet well known.
http://www.ipv6.chello.com for more generic info about our IPv6 work
here at UPC/chello.
---
Roger Jorgensen (rjorgensen@upctechnology.com)
System Engineer @ UPC Technology / IP engineering
handles: ROJO1-6BONE ROJO9-RIPE RJC10-NORID
From buddha73@newyork.com Thu Oct 17 14:13:39 2002
From: buddha73@newyork.com (Kryno Bosman)
Date: 17 Oct 2002 14:13:39 +0100
Subject: [6bone] Unsubscribe
Message-ID: <2a5d801c275df$0275b320$4701020a@corp.load.com>
This is a multi-part message in MIME format.
------=_NextPart_000_2A5D5_01C275E7.64356030
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: base64
TWUgdG9vIQ0KIA0KVEhOWCENCg0KCS0tLS0tT3JpZ2luYWwgTWVzc2FnZS0tLS0tDQoJDQoJRnJv
bTogTG9yaW4gTGlsZXM8bGlsZXNsQGNvcnAuZWFydGhsaW5rLm5ldD4NCglUbzogNmJvbmVAbWFp
bG1hbi5pc2kuZWR1IDw2Ym9uZUBtYWlsbWFuLmlzaS5lZHU+DQoJU3ViamVjdDogWzZib25lXSBV
bnN1YnNjcmliZQ0KCQ0KCSANCgkgDQoJIA0KCVBsZWFzZSB0YWtlIG1lIG9mZiB0aGlzIGRpc3Ry
by4NCgkgDQoJVEhYLA0KCUwNCg0KDQo=
------=_NextPart_000_2A5D5_01C275E7.64356030
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: base64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------=_NextPart_000_2A5D5_01C275E7.64356030--
From ferryas@cc.saga-u.ac.jp Fri Oct 18 02:39:59 2002
From: ferryas@cc.saga-u.ac.jp (ferry astika saputra)
Date: Fri, 18 Oct 2002 10:39:59 +0900
Subject: [6bone] Building FreeBSD ipv6 router problem
Message-ID: <200210181039.ADD47128.ZGSBTFS@cc.saga-u.ac.jp>
I want to make some eksperiment with ipv6. I have already get my
ipv6 address , here my ipv6 address allocation from my uplink:
3ffe:517:30a:5800:: /64. And my topology :
-- ---------
---------------| Router$B!!!!!!(B|--------------|
fxp1 ------------fxp0
Uplink Router: FreeBSD Client: WindowsXP(ipv6 ready)
4.6
Firt step I try to connet my uplink, and its succed. I can ping6
some ipv6 site like www.kame.net, www.jp.freebsd.org.
Then I modify my rc.conf :
ipv6_enable="YES"
ipv6_network_interfaces="fxp0 fxp1"
ipv6_defaultrouter="3ffe:517:30a:5800::1%fxp1"
ipv6_gateway_enable="YES"
ipv6_router_enable="YES"
ipv6_ifconfig_fxp0="3ffe:517:30a:5880::1 prefixlen 64 up"
rtadvd_enable="YES"
rtadvd_interfaces="fxp0"
With this configuration my router can advertise to downlink, but
cannot receive router advertisement from uplink.
Somebody can help me ?
--
ferry astika saputra
ferryas@cc.saga-u.ac.jp
From itojun@iijlab.net Fri Oct 18 03:28:29 2002
From: itojun@iijlab.net (itojun@iijlab.net)
Date: Fri, 18 Oct 2002 11:28:29 +0900
Subject: [6bone] Building FreeBSD ipv6 router problem
In-Reply-To: ferryas's message of Fri, 18 Oct 2002 10:39:59 +0900. <200210181039.ADD47128.ZGSBTFS@cc.saga-u.ac.jp>
Message-ID: <20021018022829.C30EF4B23@coconut.itojun.org>
>I want to make some eksperiment with ipv6. I have already get my
> ipv6 address , here my ipv6 address allocation from my uplink:
>3ffe:517:30a:5800:: /64. And my topology :
> -- ---------
>---------------| Router$B!!!!!!(B|--------------|
> fxp1 ------------fxp0
>Uplink Router: FreeBSD Client: WindowsXP(ipv6 ready)
>
> 4.6
>
>
>Firt step I try to connet my uplink, and its succed. I can ping6
> some ipv6 site like www.kame.net, www.jp.freebsd.org.
>Then I modify my rc.conf :
>
>ipv6_enable="YES"
>ipv6_network_interfaces="fxp0 fxp1"
>ipv6_defaultrouter="3ffe:517:30a:5800::1%fxp1"
>ipv6_gateway_enable="YES"
>ipv6_router_enable="YES"
>ipv6_ifconfig_fxp0="3ffe:517:30a:5880::1 prefixlen 64 up"
>rtadvd_enable="YES"
>rtadvd_interfaces="fxp0"
>
>With this configuration my router can advertise to downlink, but
> cannot receive router advertisement from uplink.
>Somebody can help me ?
i see three problems:
(1) global address cannot have scope identifier, so ipv6_defaultrouter
setting is wrong.
(2) ipv6_defaultrouter setting specifies offlink address (not on the
same subnet). it has to be on the same link as the node itself
(3) routes should have link-local address as gateway, to make icmp6
redirect work. you'd better run some routing protocol than to
specify ipv6_defaultrouter.
itojun
From pekkas@netcore.fi Sat Oct 19 12:36:32 2002
From: pekkas@netcore.fi (Pekka Savola)
Date: Sat, 19 Oct 2002 14:36:32 +0300 (EEST)
Subject: [6bone] A looking glass/traceroute page
Message-ID:
Hello,
Failing to find good pages on collected looking glass/traceroute pointers
for v6, I did some quick browsing and collected some at:
http://staff.csc.fi/psavola/lg.html
Please send updates/modifications/etc directly. In particular, I failed
to find the pages (if they exist) for the major v6 services providers..
HTH.
--
Pekka Savola "Tell me of difficulties surmounted,
Netcore Oy not those you stumble over and fall"
Systems. Networks. Security. -- Robert Jordan: A Crown of Swords
From pekkas@netcore.fi Sat Oct 19 12:42:36 2002
From: pekkas@netcore.fi (Pekka Savola)
Date: Sat, 19 Oct 2002 14:42:36 +0300 (EEST)
Subject: [6bone] Re: A looking glass/traceroute page
In-Reply-To:
Message-ID:
On Sat, 19 Oct 2002, Pekka Savola wrote:
> Failing to find good pages on collected looking glass/traceroute pointers
> for v6, I did some quick browsing and collected some at:
>
> http://staff.csc.fi/psavola/lg.html
>
> Please send updates/modifications/etc directly. In particular, I failed
> to find the pages (if they exist) for the major v6 services providers..
To clarify, because the mailing list removed the reply-to I set, do not
send any updates etc. on the list(s) but directly.
Thanks.
--
Pekka Savola "Tell me of difficulties surmounted,
Netcore Oy not those you stumble over and fall"
Systems. Networks. Security. -- Robert Jordan: A Crown of Swords
From nicolas.deffayet@ndsoftware.net Sat Oct 19 15:13:23 2002
From: nicolas.deffayet@ndsoftware.net (Nicolas DEFFAYET)
Date: 19 Oct 2002 16:13:23 +0200
Subject: [6bone] A looking glass/traceroute page
In-Reply-To:
References:
Message-ID: <1035036803.610.1726.camel@wks1.fr.corp.ndsoftware.com>
On Sat, 2002-10-19 at 13:36, Pekka Savola wrote:
Hello,
> Failing to find good pages on collected looking glass/traceroute pointers
> for v6, I did some quick browsing and collected some at:
http://www.traceroute6.org/, but jv don't update this site...
> http://staff.csc.fi/psavola/lg.html
http://noc.ndsoftwarenet.com/docs/ipv6-tools-links.php
> Please send updates/modifications/etc directly. In particular, I failed
> to find the pages (if they exist) for the major v6 services providers..
http://www.dot-god.com/resources/IPV6.html
Best Regards,
Nicolas DEFFAYET
From pekkas@netcore.fi Sat Oct 19 15:18:24 2002
From: pekkas@netcore.fi (Pekka Savola)
Date: Sat, 19 Oct 2002 17:18:24 +0300 (EEST)
Subject: [6bone] A looking glass/traceroute page
In-Reply-To: <1035036803.610.1726.camel@wks1.fr.corp.ndsoftware.com>
Message-ID:
On 19 Oct 2002, Nicolas DEFFAYET wrote:
> On Sat, 2002-10-19 at 13:36, Pekka Savola wrote:
> Hello,
>
> > Failing to find good pages on collected looking glass/traceroute pointers
> > for v6, I did some quick browsing and collected some at:
>
> http://www.traceroute6.org/, but jv don't update this site...
Didn't know of this..
> > http://staff.csc.fi/psavola/lg.html
>
> http://noc.ndsoftwarenet.com/docs/ipv6-tools-links.php
Didn't know of this either... :-)
If you add some sane groupings, I don't want to update my list anymore, as
traceroute6 + you should be sufficient; no use having too many competing
pages, best to have only one or two.
--
Pekka Savola "Tell me of difficulties surmounted,
Netcore Oy not those you stumble over and fall"
Systems. Networks. Security. -- Robert Jordan: A Crown of Swords
From jv@pilsedu.cz Sat Oct 19 16:23:37 2002
From: jv@pilsedu.cz (Jakub Vlasek)
Date: Sat, 19 Oct 2002 17:23:37 +0200
Subject: [6bone] A looking glass/traceroute page
In-Reply-To: <1035036803.610.1726.camel@wks1.fr.corp.ndsoftware.com>
References: <1035036803.610.1726.camel@wks1.fr.corp.ndsoftware.com>
Message-ID: <20021019152337.GA13353@pilsedu.cz>
On Sat, Oct 19, 2002 at 04:13:23PM +0200, Nicolas DEFFAYET wrote:
> On Sat, 2002-10-19 at 13:36, Pekka Savola wrote:
> Hello,
>
> > Failing to find good pages on collected looking glass/traceroute pointers
> > for v6, I did some quick browsing and collected some at:
>
> http://www.traceroute6.org/, but jv don't update this site...
I *do* update www.traceroute6.org page. I wanted to publish this list on 6bone
ml very soon. I always welcome submissions to the list.
Best regards
Jakub Vlasek
From dragon@tdoi.org Sat Oct 19 17:02:03 2002
From: dragon@tdoi.org (Christian Nickel)
Date: Sat, 19 Oct 2002 18:02:03 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
References: <5.1.0.14.0.20021016061927.02891008@imap2.es.net>
Message-ID: <006e01c27788$de12b220$fd04a80a@alpha>
HOWTO get a "real" ASN and a pTLA
----------------------------------------
1. Get a "real" Autonomous System Number
1.1 Setup ZEBRA on two root servers, build some fake ISP and name it NDSOFTWARE
1.2 get some pTLA/sTLA sites who accept your private AS and peer with them
1.3 whining and begging at pTLA/sTLA site which normally don't accept private
peerings, until they peer with you
1.4 peering with other private AS's and assign them some adress space of your
three /32, which you begged from other 6bone backbone sites
1.5 repeat 1.3 and 1.4 until you have 101 peerings
1.6 spam to 6bone-ml, don't forget to write your name in uppercase
1.7 make new friends like Pim B. van Pelt
http://mailman.isi.edu/pipermail/6bone/2002-May/005470.html
http://mailman.isi.edu/pipermail/6bone/2002-May/005498.html
1.8 more spam to 6bone-ml, don't forget to write your name in uppercase
1.9 create a IPv6 only Internet Exchange Point
1.10 whining and begging at your local LIR until they send your request for
a ASN to RIR
1.11 RIR doesn't accept your ASN request
1.12 spam, whining and begging at 6bone-ml
1.13 finally you got your ASN
2. Get a pTLA
2.1 request a pTLA
2.2 your request is accepted, and review period begins
2.3 ok, now you can send some bad emails to your peering partners
like this one:
PS: don't forget to write your name uppercase!!!
------------
Hello,
You receive this email because you have a BGP peering with NDSoftware.
We change our peering policy.
We will shutdown your BGP peering the 23 October 2002 because you don't
respect our new peering policy with your private ASN.
Best Regards,
Nicolas DEFFAYET
------------
oh! what a coincidence "We will shutdown your BGP peering the 23 October 2002"
and "pTLA request NDSOFTWARE - review closes 23 October 2002"
PS: great policy http://noc.ndsoftwarenet.com/docs/peering-policy.php
2.4 you finally got your pTLA
2.5 you think private ASN sucks and remove all, peerings with private ASN
the annoyed german ipv6 community
-------------------------
ok now my personal opinion:
I don't support the pTLA request of NDSOFTWARE because:
Nicolas is an inmensly irritating clueless kiddie
I think he is just a kiddie who wants to impress other 6bone / ipv6 users.
I think his company NDSOFTWARE is no real ISP
I also think the other four persons in his request are fake or they don't know
what IPv6 is. (I have never seen or heared of anyone expect Nicolas)
Christian "_DrAGON_" Nickel
TDOI Network | dragon@tdoi.org | www.tdoi.org
From paitken@cisco.com Sat Oct 19 18:03:31 2002
From: paitken@cisco.com (Paul Aitken)
Date: Sat, 19 Oct 2002 18:03:31 +0100
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
References: <5.1.0.14.0.20021016061927.02891008@imap2.es.net>
Message-ID: <3DB19063.9020105@cisco.com>
Bob,
>> d. A fully maintained, and reliable, IPv6-accessible system
>> providing, at a mimimum, one or more web pages, describing the
>> Applicant's IPv6 services. This server must be IPv6 pingable.
>> http://noc.ndsoftwarenet.com/
Over IPv4, "This site is under contruction." [sic]
While this may be very precisely "describing the Applicant's IPv6
services", I feel that it fails to meet the spirit of the requirement.
Worse, over IPv6, only the default Apache screen is displayed. So while
the system may be IPv6 reachable, it can hardly be deemed accessible.
>> b. A common mailbox for support contact purposes that all support
>> staff have acess to, pointed to with a notify attribute in the
>> ipv6-site object for the pTLA Applicant.
>>
>> ipmaster@ndsoftwarenet.com
This entry is not "pointed to with a notify attribute in the ipv6-site
object for the pTLA Applicant".
Therefore, at this time I DO NOT "find their request fully compliant
with RFC2772".
Cheers.
--
Paul Aitken
IPv6 Development, Cisco Systems Ltd, Edinburgh, Scotland. EH6 6LX
From jeroen@unfix.org Sat Oct 19 18:14:47 2002
From: jeroen@unfix.org (Jeroen Massar)
Date: Sat, 19 Oct 2002 19:14:47 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
Message-ID: <000101c27793$0734e4c0$420d640a@unfix.org>
> NDSOFTWARE has requested a pTLA allocation and I find their request
fully
> compliant with RFC2772. The open review period for this will close 23
> October 2002. Please send your comments to me or the list.
>
>
>
>
I got a couple of questions about this request:
- Where/what is he going to use this pTLA for if it gets assigned?
(his "ISP" can't use it for _commercial_ purposes as 6bone is for
expermenting)
- What/Where is he currently using his _3_ /32's for?
- What is his planned numberplan?
- What is his planned/current backbone?
(This has to do with the fact that he is probably going to do transit
over it and
that will destabilize the 6bone even more...)
- What is going to happen with the other _3_ /32's he already has?
- Can't he get some IPv6 space from his upstreams?
Greets,
Jeroen
Notez bien that _3_ /32's he _already_ has, is more than any RIR
currently has!
From nicolas.deffayet@ndsoftware.net Sat Oct 19 18:15:28 2002
From: nicolas.deffayet@ndsoftware.net (Nicolas DEFFAYET)
Date: 19 Oct 2002 19:15:28 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <006e01c27788$de12b220$fd04a80a@alpha>
References: <5.1.0.14.0.20021016061927.02891008@imap2.es.net>
<006e01c27788$de12b220$fd04a80a@alpha>
Message-ID: <1035047728.636.1776.camel@wks1.fr.corp.ndsoftware.com>
On Sat, 2002-10-19 at 18:02, Christian Nickel wrote:
> Hello,
>
> You receive this email because you have a BGP peering with NDSoftware.
>
> We change our peering policy.
>
> We will shutdown your BGP peering the 23 October 2002 because you don't
> respect our new peering policy with your private ASN.
>
> Best Regards,
>
> Nicolas DEFFAYET
Why we close peering with private ASN the 23 October 2002:
We have 101 BGP4+ peer, our current routers are full (zebra is very
unstable if i add new peer) and we want get new peer with other
pTLA/sTLA that we can't get with our old private ASN.
We have choose to delete all peers with private ASN for free BGP session
on our routers for this new peers.
> oh! what a coincidence "We will shutdown your BGP peering the 23 October 2002"
> and "pTLA request NDSOFTWARE - review closes 23 October 2002"
>
> 2.4 you finally got your pTLA
> 2.5 you think private ASN sucks and remove all, peerings with private ASN
We don't delete peers with private ASN because "private ASN sucks", we
keep peering with important private ASN like NextGenCollective or
IPNG-UK (this 2 projects projet provide a lot of tunnels to users). I
understand their status, it's why i keep peering with them.
TDOI don't provide a lot of tunnels (more than 50) to users, it's for
that reason that we shutdown your peering.
Where i send you your certificate of the best troll of 6bone
mailing-list ?
Best Regards,
Nicolas DEFFAYET
From nicolas.deffayet@ndsoftware.net Sat Oct 19 18:33:35 2002
From: nicolas.deffayet@ndsoftware.net (Nicolas DEFFAYET)
Date: 19 Oct 2002 19:33:35 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <3DB19063.9020105@cisco.com>
References: <5.1.0.14.0.20021016061927.02891008@imap2.es.net>
<3DB19063.9020105@cisco.com>
Message-ID: <1035048815.629.1796.camel@wks1.fr.corp.ndsoftware.com>
On Sat, 2002-10-19 at 19:03, Paul Aitken wrote:
> >> d. A fully maintained, and reliable, IPv6-accessible system
> >> providing, at a mimimum, one or more web pages, describing the
> >> Applicant's IPv6 services. This server must be IPv6 pingable.
>
> >> http://noc.ndsoftwarenet.com/
>
> Over IPv4, "This site is under contruction." [sic]
Where ?
I don't see under construction...
>
> While this may be very precisely "describing the Applicant's IPv6
> services", I feel that it fails to meet the spirit of the requirement.
>
> Worse, over IPv6, only the default Apache screen is displayed. So while
> the system may be IPv6 reachable, it can hardly be deemed accessible.
Over IPv6 it's work fine, i can send you the log of access with IPv6
address since 05/Sep/2002.
3ffe:c15:c000:c:203:47ff:feb6:6a8b - - [19/Oct/2002:16:31:28 +0200] "GET
/tools/traceroute6.php HTTP/1.1" 200 10974 "http://www.traceroute6.org/"
"Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.0.1) Gecko/20020918"
3ffe:c15:c000:c:203:47ff:feb6:6a8b - - [19/Oct/2002:16:31:30 +0200] "GET
/interface.css HTTP/1.1" 200 1341
"http://noc.ndsoftwarenet.com/tools/traceroute6.php" "Mozilla/5.0 (X11;
U; Linux i686; en-US; rv:1.0.1) Gecko/20020918"
3ffe:c15:c000:c:203:47ff:feb6:6a8b - - [19/Oct/2002:16:31:31 +0200] "GET
/img/wi/top_logo.gif HTTP/1.1" 200 2998
"http://noc.ndsoftwarenet.com/tools/traceroute6.php" "Mozilla/5.0 (X11;
U; Linux i686; en-US; rv:1.0.1) Gecko/20020918"
3ffe:c15:c000:c:203:47ff:feb6:6a8b - - [19/Oct/2002:16:31:31 +0200] "GET
/img/wi/pixel.gif HTTP/1.1" 200 43
"http://noc.ndsoftwarenet.com/tools/traceroute6.php" "Mozilla/5.0 (X11;
U; Linux i686; en-US; rv:1.0.1) Gecko/20020918"
3ffe:c15:c000:c:203:47ff:feb6:6a8b - - [19/Oct/2002:16:31:31 +0200] "GET
/img/wi/pixel-999999.gif HTTP/1.1" 200 49
"http://noc.ndsoftwarenet.com/tools/traceroute6.php" "Mozilla/5.0 (X11;
U; Linux i686; en-US; rv:1.0.1) Gecko/20020918"
3ffe:c15:c000:c:203:47ff:feb6:6a8b - - [19/Oct/2002:16:31:31 +0200] "GET
/img/wi/arrow-u.gif HTTP/1.1" 200 54
"http://noc.ndsoftwarenet.com/tools/traceroute6.php" "Mozilla/5.0 (X11;
U; Linux i686; en-US; rv:1.0.1) Gecko/20020918"
3ffe:c15:c000:c:203:47ff:feb6:6a8b - - [19/Oct/2002:16:31:32 +0200] "GET
/img/wi/arrow-r.gif HTTP/1.1" 200 55
"http://noc.ndsoftwarenet.com/tools/traceroute6.php" "Mozilla/5.0 (X11;
U; Linux i686; en-US; rv:1.0.1) Gecko/20020918"
3ffe:c15:c000:c:203:47ff:feb6:6a8b - - [19/Oct/2002:16:31:41 +0200] "GET
/tools/traceroute6.php?host=ipv6-lab-gw.cisco.com HTTP/1.1" 200 11182
"http://noc.ndsoftwarenet.com/tools/traceroute6.php" "Mozilla/5.0 (X11;
U; Linux i686; en-US; rv:1.0.1) Gecko/20020918"
u1000000@wks1:~$ whois 3ffe:c15:c000:c:203:47ff:feb6:6a8b
% RIPEdb(3.0.0b2) with ISI RPSL extensions
inet6num: 3FFE:C00::/24
netname: CISCO
....
> >> b. A common mailbox for support contact purposes that all support
> >> staff have acess to, pointed to with a notify attribute in the
> >> ipv6-site object for the pTLA Applicant.
> >>
> >> ipmaster@ndsoftwarenet.com
>
> This entry is not "pointed to with a notify attribute in the ipv6-site
> object for the pTLA Applicant".
Yes, we use notify@ndsoftwarenet.com on notify attribute for have a
better management.
remarks: -----------------------------------------------------
remarks: Network problems: noc@ndsoftwarenet.com
remarks: Peering requests: peering@ndsoftwarenet.com
remarks: Abuse notifications: abuse@ndsoftwarenet.com
remarks: -----------------------------------------------------
I think that this remarks in our ipv6-site object is enough for contact
us.
Best Regards,
Nicolas DEFFAYET
From rmk@arm.linux.org.uk Sat Oct 19 18:43:29 2002
From: rmk@arm.linux.org.uk (Russell King)
Date: Sat, 19 Oct 2002 18:43:29 +0100
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <3DB19063.9020105@cisco.com>; from paitken@cisco.com on Sat, Oct 19, 2002 at 06:03:31PM +0100
References: <5.1.0.14.0.20021016061927.02891008@imap2.es.net> <3DB19063.9020105@cisco.com>
Message-ID: <20021019184329.B21819@flint.arm.linux.org.uk>
On Sat, Oct 19, 2002 at 06:03:31PM +0100, Paul Aitken wrote:
> >> d. A fully maintained, and reliable, IPv6-accessible system
> >> providing, at a mimimum, one or more web pages, describing the
> >> Applicant's IPv6 services. This server must be IPv6 pingable.
>
> >> http://noc.ndsoftwarenet.com/
>
> Over IPv4, "This site is under contruction." [sic]
>
> While this may be very precisely "describing the Applicant's IPv6
> services", I feel that it fails to meet the spirit of the requirement.
>
> Worse, over IPv6, only the default Apache screen is displayed. So while
> the system may be IPv6 reachable, it can hardly be deemed accessible.
Worksforme. Both with Mozilla and lynx. IPv6 and IPv4.
Maybe your client isn't sending the HTTP Host: header?
--
Russell King (rmk@arm.linux.org.uk) The developer of ARM Linux
http://www.arm.linux.org.uk/personal/aboutme.html
From paitken@cisco.com Sat Oct 19 18:50:27 2002
From: paitken@cisco.com (Paul Aitken)
Date: Sat, 19 Oct 2002 18:50:27 +0100
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
References: <5.1.0.14.0.20021016061927.02891008@imap2.es.net> <3DB19063.9020105@cisco.com> <1035048815.629.1796.camel@wks1.fr.corp.ndsoftware.com>
Message-ID: <3DB19B63.6040908@cisco.com>
Nicolas,
>>Over IPv4, "This site is under contruction." [sic]
>
> Where ?
> I don't see under construction...
I only quoted what I saw (including the spelling mistake).
> Over IPv6 it's work fine, i can send you the log of access with IPv6
> address since 05/Sep/2002.
I only say what I saw when I pointed my browser at
http://[3ffe:81f1:11:1:2::6]/ (and FWIW, it's not changed).
> u1000000@wks1:~$ whois 3ffe:c15:c000:c:203:47ff:feb6:6a8b
Yes, that's me.
>>>>ipmaster@ndsoftwarenet.com
>>>
>>This entry is not "pointed to with a notify attribute in the ipv6-site
>>object for the pTLA Applicant".
>
> Yes, we use notify@ndsoftwarenet.com on notify attribute for have a
> better management.
Then list this in your application!
Cheers.
--
Paul Aitken
IPv6 Development, Cisco Systems Ltd, Edinburgh, Scotland. EH6 6LX
From dragon@tdoi.org Sat Oct 19 19:06:34 2002
From: dragon@tdoi.org (Christian Nickel)
Date: Sat, 19 Oct 2002 20:06:34 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
References: <5.1.0.14.0.20021016061927.02891008@imap2.es.net> <006e01c27788$de12b220$fd04a80a@alpha> <1035047728.636.1776.camel@wks1.fr.corp.ndsoftware.com>
Message-ID: <003501c2779a$42c91ae0$fd04a80a@alpha>
Hello 6bone members,
I recieved the following mail from a french LIR that wants to stay anonymous:
>Christian,
>
>to be honnest, French IPv6 community is also very annoyed by Mr. Deffayet.
>Its behaviour is starting to get on our nerves.
>
>What you've forgotten to tell in your mail to the 6bone mailing-list is
>that Mr Deffayet does not hesitate to insult the peoples who refuse to
>grant him BGP sessions or tunnels.
>
>As maintainer of the xxx pTLA and maintainer of the xxx RIPE sTLA,
>I am used to be victim of Mr Deffayet's attacks only
>because I decline to offer him tunnel or BGP4+ for legitimate reasons.
>
>> ok now my personal opinion:
>>
>> I don't support the pTLA request of NDSOFTWARE because:
>
>I do TOTALY agree with you. It is a matter of survival for the IPv6
>worldwide experimentation and deployement not to allow Mr. Deffayet to
>play with global BGP.
>
>The reason is obvious : that kid (and yes, it is truely a kid and its
>compagny does NOT exists legally in France) is taking IPv6 as a HOBBY.
>It's goal is nothing more than collecting tunnels and BGP sessions without
>having any clues about avoiding trombonning, using MEDs or wise
>communities. Some peoples believe "NDSoftware" is the root cause (or at
>least a big part of the cause) of the "ghost AS_pathes" we saw emerging a
>few months ago.
>
>A hidden AS with so many BGP sessions redistributing everything with
>no-export is not a sane thing. It is even worse if some of the peers of
>that AS are not conforming to the no-export tag.
>
>When peoples like xxx, xxx, xxx and many other are making
>efforts to limit transcontinentals tunnels, trying to have v6 topology
>matching the underlying v4 one to optimize RTTs or to even spend money and
>time to deploy native IPv6, we are truely sad to see that a kid playing
>with some PCs running Zebra could annihilate all of those efforts.
>
>Bob, I really do think that you should not give a pTLA to NDSoftware for
>all the reasons here and all the reasons other have raised.
>
>PS: RIPE has been warned too.
PS: Nicolas i think you are the troll of 6bone or should i say of the complete
ipv6 community?
Christian "_DrAGON_" Nickel
TDOI Network | dragon@tdoi.org | www.tdoi.org
From tjc@ecs.soton.ac.uk Sat Oct 19 19:18:30 2002
From: tjc@ecs.soton.ac.uk (Tim Chown)
Date: Sat, 19 Oct 2002 19:18:30 +0100
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <1035047728.636.1776.camel@wks1.fr.corp.ndsoftware.com>
References: <5.1.0.14.0.20021016061927.02891008@imap2.es.net> <006e01c27788$de12b220$fd04a80a@alpha> <1035047728.636.1776.camel@wks1.fr.corp.ndsoftware.com>
Message-ID: <20021019181830.GC16714@login.ecs.soton.ac.uk>
On Sat, Oct 19, 2002 at 07:15:28PM +0200, Nicolas DEFFAYET wrote:
>
> We have 101 BGP4+ peer, our current routers are full (zebra is very
> unstable if i add new peer) and we want get new peer with other
> pTLA/sTLA that we can't get with our old private ASN.
So how many of these 101 BGP peerings are on direct native links?
I would have thought 5-10 would be more than ample. It's not a race to see
who can get the most(!)
> TDOI don't provide a lot of tunnels (more than 50) to users, it's for
> that reason that we shutdown your peering.
Tunnels to end users is one thing, the question is more on your infrastructure
peerings to other pTLAs (or SubTLAs, if any would peer with you...)
There are many networks trying to deploy IPv6 with some structure to the
connectivity, removing multi-hop IPv4 tunnels, trying to get a more predictable
IPv6 service, as more people begin to use IPv6 daily. I'm not sure that any
ISP with 101 BGP peerings is helping that. Perhaps you can explain your
goals in building these peerings?
Tim
From rico@noris.net Sat Oct 19 19:38:16 2002
From: rico@noris.net (Rico Gloeckner)
Date: Sat, 19 Oct 2002 20:38:16 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <1035047728.636.1776.camel@wks1.fr.corp.ndsoftware.com>; from nicolas.deffayet@ndsoftware.net on Sat, Oct 19, 2002 at 07:15:28PM +0200
References: <5.1.0.14.0.20021016061927.02891008@imap2.es.net> <006e01c27788$de12b220$fd04a80a@alpha> <1035047728.636.1776.camel@wks1.fr.corp.ndsoftware.com>
Message-ID: <20021019203815.M5088@noris.de>
On Sat, Oct 19, 2002 at 07:15:28PM +0200, Nicolas DEFFAYET wrote:
> Why we close peering with private ASN the 23 October 2002:
>
> We have 101 BGP4+ peer, our current routers are full (zebra is very
> unstable if i add new peer) and we want get new peer with other
> pTLA/sTLA that we can't get with our old private ASN.
>
> We don't delete peers with private ASN because "private ASN sucks", we
> keep peering with important private ASN like NextGenCollective or
> IPNG-UK (this 2 projects projet provide a lot of tunnels to users). I
> understand their status, it's why i keep peering with them.
>
> TDOI don't provide a lot of tunnels (more than 50) to users, it's for
> that reason that we shutdown your peering.
Ok, let me rephrase you:
| I close tunnels to small local Peers because i want to peer with Large
| Peers around the world.
Did I understand you correctly?
If so, i hope all your Peers will decrease the Priority to you to a very
minimum, because this is the best Way to fuck up IPv6-Routing.
-rg
PS: I dont see how colocated Boxes qualify for the Attribute "IX",
let it be alone "First IPv6 IX in $TLD", if so there were sure some more
IPv6-only IXs before you in your TLD.
From nicolas.deffayet@ndsoftware.net Sat Oct 19 19:58:15 2002
From: nicolas.deffayet@ndsoftware.net (Nicolas DEFFAYET)
Date: 19 Oct 2002 20:58:15 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <20021019203815.M5088@noris.de>
References: <5.1.0.14.0.20021016061927.02891008@imap2.es.net>
<006e01c27788$de12b220$fd04a80a@alpha>
<1035047728.636.1776.camel@wks1.fr.corp.ndsoftware.com>
<20021019203815.M5088@noris.de>
Message-ID: <1035053895.634.1978.camel@wks1.fr.corp.ndsoftware.com>
On Sat, 2002-10-19 at 20:38, Rico Gloeckner wrote:
> Ok, let me rephrase you:
> | I close tunnels to small local Peers because i want to peer with Large
> | Peers around the world.
>
> Did I understand you correctly?
>
> If so, i hope all your Peers will decrease the Priority to you to a very
> minimum, because this is the best Way to fuck up IPv6-Routing.
I use MED, i don't have a bad routing, you can check:
http://noc.ndsoftwarenet.com/stats/aspath-tree/bgp-page-complete.php
I see all european pTLA/sTLA by European peer,...
My MED for tunnels:
500: - 10 ms
510: 10 - 25 ms
520: 25 - 50 ms
530: 50 - 100 ms
540: + 100 ms
A lot of pTLA and sTLA don't use MED, i use MED for have a good quality.
> PS: I dont see how colocated Boxes qualify for the Attribute "IX",
> let it be alone "First IPv6 IX in $TLD", if so there were sure some more
> IPv6-only IXs before you in your TLD.
FNIX6 is the first IPv6 (IPv6 only of course) Internet Exchange in
France.
SFINX is the first IPv4/IPv6 (dualstack) Internet Exchange in France.
I have a full rack for FNIX6 with a switch in it.
Best Regards,
Nicolas DEFFAYET
From nicolas.deffayet@ndsoftware.net Sat Oct 19 20:26:22 2002
From: nicolas.deffayet@ndsoftware.net (Nicolas DEFFAYET)
Date: 19 Oct 2002 21:26:22 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <003501c2779a$42c91ae0$fd04a80a@alpha>
References: <5.1.0.14.0.20021016061927.02891008@imap2.es.net>
<006e01c27788$de12b220$fd04a80a@alpha>
<1035047728.636.1776.camel@wks1.fr.corp.ndsoftware.com>
<003501c2779a$42c91ae0$fd04a80a@alpha>
Message-ID: <1035055582.636.2069.camel@wks1.fr.corp.ndsoftware.com>
On Sat, 2002-10-19 at 20:06, Christian Nickel wrote:
> I recieved the following mail from a french LIR that wants to stay anonymous:
Very serious...
> >to be honnest, French IPv6 community is also very annoyed by Mr. Deffayet.
> >Its behaviour is starting to get on our nerves.
What's the French IPv6 community ?
You have _a_ (not the) French IPv6 community: IPv6-FR (on IRC #ipv6-fr @
irc.ndsoftwarenet.com / #ipv6-fr @ IRCnet).
> >What you've forgotten to tell in your mail to the 6bone mailing-list is
> >that Mr Deffayet does not hesitate to insult the peoples who refuse to
> >grant him BGP sessions or tunnels.
I never do this.
I want see this mail, send it with _FULL header_ on this list.
> >As maintainer of the xxx pTLA and maintainer of the xxx RIPE sTLA,
It's ISDnet, why hide it ?
> >I am used to be victim of Mr Deffayet's attacks only
> >because I decline to offer him tunnel or BGP4+ for legitimate reasons.
I never attack this administrator.
> >> ok now my personal opinion:
> >>
> >> I don't support the pTLA request of NDSOFTWARE because:
> >
> >I do TOTALY agree with you. It is a matter of survival for the IPv6
> >worldwide experimentation and deployement not to allow Mr. Deffayet to
> >play with global BGP.
Very funny.
Ask many other administrators, no problem with my BGP.
When i used my private ASN, i always announced my 3 /32 with community
no-export and never sent a full transit.
> >The reason is obvious : that kid (and yes, it is truely a kid and its
> >compagny does NOT exists legally in France) is taking IPv6 as a HOBBY.
> >It's goal is nothing more than collecting tunnels and BGP sessions without
> >having any clues about avoiding trombonning, using MEDs or wise
> >communities. Some peoples believe "NDSoftware" is the root cause (or at
> >least a big part of the cause) of the "ghost AS_pathes" we saw emerging a
> >few months ago.
NDSoftware exist, IPv6-FR exist.
With many tunnels, i have do a lot of tests, i know now the limit of
Zebra, what's the best solution for manage a lot of peering,...
I use MED, i don't have a bad routing, you can check:
http://noc.ndsoftwarenet.com/stats/aspath-tree/bgp-page-complete.php
I see all european pTLA/sTLA by European peer,...
My MED for tunnels:
500: - 10 ms
510: 10 - 25 ms
520: 25 - 50 ms
530: 50 - 100 ms
540: + 100 ms
A lot of pTLA and sTLA don't use MED, i use MED for have a good quality.
I'm not the source of this "ghost AS_pathes", read archive of this list.
> >A hidden AS with so many BGP sessions redistributing everything with
> >no-export is not a sane thing. It is even worse if some of the peers of
> >that AS are not conforming to the no-export tag.
When peer didn't understand community no-export, i didn't send any
routes.
> >When peoples like xxx, xxx, xxx and many other are making
> >efforts to limit transcontinentals tunnels, trying to have v6 topology
> >matching the underlying v4 one to optimize RTTs or to even spend money and
> >time to deploy native IPv6, we are truely sad to see that a kid playing
> >with some PCs running Zebra could annihilate all of those efforts.
I use MED....
I see all european pTLA/sTLA by European peer,...
I'm not a kid.
I don't have Cisco or Juniper routers because i don't have the budget
for that. You can offer me a Cisco if you want...
> >Bob, I really do think that you should not give a pTLA to NDSoftware for
> >all the reasons here and all the reasons other have raised.
> >
> >PS: RIPE has been warned too.
I respect all 6bone rules...
> PS: Nicolas i think you are the troll of 6bone or should i say of the complete
> ipv6 community?
Yes, you are a troll expert with your friends.
From nicolas.deffayet@ndsoftware.net Sat Oct 19 20:30:14 2002
From: nicolas.deffayet@ndsoftware.net (Nicolas DEFFAYET)
Date: 19 Oct 2002 21:30:14 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <000101c27793$0734e4c0$420d640a@unfix.org>
References: <000101c27793$0734e4c0$420d640a@unfix.org>
Message-ID: <1035055814.629.2079.camel@wks1.fr.corp.ndsoftware.com>
On Sat, 2002-10-19 at 19:14, Jeroen Massar wrote:
>
> > NDSOFTWARE has requested a pTLA allocation and I find their request
> fully
> > compliant with RFC2772. The open review period for this will close 23
> > October 2002. Please send your comments to me or the list.
> >
> >
> >
> >
>
> I got a couple of questions about this request:
>
> - Where/what is he going to use this pTLA for if it gets assigned?
> (his "ISP" can't use it for _commercial_ purposes as 6bone is for
> expermenting)
I provide IPv6 connectivity to many projects with no commercial
objectives of course.
> - What/Where is he currently using his _3_ /32's for?
- Internal test
- IPv6 network
- FNIX6
- IPv6-FR
- ftp://ftp.openipv6.com
- Mirrors of website
- Delegation to many projects
- ...
> - What is his planned/current backbone?
> (This has to do with the fact that he is probably going to do transit
> over it and
> that will destabilize the 6bone even more...)
I provide transit only if peer request it.
I don't will destabilize the 6bone !
> - What is going to happen with the other _3_ /32's he already has?
I will return this 3 /32 to their owner when i will get my pTLA.
> - Can't he get some IPv6 space from his upstreams?
No, i want announce a pTLA.
Best Regards,
Nicolas DEFFAYET
From nicolas.deffayet@ndsoftware.net Sat Oct 19 20:36:55 2002
From: nicolas.deffayet@ndsoftware.net (Nicolas DEFFAYET)
Date: 19 Oct 2002 21:36:55 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <20021019181830.GC16714@login.ecs.soton.ac.uk>
References: <5.1.0.14.0.20021016061927.02891008@imap2.es.net>
<006e01c27788$de12b220$fd04a80a@alpha>
<1035047728.636.1776.camel@wks1.fr.corp.ndsoftware.com>
<20021019181830.GC16714@login.ecs.soton.ac.uk>
Message-ID: <1035056215.631.2088.camel@wks1.fr.corp.ndsoftware.com>
On Sat, 2002-10-19 at 20:18, Tim Chown wrote:
> There are many networks trying to deploy IPv6 with some structure to the
> connectivity, removing multi-hop IPv4 tunnels, trying to get a more predictable
> IPv6 service, as more people begin to use IPv6 daily. I'm not sure that any
> ISP with 101 BGP peerings is helping that. Perhaps you can explain your
> goals in building these peerings?
With many peerings, i do a lot of tests, i know now the limit of
Zebra, what's the best solution for manage a lot of peering,...
I don't provide transit to all of this peers, i can't be a problem.
Best Regards,
Nicolas DEFFAYET
From jeroen@unfix.org Sat Oct 19 20:39:43 2002
From: jeroen@unfix.org (Jeroen Massar)
Date: Sat, 19 Oct 2002 21:39:43 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <1035055814.629.2079.camel@wks1.fr.corp.ndsoftware.com>
Message-ID: <000f01c277a7$4677df20$210d640a@unfix.org>
Nicolas DEFFAYET [mailto:nicolas.deffayet@ndsoftware.net] wrote:
> On Sat, 2002-10-19 at 19:14, Jeroen Massar wrote:
> >
> > > NDSOFTWARE has requested a pTLA allocation and I find
> their request
> > fully
> > > compliant with RFC2772. The open review period for this
> will close 23
> > > October 2002. Please send your comments to me or the list.
> > >
> > >
> > >
> > >
> >
> > I got a couple of questions about this request:
> >
> > - Where/what is he going to use this pTLA for if it gets assigned?
> > (his "ISP" can't use it for _commercial_ purposes as 6bone is for
> > expermenting)
>
> I provide IPv6 connectivity to many projects with no commercial
> objectives of course.
>
> > - What/Where is he currently using his _3_ /32's for?
>
> - Internal test
> - IPv6 network
> - FNIX6
> - IPv6-FR
> - ftp://ftp.openipv6.com
> - Mirrors of website
> - Delegation to many projects
> - ...
You need _3_ /32's for a couple of MIRRORS ??????
Come on... get a grip... I know there should be enough IPv6 space
but that is ludacrist!
FNIX doesn't exist as it doesn't have any members, and even if it had
an IX doesn't need any global IP space. (see AMS-IX cases for requesting
it).
> > - What is his planned/current backbone?
> > (This has to do with the fact that he is probably going to do
transit
> > over it and
> > that will destabilize the 6bone even more...)
>
> I provide transit only if peer request it.
> I don't will destabilize the 6bone !
You don't or you will ? What is it?
I don't have a de-yoda-fier here, sorry.
> > - What is going to happen with the other _3_ /32's he already has?
>
> I will return this 3 /32 to their owner when i will get my pTLA.
That's too be expected...
> > - Can't he get some IPv6 space from his upstreams?
>
> No, i want announce a pTLA.
You want to announce a pTLA for a couple of MIRRORS from 2 colo boxes???
Cool, I can mirror some stuff, now I want my personal pTLA too!
Nicolas, mind you that the p in pTLA doesn't stand for 'personal'.
It's for experimental *deployment*, eg providing IPv6 access to users in
experimental stadia, not for having lots of IP space for ftp/www
mirrors.
Greets,
Jeroen
From nicolas.deffayet@ndsoftware.net Sat Oct 19 20:49:35 2002
From: nicolas.deffayet@ndsoftware.net (Nicolas DEFFAYET)
Date: 19 Oct 2002 21:49:35 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <000f01c277a7$4677df20$210d640a@unfix.org>
References: <000f01c277a7$4677df20$210d640a@unfix.org>
Message-ID: <1035056975.634.2109.camel@wks1.fr.corp.ndsoftware.com>
On Sat, 2002-10-19 at 21:39, Jeroen Massar wrote:
> > > - What/Where is he currently using his _3_ /32's for?
> >
> > - Internal test
> > - IPv6 network
> > - FNIX6
> > - IPv6-FR
> > - ftp://ftp.openipv6.com
> > - Mirrors of website
> > - Delegation to many projects
> > - ...
>
> You need _3_ /32's for a couple of MIRRORS ??????
> Come on... get a grip... I know there should be enough IPv6 space
> but that is ludacrist!
I will return this 3 /32 to their owner when i will get my pTLA.
> FNIX doesn't exist as it doesn't have any members, and even if it had
> an IX doesn't need any global IP space. (see AMS-IX cases for requesting
> it).
I can't request a /48 for FNIX6 to RIPE because NDSoftware is not LIR.
> > > - What is his planned/current backbone?
> > > (This has to do with the fact that he is probably going to do
> transit
> > > over it and
> > > that will destabilize the 6bone even more...)
> >
> > I provide transit only if peer request it.
> > I don't will destabilize the 6bone !
>
> You don't or you will ? What is it?
I never destabilize the 6bone, i don't destabilize the 6bone and i will
never destabilize the 6bone.
> > > - Can't he get some IPv6 space from his upstreams?
> >
> > No, i want announce a pTLA.
>
> You want to announce a pTLA for a couple of MIRRORS from 2 colo boxes???
I don't have only this...
I delegate a lof of IPv6 blocks to many projects, see my whois.
> Cool, I can mirror some stuff, now I want my personal pTLA too!
It's NOT a personal pTLA.
I request it for NDSoftware and for many projects.
> Nicolas, mind you that the p in pTLA doesn't stand for 'personal'.
> It's for experimental *deployment*, eg providing IPv6 access to users in
> experimental stadia, not for having lots of IP space for ftp/www
> mirrors.
Yes, i provide IPv6 access to users, see my whois.
Best Regards,
Nicolas DEFFAYET
From bortzmeyer@gitoyen.net Sat Oct 19 20:54:48 2002
From: bortzmeyer@gitoyen.net (Stephane Bortzmeyer)
Date: Sat, 19 Oct 2002 21:54:48 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <003501c2779a$42c91ae0$fd04a80a@alpha>
("Christian Nickel" 's message of
Sat, 19 Oct 2002 20:06:34 +0200)
Message-ID: <200210191954.g9JJsmgj023818@ludwigV.sources.org>
On Saturday 19 October 2002, at 20 h 6,
"Christian Nickel" wrote:
> I recieved the following mail from a french LIR that wants to stay anonymous:
As a French LIR myself, I can say that I find annoying that cowards' comments
are forwarded to the 6bone list.
Nicolas Deffayet is obnoxious enough with what he does publicly (such as
pretending his toy is the "first IPv6 IXP in France"). No need to add
anonymous accusations.
> >time to deploy native IPv6, we are truely sad to see that a kid playing
> >with some PCs running Zebra could annihilate all of those efforts.
That b...s...t about Zebra (which we use in production) is one more thing that
should warn 6bone readers against that anonymous comment.
From jeroen@unfix.org Sat Oct 19 21:08:39 2002
From: jeroen@unfix.org (Jeroen Massar)
Date: Sat, 19 Oct 2002 22:08:39 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <1035056975.634.2109.camel@wks1.fr.corp.ndsoftware.com>
Message-ID: <001701c277ab$513eac50$210d640a@unfix.org>
Nicolas DEFFAYET [mailto:nicolas.deffayet@ndsoftware.net] wrote:
> On Sat, 2002-10-19 at 21:39, Jeroen Massar wrote:
> > > > - What/Where is he currently using his _3_ /32's for?
> > >
> > > - Internal test
> > > - IPv6 network
> > > - FNIX6
> > > - IPv6-FR
> > > - ftp://ftp.openipv6.com
> > > - Mirrors of website
> > > - Delegation to many projects
> > > - ...
> >
> > You need _3_ /32's for a couple of MIRRORS ??????
> > Come on... get a grip... I know there should be enough IPv6 space
> > but that is ludacrist!
>
> I will return this 3 /32 to their owner when i will get my pTLA.
Have you got, or better does a numberplan even exist for any of these?
You might consider returning them now.
> > FNIX doesn't exist as it doesn't have any members, and even
> if it had
> > an IX doesn't need any global IP space. (see AMS-IX cases
> for requesting
> > it).
>
> I can't request a /48 for FNIX6 to RIPE because NDSoftware is not LIR.
No, that's the complete idea, persons should not be able to do that.
> > > > - What is his planned/current backbone?
> > > > (This has to do with the fact that he is probably going to do
> > transit
> > > > over it and
> > > > that will destabilize the 6bone even more...)
> > >
> > > I provide transit only if peer request it.
> > > I don't will destabilize the 6bone !
> >
> > You don't or you will ? What is it?
>
> I never destabilize the 6bone, i don't destabilize the 6bone
> and i will never destabilize the 6bone.
Can you make up a signed document in triplet,
postal mail that to 6bone secretariat and put up a scanned version too?
> > > > - Can't he get some IPv6 space from his upstreams?
> > >
> > > No, i want announce a pTLA.
> >
> > You want to announce a pTLA for a couple of MIRRORS from 2
> colo boxes???
>
> I don't have only this...
> I delegate a lof of IPv6 blocks to many projects, see my whois.
"I" "I" , "my" whois?
> > Cool, I can mirror some stuff, now I want my personal pTLA too!
>
> It's NOT a personal pTLA.
>
> I request it for NDSoftware and for many projects.
That's why you say "I" 4 lines above this.
And only "I" and "my" and nothing else, never "ndsoftware".
Where is the "rest" of NDSoftware (Nicolas DEFFAYET Software).
> > Nicolas, mind you that the p in pTLA doesn't stand for 'personal'.
> > It's for experimental *deployment*, eg providing IPv6
> access to users in
> > experimental stadia, not for having lots of IP space for ftp/www
> > mirrors.
>
> Yes, i provide IPv6 access to users, see my whois.
I found 10 "user" inet6nums in the 6bone whois database.
They where all sized /44, that's quite big, present a numberplan please.
You don't need a pTLA, you need a delegation from NERIM, they are your
upstream.
And please think up new answers and answer the questions.
I get quite annoyed by receiving the same nonsense over and over.
Greets,
Jeroen
From bortzmeyer@gitoyen.net Sat Oct 19 21:10:41 2002
From: bortzmeyer@gitoyen.net (Stephane Bortzmeyer)
Date: Sat, 19 Oct 2002 22:10:41 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <1035056975.634.2109.camel@wks1.fr.corp.ndsoftware.com>
(Nicolas DEFFAYET 's message of
19 Oct 2002 21:49:35 +0200)
Message-ID: <200210192010.g9JKAfgj024322@ludwigV.sources.org>
On Saturday 19 October 2002, at 21 h 49,
Nicolas DEFFAYET wrote:
> > FNIX doesn't exist as it doesn't have any members, and even if it had
> > an IX doesn't need any global IP space. (see AMS-IX cases for requesting
> > it).
>
> I can't request a /48 for FNIX6 to RIPE because NDSoftware is not LIR.
You can always go through an existing LIR. Gitoyen could certainly do it, even if I find the idea of an IPv6-only IXP absolutely wrong.
I strongly disagree with Jeroen, an IXP should really have global and routable IP addresses.
From nicolas.deffayet@ndsoftware.net Sat Oct 19 21:27:33 2002
From: nicolas.deffayet@ndsoftware.net (Nicolas DEFFAYET)
Date: 19 Oct 2002 22:27:33 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <001701c277ab$513eac50$210d640a@unfix.org>
References: <001701c277ab$513eac50$210d640a@unfix.org>
Message-ID: <1035059253.610.2135.camel@wks1.fr.corp.ndsoftware.com>
On Sat, 2002-10-19 at 22:08, Jeroen Massar wrote:
> Nicolas DEFFAYET [mailto:nicolas.deffayet@ndsoftware.net] wrote:
>
> > On Sat, 2002-10-19 at 21:39, Jeroen Massar wrote:
> > > > > - What/Where is he currently using his _3_ /32's for?
> > > >
> > > > - Internal test
> > > > - IPv6 network
> > > > - FNIX6
> > > > - IPv6-FR
> > > > - ftp://ftp.openipv6.com
> > > > - Mirrors of website
> > > > - Delegation to many projects
> > > > - ...
> > >
> > > You need _3_ /32's for a couple of MIRRORS ??????
> > > Come on... get a grip... I know there should be enough IPv6 space
> > > but that is ludacrist!
> >
> > I will return this 3 /32 to their owner when i will get my pTLA.
> Have you got, or better does a numberplan even exist for any of these?
> You might consider returning them now.
I will move the 3 /32 in the pTLA.
For the moment, i use:
/35 DELEGATION (/48, /44, /40)
/35 IPV6-FR
/36 NEXTGENCOLLECTIVE
/36 NDSOFTWARE
A lot of /24 and /28 pTLA use only one /48 in their pTLA...
I hope that you see that i need a pTLA.
> > > Cool, I can mirror some stuff, now I want my personal pTLA too!
> >
> > It's NOT a personal pTLA.
> >
> > I request it for NDSoftware and for many projects.
> That's why you say "I" 4 lines above this.
> And only "I" and "my" and nothing else, never "ndsoftware".
Don't play with words.
> Where is the "rest" of NDSoftware (Nicolas DEFFAYET Software).
I'm the founder of NDSoftware, yes NDSoftware = Nicolas DEFFAYET
Software.
NDSoftware exist since 03/2000.
I use "I" for call NDSoftware because it's my company.
> You don't need a pTLA, you need a delegation from NERIM, they are your
> upstream.
No, i need a pTLA.
Best Regards,
Nicolas DEFFAYET
From nicolas.deffayet@ndsoftware.net Sat Oct 19 21:34:50 2002
From: nicolas.deffayet@ndsoftware.net (Nicolas DEFFAYET)
Date: 19 Oct 2002 22:34:50 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <200210191954.g9JJsmgj023818@ludwigV.sources.org>
References: <200210191954.g9JJsmgj023818@ludwigV.sources.org>
Message-ID: <1035059690.626.2145.camel@wks1.fr.corp.ndsoftware.com>
On Sat, 2002-10-19 at 21:54, Stephane Bortzmeyer wrote:
> On Saturday 19 October 2002, at 20 h 6,
> "Christian Nickel" wrote:
>
> > I recieved the following mail from a french LIR that wants to stay anonymous:
>
> As a French LIR myself, I can say that I find annoying that cowards' comments
> are forwarded to the 6bone list.
LIR != god of Internet
> Nicolas Deffayet is obnoxious enough with what he does publicly (such as
> pretending his toy is the "first IPv6 IXP in France"). No need to add
> anonymous accusations.
FNIX6 is the first french IPv6 Internet Exchange.
IPv6 Internet Exchange = IPv6 only
Why do you don't want see the reality ?
Check all french IX websites, you will find that FNIX6 is the first that
offer IPv6 without IPv4.
> > >time to deploy native IPv6, we are truely sad to see that a kid playing
> > >with some PCs running Zebra could annihilate all of those efforts.
>
> That b...s...t about Zebra (which we use in production) is one more thing that
> should warn 6bone readers against that anonymous comment.
Yes, Zebra is very stable.
Best Regards,
Nicolas DEFFAYET
From jeroen@unfix.org Sat Oct 19 21:41:12 2002
From: jeroen@unfix.org (Jeroen Massar)
Date: Sat, 19 Oct 2002 22:41:12 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <1035059253.610.2135.camel@wks1.fr.corp.ndsoftware.com>
Message-ID: <001f01c277af$dd0aa640$210d640a@unfix.org>
Nicolas DEFFAYET [mailto:nicolas.deffayet@ndsoftware.net] wrote:
> On Sat, 2002-10-19 at 22:08, Jeroen Massar wrote:
> > Nicolas DEFFAYET [mailto:nicolas.deffayet@ndsoftware.net] wrote:
> >
> > > On Sat, 2002-10-19 at 21:39, Jeroen Massar wrote:
> > > > > > - What/Where is he currently using his _3_ /32's for?
> > > > >
> > > > > - Internal test
> > > > > - IPv6 network
> > > > > - FNIX6
> > > > > - IPv6-FR
> > > > > - ftp://ftp.openipv6.com
> > > > > - Mirrors of website
> > > > > - Delegation to many projects
> > > > > - ...
> > > >
> > > > You need _3_ /32's for a couple of MIRRORS ??????
> > > > Come on... get a grip... I know there should be enough IPv6
space
> > > > but that is ludacrist!
> > >
> > > I will return this 3 /32 to their owner when i will get my pTLA.
> > Have you got, or better does a numberplan even exist for
> any of these?
> > You might consider returning them now.
>
> I will move the 3 /32 in the pTLA.
>
> For the moment, i use:
_I_ use.
> /35 DELEGATION (/48, /44, /40)
For what? Llama's? Vhosts? Mirrors?
> /35 IPV6-FR
> /36 NEXTGENCOLLECTIVE
> /36 NDSOFTWARE
>
> A lot of /24 and /28 pTLA use only one /48 in their pTLA...
>
> I hope that you see that i need a pTLA.
No I don't see any reason for a pTLA.
I do see a reason for making a numberplan.
How many users are in this 3x /32?
A very big ISP can provide IPv6 to whole china with it.
How many 'users' are there in NDSOFTWARE that you need
and can justify a /36 ???
Where did FNIX6 go? Oh yeah it doesn't exist checking the site.
If all _companies_ around this globe allocate like that we will for sure
be needing a 512bits address space.
> > > > Cool, I can mirror some stuff, now I want my personal pTLA too!
> > >
> > > It's NOT a personal pTLA.
> > >
> > > I request it for NDSoftware and for many projects.
> > That's why you say "I" 4 lines above this.
> > And only "I" and "my" and nothing else, never "ndsoftware".
>
> Don't play with words.
_I_ don't play with words. You are, I did watch Starwars I and
understood
Yoda then, but I didn't get a clue about what he is saying in part 6.
> > Where is the "rest" of NDSoftware (Nicolas DEFFAYET Software).
>
> I'm the founder of NDSoftware, yes NDSoftware = Nicolas DEFFAYET
> Software.
> NDSoftware exist since 03/2000.
> I use "I" for call NDSoftware because it's my company.
And nobody else now is there?
Who are the other people mentioned in _your_ whois ?
> > You don't need a pTLA, you need a delegation from NERIM,
> they are your
> > upstream.
>
> No, i need a pTLA.
I again. YOU don't need a pTLA.
WHY ? Oh I'll spell it again : W.H.Y.
Thats: Double-U, Ache, Why: Why.
First justify WHY _your company_ needs it.
Then rememeber that it's for EXPERIMENTAL purposes
So you can't use it COMMERCIALLY.
And again I ask you, if you reply and _your company_ still
want to request a pTLA: answer the above questions in full without
cutting.
Greets,
Jeroen
From bortzmeyer@gitoyen.net Sat Oct 19 21:46:47 2002
From: bortzmeyer@gitoyen.net (Stephane Bortzmeyer)
Date: Sat, 19 Oct 2002 22:46:47 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <1035059690.626.2145.camel@wks1.fr.corp.ndsoftware.com>
(Nicolas DEFFAYET 's message of
19 Oct 2002 22:34:50 +0200)
Message-ID: <200210192046.g9JKklgj025269@ludwigV.sources.org>
On Saturday 19 October 2002, at 22 h 34,
Nicolas DEFFAYET wrote:
> LIR != god of Internet
Yes, to become a LIR, you need to be a demi-god, you need to be able to
understand RIPE documents and to format properly a request.
> FNIX6 is the first french IPv6 Internet Exchange.
> IPv6 Internet Exchange = IPv6 only
Unlike what happens in the novel "1984", you cannot redefine the language at
will. For everybody, a "IPv6 Internet Exchange" is an IX where you can run
IPv6. You know it fairly well but you choosed, by pure dishonesty, to use the
sentence "the first french IPv6 Internet Exchange" knowing it would confuse
people.
BTW, I have just started to manufacture the first floating-point processor in
the world (yes, other processors do floating-point calculations but mine is
the first to perform only FP operations).
> Check all french IX websites, you will find that FNIX6 is the first that
> offer IPv6 without IPv4.
I agree. If you are intellectually honest, you will change your signature to :
FNIX6 is the first french Internet Exchange without IPv4
From jeroen@unfix.org Sat Oct 19 21:49:50 2002
From: jeroen@unfix.org (Jeroen Massar)
Date: Sat, 19 Oct 2002 22:49:50 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <200210192010.g9JKAfgj024322@ludwigV.sources.org>
Message-ID: <002701c277b1$12200400$210d640a@unfix.org>
Stephane Bortzmeyer wrote:
> On Saturday 19 October 2002, at 21 h 49,
> Nicolas DEFFAYET wrote:
>
> > > FNIX doesn't exist as it doesn't have any members, and
> even if it had
> > > an IX doesn't need any global IP space. (see AMS-IX cases
> for requesting
> > > it).
> I strongly disagree with Jeroen, an IXP should really
> have global and routable IP addresses.
I have to rectify that you misread me Stephane.
I do want to see the ability for an IXP to get address space that is
routable. Which is why I mentioned the AMS-IX case, which is a
"non-profit, neutral and independent association" and thus can't rely on
other peoples
IP space. The shared medium can run only on the /48, but the services
provided (website, lookingglasses etc) can't. AMS-IX has major
difficulties
getting that space. Notez bien that even RIPE can't request space from
thereselves as they are not a LIR. Odd rules in a odd world.
Greets,
Jeroen
From nicolas.deffayet@ndsoftware.net Sat Oct 19 22:00:37 2002
From: nicolas.deffayet@ndsoftware.net (Nicolas DEFFAYET)
Date: 19 Oct 2002 23:00:37 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <001f01c277af$dd0aa640$210d640a@unfix.org>
References: <001f01c277af$dd0aa640$210d640a@unfix.org>
Message-ID: <1035061237.636.2158.camel@wks1.fr.corp.ndsoftware.com>
On Sat, 2002-10-19 at 22:41, Jeroen Massar wrote:
> >
> > I will move the 3 /32 in the pTLA.
> >
> > For the moment, i use:
> _I_ use.
Why do you play with words ?
It's funny for you ?
> > /35 DELEGATION (/48, /44, /40)
> For what? Llama's? Vhosts? Mirrors?
read my request of pTLA to Bob Fink.
> > /35 IPV6-FR
> > /36 NEXTGENCOLLECTIVE
> > /36 NDSOFTWARE
> >
> > A lot of /24 and /28 pTLA use only one /48 in their pTLA...
> >
> > I hope that you see that i need a pTLA.
>
> No I don't see any reason for a pTLA.
> I do see a reason for making a numberplan.
> How many users are in this 3x /32?
> A very big ISP can provide IPv6 to whole china with it.
> How many 'users' are there in NDSOFTWARE that you need
> and can justify a /36 ???
Why you need know my number plan ?
Other pTLA didn't published their number plan for get their pTLA...
> Where did FNIX6 go? Oh yeah it doesn't exist checking the site.
http://www.fnix6.net.net/, this URL is in my request of pTLA
> > > Where is the "rest" of NDSoftware (Nicolas DEFFAYET Software).
> >
> > I'm the founder of NDSoftware, yes NDSoftware = Nicolas DEFFAYET
> > Software.
> > NDSoftware exist since 03/2000.
> > I use "I" for call NDSoftware because it's my company.
> And nobody else now is there?
> Who are the other people mentioned in _your_ whois ?
It's like you said "my family".
It's people who work on the IPv6 project with me.
> > > You don't need a pTLA, you need a delegation from NERIM,
> > they are your
> > > upstream.
> >
> > No, i need a pTLA.
>
> I again. YOU don't need a pTLA.
> WHY ? Oh I'll spell it again : W.H.Y.
> Thats: Double-U, Ache, Why: Why.
>
> First justify WHY _your company_ needs it.
> Then rememeber that it's for EXPERIMENTAL purposes
> So you can't use it COMMERCIALLY.
>
> And again I ask you, if you reply and _your company_ still
> want to request a pTLA: answer the above questions in full without
> cutting.
I need a pTLA, i have sent my request with my justification to Bob Fink
and he had sent to this mailing list.
Read it before ask why why why !
Ask other pTLA why they have a pTLA...
From nicolas.deffayet@ndsoftware.net Sat Oct 19 22:04:33 2002
From: nicolas.deffayet@ndsoftware.net (Nicolas DEFFAYET)
Date: 19 Oct 2002 23:04:33 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <200210192046.g9JKklgj025269@ludwigV.sources.org>
References: <200210192046.g9JKklgj025269@ludwigV.sources.org>
Message-ID: <1035061473.624.2163.camel@wks1.fr.corp.ndsoftware.com>
On Sat, 2002-10-19 at 22:46, Stephane Bortzmeyer wrote:
> > FNIX6 is the first french IPv6 Internet Exchange.
> > IPv6 Internet Exchange = IPv6 only
>
> Unlike what happens in the novel "1984", you cannot redefine the language at
> will. For everybody, a "IPv6 Internet Exchange" is an IX where you can run
> IPv6. You know it fairly well but you choosed, by pure dishonesty, to use the
> sentence "the first french IPv6 Internet Exchange" knowing it would confuse
> people.
IPv4 Internet Exchange = IPv4 only
IPv4 and IPv6 Internet Exchange = IPv4 and IPv6 (dualstack)
IPv6 Internet Exchange = IPv6 only
> BTW, I have just started to manufacture the first floating-point processor in
> the world (yes, other processors do floating-point calculations but mine is
> the first to perform only FP operations).
>
> > Check all french IX websites, you will find that FNIX6 is the first that
> > offer IPv6 without IPv4.
>
> I agree. If you are intellectually honest, you will change your signature to :
>
> FNIX6 is the first french Internet Exchange without IPv4
Best Regards,
Nicolas DEFFAYET
FNIX6 is the first french Internet Exchange without IPv4:
http://www.fnix6.net/
From pekkas@netcore.fi Sat Oct 19 22:03:23 2002
From: pekkas@netcore.fi (Pekka Savola)
Date: Sun, 20 Oct 2002 00:03:23 +0300 (EEST)
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <1035053895.634.1978.camel@wks1.fr.corp.ndsoftware.com>
Message-ID:
On 19 Oct 2002, Nicolas DEFFAYET wrote:
> On Sat, 2002-10-19 at 20:38, Rico Gloeckner wrote:
>
> > Ok, let me rephrase you:
> > | I close tunnels to small local Peers because i want to peer with Large
> > | Peers around the world.
> >
> > Did I understand you correctly?
> >
> > If so, i hope all your Peers will decrease the Priority to you to a very
> > minimum, because this is the best Way to fuck up IPv6-Routing.
>
> I use MED, i don't have a bad routing, you can check:
>
> http://noc.ndsoftwarenet.com/stats/aspath-tree/bgp-page-complete.php
>
> I see all european pTLA/sTLA by European peer,...
>
> My MED for tunnels:
>
> 500: - 10 ms
> 510: 10 - 25 ms
> 520: 25 - 50 ms
> 530: 50 - 100 ms
> 540: + 100 ms
>
> A lot of pTLA and sTLA don't use MED, i use MED for have a good quality.
Are you aware of the fact that unless you have 'always-compare-med' (or
equivalent), MED is only used when comparing paths for which the neighbor
AS is the same? So, unless you have two connections to a neighbor AS,
it's of no use?
Pretty much useless in the 6bone context otherwise.
--
Pekka Savola "Tell me of difficulties surmounted,
Netcore Oy not those you stumble over and fall"
Systems. Networks. Security. -- Robert Jordan: A Crown of Swords
From nicolas.deffayet@ndsoftware.net Sat Oct 19 22:07:25 2002
From: nicolas.deffayet@ndsoftware.net (Nicolas DEFFAYET)
Date: 19 Oct 2002 23:07:25 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To:
References:
Message-ID: <1035061646.636.2166.camel@wks1.fr.corp.ndsoftware.com>
On Sat, 2002-10-19 at 23:03, Pekka Savola wrote:
> On 19 Oct 2002, Nicolas DEFFAYET wrote:
> > On Sat, 2002-10-19 at 20:38, Rico Gloeckner wrote:
> >
> > > Ok, let me rephrase you:
> > > | I close tunnels to small local Peers because i want to peer with Large
> > > | Peers around the world.
> > >
> > > Did I understand you correctly?
> > >
> > > If so, i hope all your Peers will decrease the Priority to you to a very
> > > minimum, because this is the best Way to fuck up IPv6-Routing.
> >
> > I use MED, i don't have a bad routing, you can check:
> >
> > http://noc.ndsoftwarenet.com/stats/aspath-tree/bgp-page-complete.php
> >
> > I see all european pTLA/sTLA by European peer,...
> >
> > My MED for tunnels:
> >
> > 500: - 10 ms
> > 510: 10 - 25 ms
> > 520: 25 - 50 ms
> > 530: 50 - 100 ms
> > 540: + 100 ms
> >
> > A lot of pTLA and sTLA don't use MED, i use MED for have a good quality.
>
> Are you aware of the fact that unless you have 'always-compare-med' (or
> equivalent), MED is only used when comparing paths for which the neighbor
> AS is the same? So, unless you have two connections to a neighbor AS,
> it's of no use?
I use:
bgp always-compare-med
bgp dampening
no synchronization
Best Regards,
Nicolas DEFFAYET
From paitken@cisco.com Sat Oct 19 22:20:01 2002
From: paitken@cisco.com (Paul Aitken)
Date: Sat, 19 Oct 2002 22:20:01 +0100
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
References: <5.1.0.14.0.20021016061927.02891008@imap2.es.net> <006e01c27788$de12b220$fd04a80a@alpha> <1035047728.636.1776.camel@wks1.fr.corp.ndsoftware.com> <003501c2779a$42c91ae0$fd04a80a@alpha> <1035055582.636.2069.camel@wks1.fr.corp.ndsoftware.com>
Message-ID: <3DB1CC81.5080404@cisco.com>
Nicolas DEFFAYET wrote:
> I don't have Cisco or Juniper routers because i don't have the budget
> for that. You can offer me a Cisco if you want...
http://www.cisco.com/warp/public/csc/refurb_equipment
--
Paul Aitken
IPv6 Development, Cisco Systems Ltd, Edinburgh, Scotland. EH6 6LX
From jeroen@unfix.org Sat Oct 19 22:28:34 2002
From: jeroen@unfix.org (Jeroen Massar)
Date: Sat, 19 Oct 2002 23:28:34 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <1035061237.636.2158.camel@wks1.fr.corp.ndsoftware.com>
Message-ID: <002f01c277b6$7b68e3a0$210d640a@unfix.org>
Nicolas DEFFAYET [mailto:nicolas.deffayet@ndsoftware.net] wrote:
> On Sat, 2002-10-19 at 22:41, Jeroen Massar wrote:
> > >
> > > I will move the 3 /32 in the pTLA.
> > >
> > > For the moment, i use:
> > _I_ use.
>
> Why do you play with words ?
> It's funny for you ?
No I can't imagine that it's funny having to see somebody apply for a
pTLA
who only has 2 colo'd boxes and calling _himself_ a ISP and an IX
without any justification.
> > > /35 DELEGATION (/48, /44, /40)
> > For what? Llama's? Vhosts? Mirrors?
>
> read my request of pTLA to Bob Fink.
Guess what I did? I read it while ignoring the numberous grammar errors
and needing a de-yoda-fier to actually understand it a bit.
There is nothing in it about where it's used for.
> > > /35 IPV6-FR
> > > /36 NEXTGENCOLLECTIVE
> > > /36 NDSOFTWARE
> > >
> > > A lot of /24 and /28 pTLA use only one /48 in their pTLA...
> > >
> > > I hope that you see that i need a pTLA.
> >
> > No I don't see any reason for a pTLA.
> > I do see a reason for making a numberplan.
> > How many users are in this 3x /32?
> > A very big ISP can provide IPv6 to whole china with it.
> > How many 'users' are there in NDSOFTWARE that you need
> > and can justify a /36 ???
>
> Why you need know my number plan ?
> Other pTLA didn't published their number plan for get their pTLA...
They ARE legit companies who have existed for multiple years.
And they DO have actual REAL users and they also have a REAL plan
for the usage of the pTLA and also a will to make it production after
the experimenting is over.
So I ask you _again_: Can you justify a /36 for a single company?
A /36, how many employees do you have? Or are they all getting a /42?
> > Where did FNIX6 go? Oh yeah it doesn't exist checking the site.
>
> http://www.fnix6.net.net/, this URL is in my request of pTLA
I said I checked _your_ site, not a non-existing URL.
I checked http://www.fnix6.net/members/by-name.php
8<------------------------
FNIX6 Members Index by Name
------------------------>8
Empty -> No Members -> No IX.
Even your _own_ NDSOFTWARE doesn't want to be a member apparently.
Sorry www.fnix6.net.net doesn't exist but you probably didn't register
that domain. And yes your FAQ (http://www.fnix6.net/about/faq.php) says
it's:
8<------------------
9: Who are the members of FNIX6 ?
A full and up to date list is available on the FNIX6 website.
----------------->8
Good, no users there, now where did you need that pTLA for?
Check: www.tunnel-ix.nl at least they say they are tunnel'd.
> > > Where is the "rest" of NDSoftware (Nicolas DEFFAYET Software).
> >
> > I'm the founder of NDSoftware, yes NDSoftware = Nicolas DEFFAYET
> > Software.
> > NDSoftware exist since 03/2000.
> > I use "I" for call NDSoftware because it's my company.
> And nobody else now is there?
> Who are the other people mentioned in _your_ whois ?
>
> It's like you said "my family".
The janitor can't manage a network. Oh oops there is no network?
Or do you have any cabling coming out of that colo box to other places?
No private home adsl lines don't count, especially if they are operated
by other ISP's.
> It's people who work on the IPv6 project with me.
Actual reallife people?
> > > You don't need a pTLA, you need a delegation from NERIM,
> > they are your
> > > upstream.
> >
> > No, i need a pTLA.
>
> > I again. YOU don't need a pTLA.
> > WHY ? Oh I'll spell it again : W.H.Y.
> > Thats: Double-U, Ache, Why: Why.
>
> > First justify WHY _your company_ needs it.
> > Then rememeber that it's for EXPERIMENTAL purposes
> > So you can't use it COMMERCIALLY.
> >
> > And again I ask you, if you reply and _your company_ still
> > want to request a pTLA: answer the above questions in full without
> > cutting.
>
> I need a pTLA, i have sent my request with my justification to Bob
Fink
> and he had sent to this mailing list.
> Read it before ask why why why !
http://mailman.isi.edu/pipermail/6bone/2002-October/006364.html
I read it a couple of times and see no valid data,
the persons apparently are your family, you have no infrastructure,
no planning what you are going to do with it.
And the biggest problem of all: No users.
I also wonder why ATI doesn't get space from APNIC and needs to get
it from a one-person non-existing 'company' from france, which is at
the other side of the world and doesn't have his own line to TUNESIA?
Are you mad? It's the same as Class-A style assignments in IPv4.
> Ask other pTLA why they have a pTLA...
I asked a number of them and they all came up with valid reasons.
But this is about YOUR (or was it NDSoftware's?) request for a pTLA.
Not theirs. And thus it's now YOUR (?NDSoftware?)'s turn to answer
questions.
And why _I_ need to know about this: simply because I don't want to see
everybody requesting and getting a pTLA (and after that an sTLA?).
If you can do it, every person in this world can rectify it and then
we can really start making those addresses 512bits, ohno those same
persons will be wanting a /32 out of the 512bits for their 2 colo'd
boxes.
And that's why I want to see a numberplan.
I also find it remarkable that RIPE gave you a ASN just because you
whined about it.
Too much and's, too much questions unanswered or answered with the same
answers.
Greets,
Jeroen
From nicolas.deffayet@ndsoftware.net Sat Oct 19 22:37:19 2002
From: nicolas.deffayet@ndsoftware.net (Nicolas DEFFAYET)
Date: 19 Oct 2002 23:37:19 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <002f01c277b6$7b68e3a0$210d640a@unfix.org>
References: <002f01c277b6$7b68e3a0$210d640a@unfix.org>
Message-ID: <1035063439.631.2174.camel@wks1.fr.corp.ndsoftware.com>
On Sat, 2002-10-19 at 23:28, Jeroen Massar wrote:
I don't reply anymore to your stupid questions.
From tjc@ecs.soton.ac.uk Sat Oct 19 22:38:05 2002
From: tjc@ecs.soton.ac.uk (Tim Chown)
Date: Sat, 19 Oct 2002 22:38:05 +0100
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <1035059253.610.2135.camel@wks1.fr.corp.ndsoftware.com>
References: <001701c277ab$513eac50$210d640a@unfix.org> <1035059253.610.2135.camel@wks1.fr.corp.ndsoftware.com>
Message-ID: <20021019213804.GB18889@login.ecs.soton.ac.uk>
On Sat, Oct 19, 2002 at 10:27:33PM +0200, Nicolas DEFFAYET wrote:
>
> No, i need a pTLA.
So what exactly are you doing that a typical university that performs IPv6
experiments and research is not doing? A university doesn't get a pTLA,
just a /48. We offer connectivity to end users. We have FTP servers, we
run BGP, etc...
A similar question came up about a Swedish university a while back that
wanted a pTLA to "do BGP" and have a lot of peers. I don't recall the
result of their pTLA application.
That aside, I think the style of presentation of routing info/etc on your
site is good, and more "genuine" providers could usefully make such views
of their routing info public.
Tim
From tjc@ecs.soton.ac.uk Sat Oct 19 22:42:19 2002
From: tjc@ecs.soton.ac.uk (Tim Chown)
Date: Sat, 19 Oct 2002 22:42:19 +0100
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <001f01c277af$dd0aa640$210d640a@unfix.org>
References: <1035059253.610.2135.camel@wks1.fr.corp.ndsoftware.com> <001f01c277af$dd0aa640$210d640a@unfix.org>
Message-ID: <20021019214219.GC18889@login.ecs.soton.ac.uk>
On Sat, Oct 19, 2002 at 10:41:12PM +0200, Jeroen Massar wrote:
>
> How many users are in this 3x /32?
> A very big ISP can provide IPv6 to whole china with it.
I think China needs more than 196,000 /48's :)
Tim
From nicolas.deffayet@ndsoftware.net Sat Oct 19 22:44:45 2002
From: nicolas.deffayet@ndsoftware.net (Nicolas DEFFAYET)
Date: 19 Oct 2002 23:44:45 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <002f01c277b6$7b68e3a0$210d640a@unfix.org>
References: <002f01c277b6$7b68e3a0$210d640a@unfix.org>
Message-ID: <1035063885.610.2182.camel@wks1.fr.corp.ndsoftware.com>
On Sat, 2002-10-19 at 23:28, Jeroen Massar wrote:
> Nicolas DEFFAYET [mailto:nicolas.deffayet@ndsoftware.net] wrote:
>
> > On Sat, 2002-10-19 at 22:41, Jeroen Massar wrote:
> > > >
> > > > I will move the 3 /32 in the pTLA.
> > > >
> > > > For the moment, i use:
> > > _I_ use.
> >
> > Why do you play with words ?
> > It's funny for you ?
>
> No I can't imagine that it's funny having to see somebody apply for a
> pTLA
> who only has 2 colo'd boxes and calling _himself_ a ISP and an IX
> without any justification.
We don't have only "2 colo'd".
See my previous mail...
> They ARE legit companies who have existed for multiple years.
> And they DO have actual REAL users and they also have a REAL plan
> for the usage of the pTLA and also a will to make it production after
> the experimenting is over.
We have REAL users.
> > I need a pTLA, i have sent my request with my justification to Bob
> Fink
> > and he had sent to this mailing list.
> > Read it before ask why why why !
> http://mailman.isi.edu/pipermail/6bone/2002-October/006364.html
>
> I read it a couple of times and see no valid data,
> the persons apparently are your family, you have no infrastructure,
Not of my family, i have an infrasture.
> no planning what you are going to do with it.
> And the biggest problem of all: No users.
WE HAVE USERS.
PLEASE READ MY PTLA REQUEST !!!
> I also wonder why ATI doesn't get space from APNIC and needs to get
> it from a one-person non-existing 'company' from france, which is at
> the other side of the world and doesn't have his own line to TUNESIA?
> Are you mad? It's the same as Class-A style assignments in IPv4.
Visit http://www.ipv6net.tn (URL IN MY PTLA REQUEST !!!)
> Too much and's, too much questions unanswered or answered with the same
> answers.
All your question have a reply in the archive of this list or in my pTLA
request.
I need a pTLA, i justify it.
Why IPNg.nl have a pTLA and a sTLA ?
Please justify the both.
Regards,
Nicolas DEFFAYET
From tjc@ecs.soton.ac.uk Sat Oct 19 22:45:15 2002
From: tjc@ecs.soton.ac.uk (Tim Chown)
Date: Sat, 19 Oct 2002 22:45:15 +0100
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <002701c277b1$12200400$210d640a@unfix.org>
References: <200210192010.g9JKAfgj024322@ludwigV.sources.org> <002701c277b1$12200400$210d640a@unfix.org>
Message-ID: <20021019214514.GD18889@login.ecs.soton.ac.uk>
On Sat, Oct 19, 2002 at 10:49:50PM +0200, Jeroen Massar wrote:
>
> I do want to see the ability for an IXP to get address space that is
> routable. Which is why I mentioned the AMS-IX case, which is a
> "non-profit, neutral and independent association" and thus can't rely on
> other peoples
> IP space. The shared medium can run only on the /48, but the services
> provided (website, lookingglasses etc) can't. AMS-IX has major
> difficulties
> getting that space. Notez bien that even RIPE can't request space from
> thereselves as they are not a LIR. Odd rules in a odd world.
Euro6IX has a pTLA, to allow customer allocation of addresses from its
address space. Euro6IX is not an LIR, nor even a legal entity.
Tim
From tjc@ecs.soton.ac.uk Sat Oct 19 22:46:31 2002
From: tjc@ecs.soton.ac.uk (Tim Chown)
Date: Sat, 19 Oct 2002 22:46:31 +0100
Subject: [6bone] Re: IPv6-only IXP's are absolutely wrong
In-Reply-To: <200210192010.g9JKAfgj024322@ludwigV.sources.org>
References: <1035056975.634.2109.camel@wks1.fr.corp.ndsoftware.com> <200210192010.g9JKAfgj024322@ludwigV.sources.org>
Message-ID: <20021019214631.GE18889@login.ecs.soton.ac.uk>
On Sat, Oct 19, 2002 at 10:10:41PM +0200, Stephane Bortzmeyer wrote:
>
> You can always go through an existing LIR. Gitoyen could certainly do it, even if I find the idea of an IPv6-only IXP absolutely wrong.
Hi Stephane,
I'm curious as to why you think this?
Is the UK6X heading down the wrong path?
Tim
From jeroen@unfix.org Sat Oct 19 22:48:34 2002
From: jeroen@unfix.org (Jeroen Massar)
Date: Sat, 19 Oct 2002 23:48:34 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <1035063439.631.2174.camel@wks1.fr.corp.ndsoftware.com>
Message-ID: <003701c277b9$465d4d10$210d640a@unfix.org>
Nicolas DEFFAYET [mailto:nicolas.deffayet@ndsoftware.net] wrote:
> On Sat, 2002-10-19 at 23:28, Jeroen Massar wrote:
>
> I don't reply anymore to your stupid questions.
Does this also mean that you DROP your pTLA request?
If you are not capable of answering simple questions you should not be
eligible for a pTLA either.
Now finally for some constructive message to Nico:
You are trying and wanting to do constructive things for IPv6.
But you really really have to realize that NOT everybody can have a
pTLA.
This is simply because of the biggest troubles seen in the IPv4 land:
Address depletion and Routing table size.
If one then also adds to that a complete tunneled environment you would
simply blow it all up for everybody and that's not what we want now do
we?
Greets,
Jeroen
From nicolas.deffayet@ndsoftware.net Sat Oct 19 22:55:33 2002
From: nicolas.deffayet@ndsoftware.net (Nicolas DEFFAYET)
Date: 19 Oct 2002 23:55:33 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <003701c277b9$465d4d10$210d640a@unfix.org>
References: <003701c277b9$465d4d10$210d640a@unfix.org>
Message-ID: <1035064533.606.2187.camel@wks1.fr.corp.ndsoftware.com>
On Sat, 2002-10-19 at 23:48, Jeroen Massar wrote:
> Nicolas DEFFAYET [mailto:nicolas.deffayet@ndsoftware.net] wrote:
>
> > On Sat, 2002-10-19 at 23:28, Jeroen Massar wrote:
> >
> > I don't reply anymore to your stupid questions.
>
> Does this also mean that you DROP your pTLA request?
No, I need always a pTLA.
Why you are jealous ?
> If you are not capable of answering simple questions you should not be
> eligible for a pTLA either.
You are funny.
And if i don't want answer to your stupid questions ?
It's not in RFC2772.
> Now finally for some constructive message to Nico:
>
> You are trying and wanting to do constructive things for IPv6.
> But you really really have to realize that NOT everybody can have a
> pTLA.
Why IPNG.NL have a pTLA and a sTLA ?
Please justify this.
Regards,
Nicolas DEFFAYET
From tvo@EnterZone.Net Sat Oct 19 22:58:20 2002
From: tvo@EnterZone.Net (John Fraizer)
Date: Sat, 19 Oct 2002 17:58:20 -0400 (EDT)
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <1035047728.636.1776.camel@wks1.fr.corp.ndsoftware.com>
Message-ID:
On 19 Oct 2002, Nicolas DEFFAYET wrote:
> Why we close peering with private ASN the 23 October 2002:
>
> We have 101 BGP4+ peer, our current routers are full (zebra is very
> unstable if i add new peer) and we want get new peer with other
> pTLA/sTLA that we can't get with our old private ASN.
>
> We have choose to delete all peers with private ASN for free BGP session
> on our routers for this new peers.
>
> > oh! what a coincidence "We will shutdown your BGP peering the 23 October 2002"
> > and "pTLA request NDSOFTWARE - review closes 23 October 2002"
> >
> > 2.4 you finally got your pTLA
> > 2.5 you think private ASN sucks and remove all, peerings with private ASN
>
> We don't delete peers with private ASN because "private ASN sucks", we
> keep peering with important private ASN like NextGenCollective or
> IPNG-UK (this 2 projects projet provide a lot of tunnels to users). I
> understand their status, it's why i keep peering with them.
>
Nicolas,
If your routers are full, BUILD ANOTHER ROUTER to add to your v6 core.
You don't drop one peer to bring on another. Your new "policy" sucks,
plain and simple and for this reason, I can NOT support your request for a
pTLA.
> TDOI don't provide a lot of tunnels (more than 50) to users, it's for
> that reason that we shutdown your peering.
OK. I see a total of 2 native v6 sessions listed in your site
object. You're shutting down existing peering sessions for what reason?
Just an FYI: You violated OUR peering policy when you dropped your peering
session to change your ASN. If you investigate a bit, you'll notice that
it is possible to have Zebra speak as multiple ASNs at once. You simply
changed your ASN and "poof" the session dropped. No biggie. Just thought
I would bring that to your attention.
>
> Where i send you your certificate of the best troll of 6bone
> mailing-list ?
>
And where do I send your certificate for being childish on the 6bone
mailing list? Yet another reason I do NOT support your application for a
pTLA.
---
John Fraizer | High-Security Datacenter Services |
President | Dedicated circuits 64k - 155M OC3 |
EnterZone, Inc | Virtual, Dedicated, Colocation |
http://www.enterzone.net/ | Network Consulting Services |
From jeroen@unfix.org Sat Oct 19 23:10:17 2002
From: jeroen@unfix.org (Jeroen Massar)
Date: Sun, 20 Oct 2002 00:10:17 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <1035063885.610.2182.camel@wks1.fr.corp.ndsoftware.com>
Message-ID: <004101c277bc$4fdcc2a0$210d640a@unfix.org>
Nicolas DEFFAYET [mailto:nicolas.deffayet@ndsoftware.net] wrote:
Oh I thought that you:
I don't reply anymore to your stupid questions.
But okay you still want to request your personal TLA.
> On Sat, 2002-10-19 at 23:28, Jeroen Massar wrote:
> > Nicolas DEFFAYET [mailto:nicolas.deffayet@ndsoftware.net] wrote:
> >
> > > On Sat, 2002-10-19 at 22:41, Jeroen Massar wrote:
> > > > >
> > > > > I will move the 3 /32 in the pTLA.
> > > > >
> > > > > For the moment, i use:
> > > > _I_ use.
> > >
> > > Why do you play with words ?
> > > It's funny for you ?
> >
> > No I can't imagine that it's funny having to see somebody
> apply for a
> > pTLA
> > who only has 2 colo'd boxes and calling _himself_ a ISP and an IX
> > without any justification.
>
> We don't have only "2 colo'd".
>
> See my previous mail...
You got a "full" rack with 2 colo'd boxes?
And where do the members for your IX go?
> > They ARE legit companies who have existed for multiple years.
> > And they DO have actual REAL users and they also have a REAL plan
> > for the usage of the pTLA and also a will to make it
> production after
> > the experimenting is over.
>
> We have REAL users.
First you must define we, is it your IRC channel or is it your ISP you
are talking about or is it your IX?
> > > I need a pTLA, i have sent my request with my justification to Bob
> > Fink
> > > and he had sent to this mailing list.
> > > Read it before ask why why why !
> > http://mailman.isi.edu/pipermail/6bone/2002-October/006364.html
> >
> > I read it a couple of times and see no valid data,
> > the persons apparently are your family, you have no infrastructure,
>
> Not of my family, i have an infrasture.
Where is that infrastructure? Telecity is not yours.
> > no planning what you are going to do with it.
> > And the biggest problem of all: No users.
>
> WE HAVE USERS.
>
> PLEASE READ MY PTLA REQUEST !!!
Define We and why NDSoftware needs that /36?
Oh and you don't need any caps, I and all other people
can read lowercase quite well.
> > I also wonder why ATI doesn't get space from APNIC and needs to get
> > it from a one-person non-existing 'company' from france, which is at
> > the other side of the world and doesn't have his own line
> to TUNESIA?
> > Are you mad? It's the same as Class-A style assignments in IPv4.
>
> Visit http://www.ipv6net.tn (URL IN MY PTLA REQUEST !!!)
Tunesia falls under: APNIC, not RIPE, guess why that is.
1 3ffe:80ee:b3f:1::1 (3ffe:80ee:b3f:1::1) 0.886 ms * 1.309 ms
2 gw-ati.ipv6.tn (3ffe:80ee:b3f::1) 2.46 ms * 1.921 ms
3 defra-03-01.pop.xs26.net (3ffe:80ef:301::) 152.604 ms 149.667 ms
149.901 ms
4 bb1v6-rkp-tu25.muc.ipv6.eurocyber.net (2001:768:e:20::1) 160.374 ms
161.443 ms 159.434 ms
5 rap.ipv6.viagenie.qc.ca (3ffe:b00:c18:1:290:27ff:fe17:fc0f) 310.873
ms 312.562 ms 315.375 ms
6 www.6bone.net (3ffe:b00:c18:1::10) 310.767 ms 321.692 ms 318.758
ms
Hmmm xs26, no "ndsoftware" there.
> > Too much and's, too much questions unanswered or answered
> with the same
> > answers.
>
> All your question have a reply in the archive of this list or
> in my pTLA
> request.
>
> I need a pTLA, i justify it.
>
> Why IPNg.nl have a pTLA and a sTLA ?
> Please justify the both.
Why do I have to justify a pTLA and a sTLA for IPng.nl ?
They are not mine. Also IPng.nl doesn't have a pTLA nor a sTLA.
Greets,
Jeroen
From jeroen@unfix.org Sat Oct 19 23:19:19 2002
From: jeroen@unfix.org (Jeroen Massar)
Date: Sun, 20 Oct 2002 00:19:19 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <1035064533.606.2187.camel@wks1.fr.corp.ndsoftware.com>
Message-ID: <004901c277bd$927d6190$210d640a@unfix.org>
Nicolas DEFFAYET [mailto:nicolas.deffayet@ndsoftware.net] wrote:
> On Sat, 2002-10-19 at 23:48, Jeroen Massar wrote:
> > Nicolas DEFFAYET [mailto:nicolas.deffayet@ndsoftware.net] wrote:
> >
> > > On Sat, 2002-10-19 at 23:28, Jeroen Massar wrote:
> > >
> > > I don't reply anymore to your stupid questions.
> >
> > Does this also mean that you DROP your pTLA request?
>
> No, I need always a pTLA.
Again: explain why.
> Why you are jealous ?
Jealous? I can't be jealous at somebody who requires a pTLA for itself.
I am quite happy with the small assignement I got from my uplink.
Which is quite enough.
> > If you are not capable of answering simple questions you
> should not be
> > eligible for a pTLA either.
>
> You are funny.
I wonder what is so funny about all of this?
> And if i don't want answer to your stupid questions ?
>
> It's not in RFC2772.
Then don't. That's your choice. You requested it.
Also RFC stands for Request for comments, it isn't a STD yet.
> > Now finally for some constructive message to Nico:
> >
> > You are trying and wanting to do constructive things for IPv6.
> > But you really really have to realize that NOT everybody can have a
> > pTLA.
>
> Why IPNG.NL have a pTLA and a sTLA ?
> Please justify this.
As I said before, they don't have any.
Greets,
Jeroen
From hansolofalcon@worldnet.att.net Sat Oct 19 23:24:11 2002
From: hansolofalcon@worldnet.att.net (Gregg C Levine)
Date: Sat, 19 Oct 2002 18:24:11 -0400
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <5.1.0.14.0.20021016061927.02891008@imap2.es.net>
Message-ID: <000001c277be$45fe9900$5c59580c@who>
Hello from Gregg C Levine
I have been lurking on this list, for a good number of years now.
Sometimes even posting a comment, or a gripe. This is more along the
lines of both. I have been monitoring the traffic discussing
NDSOFTWARE's request. Both finding the original message to Bob Fink, and
the list, and everything. Sorry Mr. Deffayet, I disagree. For one, you
do need to spell out who will be using the service. Is it for your
company? Yourself? What? Who, even? Unless you can spell out neatly the
answers to my questions, I am inclined to agree with everyone else. I am
also agreeing with the people I have disagreed with early on. I might
also, add, even Master Yoda's methods of speaking isn't that confusing.
-------------------
Gregg C Levine hansolofalcon@worldnet.att.net
------------------------------------------------------------
"The Force will be with you...Always." Obi-Wan Kenobi
"Use the Force, Luke." Obi-Wan Kenobi
(This company dedicates this E-Mail to General Obi-Wan Kenobi )
(This company dedicates this E-Mail to Master Yoda )
> -----Original Message-----
> From: 6bone-admin@mailman.isi.edu [mailto:6bone-admin@mailman.isi.edu]
On
> Behalf Of Bob Fink
> Sent: Wednesday, October 16, 2002 9:26 AM
> To: 6BONE List
> Cc: Nicolas DEFFAYET
> Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October
2002
>
> 6bone Folk,
>
> NDSOFTWARE has requested a pTLA allocation and I find their request
fully
> compliant with RFC2772. The open review period for this will close 23
> October 2002. Please send your comments to me or the list.
>
>
>
>
>
>
> Thanks,
>
> Bob
>
> =====
> >Hello,
> >
> >On behalf of NDSoftware, I would like to submit our application for a
> >pTLA.
> >
> >Best Regards,
> >
> >Nicolas DEFFAYET
> >
>
>-----------------------------------------------------------------------
-
> >
> > From RFC 2772
> >
> >
> >7. Guidelines for 6Bone pTLA sites
> >
> >
> > The following rules apply to qualify for a 6Bone pTLA allocation.
It
> > should be recognized that holders of 6Bone pTLA allocations are
> > expected to provide production quality backbone network services
for
> > the 6Bone.
> >
> >
> > 1. The pTLA Applicant must have a minimum of three (3) months
> > qualifying experience as a 6Bone end-site or pNLA transit.
> >During
> > the entire qualifying period the Applicant must be
operationally
> > providing the following:
> >
> >Our ipv6-site is operational since 17 january 2001 on 6bone.
> >
> > a. Fully maintained, up to date, 6Bone Registry entries for
their
> > ipv6-site inet6num, mntner, and person objects, including
each
> > tunnel that the Applicant has.
> >
> >http://whois.6bone.net/cgi-bin/whois?NDSOFTWARE
> >
> >
> > b. Fully maintained, and reliable, BGP4+ peering and
connectivity
> > between the Applicant's boundary router and the
appropriate
> > connection point into the 6Bone. This router must be IPv6
> > pingable. This criteria is judged by members of the 6Bone
> > Operations Group at the time of the Applicant's pTLA
request.
> >
> >We have currently 101 BGP4+ sessions.
> >
> >Our ASN is AS25358:
> >aut-num: AS25358
> >as-name: NDSOFTWARE-AS
> >descr: NDSoftware IP Network
> >
> >We use 2 routers:
> > - parcr1.fr.ndsoftwarenet.net
> > - parcr2.fr.ndsoftwarenet.net
> >Looking Glass: http://noc.ndsoftwarenet.com/lg/
> >
> >
> > c. Fully maintained DNS forward (AAAA) and reverse (ip6.int)
> > entries for the Applicant's router(s) and at least one
host
> > system.
> >
> >We have 3 nameservers:
> > - ns1.ndsoftwarenet.com
> > - ns2.ndsoftwarenet.com
> > - ns3.ndsoftwarenet.com
> >
> > d. A fully maintained, and reliable, IPv6-accessible system
> > providing, at a mimimum, one or more web pages, describing
the
> > Applicant's IPv6 services. This server must be IPv6
pingable.
> >
> >http://noc.ndsoftwarenet.com/
> >
> > 2. The pTLA Applicant MUST have the ability and intent to provide
> > "production-quality" 6Bone backbone service. Applicants must
> > provide a statement and information in support of this claim.
> > This MUST include the following:
> >
> >
> > a. A support staff of two persons minimum, three preferable,
with
> > person attributes registered for each in the ipv6-site
object
> > for the pTLA applicant.
> >
> >NDN1-6BONE
> >CB2-6BONE
> >BN3-6BONE
> >MM14-6BONE
> >MC7-6BONE
> >
> > b. A common mailbox for support contact purposes that all
support
> > staff have acess to, pointed to with a notify attribute in
the
> > ipv6-site object for the pTLA Applicant.
> >
> >ipmaster@ndsoftwarenet.com
> >
> > 3. The pTLA Applicant MUST have a potential "user community" that
> > would be served by its becoming a pTLA, e.g., the Applicant
is a
> > major provider of Internet service in a region, country, or
focus
> > of interest. Applicant must provide a statement and
information
> > in support this claim.
> >
> >NDSoftware operates an IPv6 network and provide a lot of IPv6
services
> >to many projects.
> >
> >We provide to:
> >
> >IPv6-FR (a non profit organisation for the developement of IPv6 in
France
> > 200 users, each user have a /48.
> >
> >NexGenCollective (http://www.nexgencollective.net/)
> > 150 users, each user have a /48.
> >
> >ATI (A tunisian ISP, http://www.ipv6net.tn/)
> >
> >and a lot of others (see our whois), this services: IPv6 connectivity
> >(STATIC or BGP with a IPv6 block), IPv6 newsfeeds/newsread,...
> >
> >We do many actions in IPv6 research, we created FNIX6 (French
> >International Internet Exchange IPv6, http://www.fnix6.net/), we host
many
> >mirrors
> >available in IPv6, we created ftp://ftp.openipv6.com/ (a FTP with a
lot
> >of IPv6 stuff).
> >
> > 4. The pTLA Applicant MUST commit to abide by the current 6Bone
> > operational rules and policies as they exist at time of its
> > application, and agree to abide by future 6Bone backbone
> > operational rules and policies as they evolve by consensus of
the
> > 6Bone backbone and user community.
> >
> >
> >We agree to all current and future rules and policies.
> >
> >----
>
> -end
>
> _______________________________________________
> 6bone mailing list
> 6bone@mailman.isi.edu
> http://mailman.isi.edu/mailman/listinfo/6bone
From jeroen@unfix.org Sat Oct 19 23:30:19 2002
From: jeroen@unfix.org (Jeroen Massar)
Date: Sun, 20 Oct 2002 00:30:19 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <20021019214514.GD18889@login.ecs.soton.ac.uk>
Message-ID: <005101c277bf$1bdafaf0$210d640a@unfix.org>
Tim Chown wrote:
> On Sat, Oct 19, 2002 at 10:49:50PM +0200, Jeroen Massar wrote:
> >
> > I do want to see the ability for an IXP to get address space that is
> > routable. Which is why I mentioned the AMS-IX case, which is a
> > "non-profit, neutral and independent association" and thus
> can't rely on
> > other peoples
> > IP space. The shared medium can run only on the /48, but
> the services
> > provided (website, lookingglasses etc) can't. AMS-IX has major
> > difficulties
> > getting that space. Notez bien that even RIPE can't request
> space from
> > thereselves as they are not a LIR. Odd rules in a odd world.
>
> Euro6IX has a pTLA, to allow customer allocation of addresses from its
> address space. Euro6IX is not an LIR, nor even a legal entity.
AMS-IX also has a pTLA:
8<---------------
inet6num: 3FFE:3000::/24
netname: AMS-IX
descr: pTLA delegation for the 6bone
--------------->8
But a pTLA is experimental, IX's are (hopefully :) productional,
which is why they need/want to have sTLA's.
Also note that looking at:
http://www.euro6ix.org/ingles/partners/partners.php
makes Euro6IX much more a legal entity than anything "NDSoftware" is
doing.
Greets,
Jeroen
From fink@es.net Sat Oct 19 23:32:01 2002
From: fink@es.net (Bob Fink)
Date: Sat, 19 Oct 2002 15:32:01 -0700
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October
2002
In-Reply-To: <004101c277bc$4fdcc2a0$210d640a@unfix.org>
References: <1035063885.610.2182.camel@wks1.fr.corp.ndsoftware.com>
Message-ID: <5.1.0.14.0.20021019152934.028d9068@imap2.es.net>
6bone Folk,
Please keep the discussion from getting personal, or becoming defamatory or
using swear words.
Many folks watch how we carry out our 6bone business; it is important that
we remain open, objective and willing to hear all sides.
Thanks,
Bob
From nicolas.deffayet@ndsoftware.net Sat Oct 19 23:43:47 2002
From: nicolas.deffayet@ndsoftware.net (Nicolas DEFFAYET)
Date: 20 Oct 2002 00:43:47 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To:
References:
Message-ID: <1035067427.610.2230.camel@wks1.fr.corp.ndsoftware.com>
On Sat, 2002-10-19 at 23:58, John Fraizer wrote:
John,
> If your routers are full, BUILD ANOTHER ROUTER to add to your v6 core.
> You don't drop one peer to bring on another. Your new "policy" sucks,
> plain and simple and for this reason, I can NOT support your request for a
> pTLA.
We don't have budget for build anoter 6bone router.
We want start a pre-production network, all new routers will be for
pre-production network.
Yes, our new policy sucks but i don't have better solution.
I know that my email to this peers wasn't diplomatic.
You provide to an user a BGP4+ peering, after you can't keep the tunnel
for technical reasons (routers full), do you will like that this user
troll about you on public mailing-list ?
One of users who have received my email, contact me (no on public
mailing-list), for get many informations about this problem.
He reply to me "I understand your problem, good luck for your IPv6
network.". If we have free bgp session on our routers later, i will
recontact him.
It's a normal reply. Christian, don't contact me but troll on public
mailing-listing about me, it's not a normal reply, it's a revenge.
> Just an FYI: You violated OUR peering policy when you dropped your peering
> session to change your ASN. If you investigate a bit, you'll notice that
> it is possible to have Zebra speak as multiple ASNs at once. You simply
> changed your ASN and "poof" the session dropped. No biggie. Just thought
> I would bring that to your attention.
With Zebra 0.93b multi-as on the same router don't work fine.
I have send to you a email, i wait 24 hours, no reply, reply of other, i
switch the ASN of the router.
Best Regards,
Nicolas DEFFAYET
From nicolas.deffayet@ndsoftware.net Sat Oct 19 23:45:41 2002
From: nicolas.deffayet@ndsoftware.net (Nicolas DEFFAYET)
Date: 20 Oct 2002 00:45:41 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <20021019213804.GB18889@login.ecs.soton.ac.uk>
References: <001701c277ab$513eac50$210d640a@unfix.org>
<1035059253.610.2135.camel@wks1.fr.corp.ndsoftware.com>
<20021019213804.GB18889@login.ecs.soton.ac.uk>
Message-ID: <1035067541.610.2233.camel@wks1.fr.corp.ndsoftware.com>
On Sat, 2002-10-19 at 23:38, Tim Chown wrote:
> A similar question came up about a Swedish university a while back that
> wanted a pTLA to "do BGP" and have a lot of peers. I don't recall the
> result of their pTLA application.
pTLA allocated: http://whois.6bone.net/cgi-bin/whois?SSVL
Best Regards,
Nicolas DEFFAYET
From solaris@swissirc.ch Sat Oct 19 23:51:23 2002
From: solaris@swissirc.ch (Marcel Stutz)
Date: Sun, 20 Oct 2002 00:51:23 +0200
Subject: [6bone] How i get IP V6 Addresses ?
Message-ID:
What is the best way to get own IP V6 Addresses for a smale IRC Network in
switzerland ?
What i need to pay ?
Thanks Marcel
From solaris@swissirc.ch Sat Oct 19 23:51:23 2002
From: solaris@swissirc.ch (Marcel Stutz)
Date: Sun, 20 Oct 2002 00:51:23 +0200
Subject: [6bone] How i get IP V6 Addresses ?
Message-ID:
What is the best way to get own IP V6 Addresses for a smale IRC Network in
switzerland ?
What i need to pay ?
Thanks Marcel
From pekkas@netcore.fi Sun Oct 20 00:22:48 2002
From: pekkas@netcore.fi (Pekka Savola)
Date: Sun, 20 Oct 2002 02:22:48 +0300 (EEST)
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <3DB1CC81.5080404@cisco.com>
Message-ID:
On Sat, 19 Oct 2002, Paul Aitken wrote:
> Nicolas DEFFAYET wrote:
>
> > I don't have Cisco or Juniper routers because i don't have the budget
> > for that. You can offer me a Cisco if you want...
>
> http://www.cisco.com/warp/public/csc/refurb_equipment
Well, to be frank, I'm not sure why anyone would want Cisco equipment for
IPv6, old or new. They hardly seem to be able to manage 30 Mbit/s of IPv6
traffic :-(. I guess this is enough for some, for us it isn't :-(.
And no, we're not using "crap" (for some, usefull stuf for others) like
4x00's, 2x00's etc. like many seem to be doing: rather, like 7200, 7500,
12xxx, etc.
And yes, I've tried to contact ipv6-support@cisco.com to try to find out
whether these are really the bottlenecks, no replies.
--
Pekka Savola "Tell me of difficulties surmounted,
Netcore Oy not those you stumble over and fall"
Systems. Networks. Security. -- Robert Jordan: A Crown of Swords
From pekkas@netcore.fi Sun Oct 20 00:25:14 2002
From: pekkas@netcore.fi (Pekka Savola)
Date: Sun, 20 Oct 2002 02:25:14 +0300 (EEST)
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To:
Message-ID:
On Sat, 19 Oct 2002, John Fraizer wrote:
[...]
> Yet another reason I do NOT support your application for a
> pTLA.
Past experience seems to have shown that unless it's in RFC2772, support
for or against have little impact.
I suggest one should search for counter-arguments there if you want to
block the pTLA allocation.
--
Pekka Savola "Tell me of difficulties surmounted,
Netcore Oy not those you stumble over and fall"
Systems. Networks. Security. -- Robert Jordan: A Crown of Swords
From infox@infox.ath.cx Sun Oct 20 01:11:12 2002
From: infox@infox.ath.cx (Matt Prall)
Date: Sat, 19 Oct 2002 17:11:12 -0700
Subject: [6bone] How i get IP V6 Addresses ?
References:
Message-ID: <3DB1F4A0.7020602@infox.ath.cx>
It is free, I would recommend you start here:
http://www.6bone.net/6bone_hookup.html
-Matt
Marcel Stutz wrote:
>What is the best way to get own IP V6 Addresses for a smale IRC Network in
>switzerland ?
>
>What i need to pay ?
>
>Thanks Marcel
>
>
>
>
>_______________________________________________
>6bone mailing list
>6bone@mailman.isi.edu
>http://mailman.isi.edu/mailman/listinfo/6bone
>
>
>
>
From pekkas@netcore.fi Sun Oct 20 01:51:22 2002
From: pekkas@netcore.fi (Pekka Savola)
Date: Sun, 20 Oct 2002 03:51:22 +0300 (EEST)
Subject: [6bone] very drafty draft on 6bone routing mess
Message-ID:
Hello,
The so-called 6bone-mess has been discussed here back and forth, with no
apparent result or success.
Based on experiences gained in 6NET and seeing how others are doing, I got
motivated enough to write something down on the subject.
A very drafty draft (result of 3 hours of torturing the keyboard in middle
of the night :-) on 6bone routing policy issues which I believe are
causing current problems is available at:
http://www.netcore.fi/pekkas/ietf/6bone-mess.txt
There are also some ideas, but nothing specific, how one could get around
those.
Before I want to go make this something a bit less drafty I'd like to get
opinions and thoughts on this: should we try to do something to try to
ensure IPv6 Internet would actually get usable one of these years (it sure
ain't now!)
Comments, please.
(Anyone with more than 5 peers with transit should feel the sting of guilt
now. :-)
--
Pekka Savola "Tell me of difficulties surmounted,
Netcore Oy not those you stumble over and fall"
Systems. Networks. Security. -- Robert Jordan: A Crown of Swords
From tvo@EnterZone.Net Sun Oct 20 09:42:20 2002
From: tvo@EnterZone.Net (John Fraizer)
Date: Sun, 20 Oct 2002 04:42:20 -0400 (EDT)
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <1035067427.610.2230.camel@wks1.fr.corp.ndsoftware.com>
Message-ID:
On 20 Oct 2002, Nicolas DEFFAYET wrote:
> On Sat, 2002-10-19 at 23:58, John Fraizer wrote:
>
> John,
>
> > If your routers are full, BUILD ANOTHER ROUTER to add to your v6 core.
> > You don't drop one peer to bring on another. Your new "policy" sucks,
> > plain and simple and for this reason, I can NOT support your request for a
> > pTLA.
>
> We don't have budget for build anoter 6bone router.
> We want start a pre-production network, all new routers will be for
> pre-production network.
>
OK. What keeps you from putting 6bone peers and your
"pre-production" peers on the same router? Personal choice. There is no
technical barrior to doing so. You are self imposing the barrior based on
policy.
> Yes, our new policy sucks but i don't have better solution.
>
> I know that my email to this peers wasn't diplomatic.
>
> You provide to an user a BGP4+ peering, after you can't keep the tunnel
> for technical reasons (routers full), do you will like that this user
> troll about you on public mailing-list ?
First of all, trolling is the act of LOOKING for something. I don't think
that anyone has been LOOKING for anything from you besides answers to
questions to which you have been highly evasive. Second of all, if you
can't take the heat of having a user talk about you in a public forum, you
had might as well drop out and cut your losses. If you ever run a
_real_ network, it's going to happen. I don't care how perfect your
network performs. Some customer somewhere is going to experience a
problem, be it theirs, yours or something 15 as-hops away from you and
they're going to blame YOU for it in a very loud way on a public
forum. Childish replies such as those you've presented the list with
thusfar only serve to prove their point and do NOTHING to help convince
myself or others that you are a good candidate for a peer, let alone one
with a pTLA.
>
> One of users who have received my email, contact me (no on public
> mailing-list), for get many informations about this problem.
> He reply to me "I understand your problem, good luck for your IPv6
> network.". If we have free bgp session on our routers later, i will
> recontact him.
> It's a normal reply. Christian, don't contact me but troll on public
> mailing-listing about me, it's not a normal reply, it's a revenge.
My, you're full of yourself, aren't you? Do you for some reason think
that the person in question can't get a peering session with any number of
other 6bone participants? From what I saw, he was venting about the WAY
you did it rather than the fact that you did it.
>
> > Just an FYI: You violated OUR peering policy when you dropped your peering
> > session to change your ASN. If you investigate a bit, you'll notice that
> > it is possible to have Zebra speak as multiple ASNs at once. You simply
> > changed your ASN and "poof" the session dropped. No biggie. Just thought
> > I would bring that to your attention.
>
> With Zebra 0.93b multi-as on the same router don't work fine.
>
Strange. I haven't seen anyone make a bug report to the Zebra list
regarding this problem. I have also personally used this feature in our
testbed without issue. If you're having issues with the _FREE_ routing
software that you're using, don't you suppose that you have an obligation
to at least post to the list regarding the issue so that those who work on
the code can actually try to diagnose and fix the problem?
> I have send to you a email, i wait 24 hours, no reply, reply of other, i
> switch the ASN of the router.
>
The only email I received from you cam AFTER you had changed your ASN on
your bgp config. I am fairly certain that there was no issue with our
mailserver(s) as my standard 200-300 emails per day have been constant for
several years now.
Now. let me pose some questions to you. Your open, honest answers to
these questions will determine my vote (and perhaps many others) on your
pTLA application:
(1) Who are the following and what are their qualifications to be
technical contacts? The last thing I want to hear on the other end of the
phone line when I contact someones technical contacts is "I'm sorry.
He's at school. I'll tell him you called." Do these people all have
enable on your routers? Do they understand v6 routing? Would they know
what I was talking about if I told them that you were leaking leaking
routes or that your peering session was flapping?
person: Chris BURTON
address: NDSoftware
address: 57 rue du president Wilson
address: 92300 Levallois-Perret
address: France
phone: +33 671887502
e-mail: chris.burton@ndsoftware.net
nic-hdl: CB2-6BONE
notify: notify@ndsoftwarenet.com
mnt-by: MNT-NDSOFTWARE
changed: chris.burton@ndsoftware.net 20021015
source: 6BONE
person: Bruno NASH
address: NDSoftware
address: 57 rue du president Wilson
address: 92300 Levallois-Perret
address: France
phone: +33 671887502
e-mail: bruno.nash@ndsoftware.net
nic-hdl: BN3-6BONE
notify: notify@ndsoftwarenet.com
mnt-by: MNT-NDSOFTWARE
changed: bruno.nash@ndsoftware.net 20021015
source: 6BONE
person: Myriam MOREL
address: NDSoftware
address: 57 rue du president Wilson
address: 92300 Levallois-Perret
address: France
phone: +33 671887502
e-mail: myriam.morel@ndsoftware.net
nic-hdl: MM14-6BONE
notify: notify@ndsoftwarenet.com
mnt-by: MNT-NDSOFTWARE
changed: myriam.morel@ndsoftware.net 20021015
source: 6BONE
person: Mike CHENEY
address: NDSoftware
address: 57 rue du president Wilson
address: 92300 Levallois-Perret
address: France
phone: +33 671887502
e-mail: mike.cheney@ndsoftware.net
nic-hdl: MC7-6BONE
notify: notify@ndsoftwarenet.com
mnt-by: MNT-NDSOFTWARE
changed: mike.cheney@ndsoftware.net 20021015
source: 6BONE
(2) Do you have a network plan? IE; How are your current /32's
dispersed?
(3) Do _you_ have a network or are you simply colocated someplace on
someone elses network?
If you're colocated, do you #1 have 24hr _physical_ access to the
equipment? Can you be onsite within a reasonable amount of time in the
event that physical access to equipment is required to remedy a
problem? If not, do you have a "remote hands" contract in place?
This is a serious pet peeve with me. There are far too many
"posers" claiming to have networks and datacenters that in truth have a
19" x 19" space in someone elses datacenter.
Looking a bit at v4 address registries, I see the following:
parcr1.fr.ndsoftwarenet.net (213.91.4.3)
inetnum: 213.91.4.0 - 213.91.4.127
netname: FR-TEASER
descr: FIRSTREAM.NET hosted servers
country: FR
admin-c: HRA81-RIPE
tech-c: HRA81-RIPE
rev-srv: ns0.teaser.net
rev-srv: ns1.teaser.net
status: ASSIGNED PA
remarks: Please report all abuse related issues to
remarks: abuse@firstream.net
notify: hostmaster@teaser.net
mnt-by: FT-NOC
changed: at@teaser.net 20011220
source: RIPE
parcr2.fr.ndsoftwarenet.net (62.4.18.114)
inetnum: 62.4.16.0 - 62.4.20.255
netname: NERIM-1
descr: Nerim is an Internet Service Provider
descr: based in France
country: FR
admin-c: RB7192-RIPE
tech-c: RB7192-RIPE
rev-srv: maridia.nerim.net
rev-srv: metroid.nerim.net
rev-srv: noemie.nerim.net
status: ASSIGNED PA
notify: ripe@isdnet.net
mnt-by: ISDNET-NOC
changed: dly@isdnet.net 19991201
changed: dly@isdnet.net 20001012
changed: kbrebion@isdnet.net 20001018
source: RIPE
(4) If you don't have your own network, how do you propose to provide
"production quality" 6bone backbone services?
I submit that without your own portable v4 address space for an endpoint
of tunnels, you're at the mercy of your upstreams. If they require you to
renumber, every one of your peers will have to reconfigure their tunnels.
(5) I find this strange. Can you explain it?
$ traceroute6 parcr1.fr.ndsoftwarenet.net
traceroute to parcr1.fr.ndsoftwarenet.net (3ffe:81f1:2:1::1) from 3ffe:4010:ff09::1, 30 hops max, 16 byte packets
1 enterzone-ndsoftware-gw.paris.ipv6.enterzone.net (3ffe:4010:ff09::2) 421.357 ms 438.313 ms 419.447 ms
2 bb1v6-sgr-tu0.muc.ipv6.eurocyber.net (2001:768:e:12::2) 492.543 ms 555.586 ms 530.186 ms
3 bb1v6-rkp-tu7.muc.ipv6.eurocyber.net (2001:768:e:14::1) 531.863 ms 603.233 ms 606.157 ms
4 feth0-1-parcr1.fr.ndsoftwarenet.net (3ffe:81f1:0:1::1) 1030.55 ms 1141.35 ms 1132.64 ms
5 bb1v6-rkp-tu7.muc.ipv6.eurocyber.net (2001:768:e:14::1) 724.858 ms 772.424 ms 725.722 ms
6 feth0-1-parcr1.fr.ndsoftwarenet.net (3ffe:81f1:0:1::1) 1212.88 ms 1316.88 ms 1484.25 ms
7 bb1v6-rkp-tu7.muc.ipv6.eurocyber.net (2001:768:e:14::1) 881.822 ms 924.786 ms 892.842 ms
Nice routing loop there. Have you considered: (1) Not having a v6 default on your border
router. (2) Having a connection between your two border routers and running an IGP between them?
(6) It can't be a good sign for a "production quality" network when your
route-server can't maintain a BGP peering session with your own routers:
route-server.ndsoftwarenet.net> sh ipv6 bgp sum
BGP router identifier 10.0.1.2, local AS number 25358
246 BGP AS-PATH entries
9 BGP community entries
Neighbor V AS MsgRcvd MsgSent TblVer InQ OutQ Up/Down State/PfxRcd
3ffe:81f1:0:1::1
4 25358 0 2139 0 0 0 00:04:12 OpenSent
3ffe:81f1:0:2::1
4 25358 13207 5127 0 0 0 3d08h30m 310
Total number of neighbors 2
(7) "a. Fully maintained, up to date, 6Bone Registry entries for their
ipv6-site inet6num, mntner, and person objects, including each
tunnel that the Applicant has."
You've got parcr3.fr.ndsoftwarenet.net listed in your ipv6-site object but:
$ traceroute parcr3.fr.ndsoftwarenet.net
traceroute: unknown host parcr3.fr.ndsoftwarenet.net
$ traceroute6 parcr3.fr.ndsoftwarenet.net
traceroute: unknown host parcr3.fr.ndsoftwarenet.net
(8) With regards to #7 above, I suggest that with your recent policy
change regarding BGP peers, you remove the following line from your
ipb6-site object:
remarks: NDSoftware have an open peering policy.
(9) What is your "potential user community" IE; What gap are you going to
be filling in the service delivery arena that is not already served by
other pTLAs?
(10) What purpose will having your OWN pTLA serve that your current 3
/32's don't already serve? Keep in mind that _wanting_ your own pTLA !=
_NEEDING_ your own pTLA and _NEEDING_ to announce a pTLA into the DFZ
because it's a requirement for you to have your own ASN is _not_
sufficient justification for you to be issued a pTLA.
(11) "d. A fully maintained, and reliable, IPv6-accessible system
providing, at a mimimum, one or more web pages, describing the
Applicant's IPv6 services. This server must be IPv6 pingable."
Looking at http://noc.ndsoftwarenet.com, information about what
NDSOFTWARE actually *does* is strangely absent. Your peering-policy link
returns a 404 error. Your route-filtering link returns a 404 error. Your
usenet-policy link returns a 404 error. Register, Login and Help all
point to your bgp-communities page, as do your "go" button and the
advanced-search link. Home, Products & Services, Support, Download, Buy
and Contact links at the top page simply link the whatever page you're
currently viewing. There is no information about what your
"company?" actually does or offers to do even.
(12) "b. Fully maintained, and reliable, BGP4+ peering and connectivity
between the Applicant's boundary router and the appropriate
connection point into the 6Bone. This router must be IPv6
pingable. This criteria is judged by members of the 6Bone
Operations Group at the time of the Applicant's pTLA request.
We have currently 101 BGP4+ sessions."
Looking at parcr1.fr.ndsoftwarenet.net, I count 87 neighbors, 15 of which
are down, 3 of which have never established an adjacency, two of the 87
peering sessions being yourself. (84 ?real? sessions on this router.)
Looking at parcr2.fr.ndsoftwarenet.net, I count 11 neighbors, 4 of which
are down with two of the 11 peering sessions being yourself. (9
?real? peering sessions.)
I don't know about in France but, in the US, 84 + 9 = 93 peering sessions,
not 101 peering sessions.
Can you perhaps explain your math to us?
(13) Of those 84 peering sessions, have you verified that they have
appropriate entries in their ipv6-site objects for the tunnel/connection
or that they have ipv6-site objects AT-ALL? Before you answer this, take
a look at this:
IPv6 in IPv4 parcr1.fr.ndsoftwarenet.net -> 213.121.24.91 UK6X BGP4+
*** UK6X: Destination site does not appear in registry - check spelling
against registry and remove any extra text
IPv6 in IPv4 parcr1.fr.ndsoftwarenet.net -> ge3-1.er1a.fra2.de.mfnx.net
DE-TRMD-SBI-1 BGP4+
*** DE-TRMD-SBI-1: Destination site does not appear in registry - check
spelling against registry and remove any extra text
IPv6 in IPv4 parcr1.fr.ndsoftwarenet.net ->
modemcable049.63-201-24.mtl.mc.videotron.ca SYNCROS BGP4+
*** SYNCROS: Destination site does not appear in registry - check spelling
against registry and remove any extra text
IPv6 in IPv4 parcr1.fr.ndsoftwarenet.net -> worldakt.com WORLDAKT BGP4+
*** WORLDAKT: Destination site does not appear in registry - check
spelling against registry and remove any extra text
IPv6 in IPv4 parcr1.fr.ndsoftwarenet.net ->
user-212-88-249-12.tvcablenet.be WOOF BGP4+
*** WOOF: Destination site does not appear in registry - check spelling
against registry and remove any extra text
IPv6 in IPv4 parcr1.fr.ndsoftwarenet.net -> tms.dicp.de DICP BGP4+
*** DICP: Destination site does not appear in registry - check spelling
against registry and remove any extra text
IPv6 in IPv4 parcr1.fr.ndsoftwarenet.net -> 194.139.3.28 DE-TRMD BGP4+
*** DE-TRMD: Destination site does not appear in registry - check spelling
against registry and remove any extra text
IPv6 in IPv4 parcr1.fr.ndsoftwarenet.net -> pc65.bydgoszcz.sdi.tpnet.pl
MUSIALEK STATIC
*** MUSIALEK: Destination site does not appear in registry - check
spelling against registry and remove any extra text
IPv6 in IPv4 parcr1.fr.ndsoftwarenet.net -> thunderbird.e-concepts.be
ERALY STATIC
*** ERALY: Destination site does not appear in registry - check spelling
against registry and remove any extra text
IPv6 in IPv4 parcr1.fr.ndsoftwarenet.net -> mirror.seabone.net CACI BGP4+
*** CACI: Destination site does not appear in registry - check spelling
against registry and remove any extra text
IPv6 in IPv4 parcr1.fr.ndsoftwarenet.net ->
port5.ds1-sby.adsl.cybercity.dk FABBIONE BGP4+
*** FABBIONE: Destination site does not appear in registry - check
spelling against registry and remove any extra text
IPv6 in IPv4 parcr1.fr.ndsoftwarenet.net -> 212.81.112.99 DARKSNOW BGP4+
*** DARKSNOW: Destination site does not appear in registry - check
spelling against registry and remove any extra text
IPv6 in IPv4 parcr1.fr.ndsoftwarenet.net -> ipv4.nerime.net NERIME BGP4+
*** NERIME: Destination site does not appear in registry - check spelling
against registry and remove any extra text
IPv6 in IPv4 parcr1.fr.ndsoftwarenet.net ->
pa76.miedzyrzec-podlaski.sdi.tpnet.pl EXORCIST STATIC
*** EXORCIST: Destination site does not appear in registry - check
spelling against registry and remove any extra text
IPv6 in IPv4 parcr1.fr.ndsoftwarenet.net -> quint.netisland.net TOREN
STATIC
*** TOREN: Destination site does not appear in registry - check spelling
against registry and remove any extra text
IPv6 in IPv4 parcr1.fr.ndsoftwarenet.net -> chabrowa.net CHABROWA STATIC
*** CHABROWA: Destination site does not appear in registry - check
spelling against registry and remove any extra text
IPv6 in IPv4 parcr1.fr.ndsoftwarenet.net -> pa1.wroclaw.sdi.tpnet.pl
ANDRE-WRO STATIC
*** ANDRE-WRO: Destination site does not appear in registry - check
spelling against registry and remove any extra text
IPv6 in IPv4 parcr1.fr.ndsoftwarenet.net -> b1rtr.cyf-kr.edu.pl CYFRONET
BGP4+
*** CYFRONET: Destination site does not appear in registry - check
spelling against registry and remove any extra text
IPv6 in IPv4 parcr1.fr.ndsoftwarenet.net -> 212.49.128.151 BTIGNITE BGP4+
*** BTIGNITE: Destination site does not appear in registry - check
spelling against registry and remove any extra text
IPv6 in IPv4 parcr1.fr.ndsoftwarenet.net -> br2.den1.amisp.net AISP BGP4+
*** AISP: Destination site does not appear in registry - check spelling
against registry and remove any extra text
IPv6 in IPv4 parcr1.fr.ndsoftwarenet.net -> ipv6.asol.com.ph ASPI-PH BGP4+
*** ASPI-PH: Destination site does not appear in registry - check spelling
against registry and remove any extra text
IPv6 in IPv4 parcr1.fr.ndsoftwarenet.net -> ns1.labatut.net LABATUT BGP4+
*** LABATUT: Destination site does not appear in registry - check spelling
against registry and remove any extra text
Part of properly maintaining _YOUR_ ipv6-site object is making sure that
you don't reference an object that doesn't exist. If someone is unable or
unwilling to create & maintain an ipv6-site object, do you really feel
that they are a good peering candidate? I certainly don't.
Last but not least:
(13) Just for my own personal amusement... You have a VPI/VCI pair field
in your list of public peering points that you participate in or plan to
participate in on your website but, your interconnects are all listed as
100M FE. Um, what kind of ethernet are you using that supports VPI/VCI or
did you just think it would look "cool" to have that field?
Nicolas, I spent quite a bit of time and effort composing this email and I
feel that I have asked valid questions of you. Please answer them
concisely and accurately.
---
John Fraizer | High-Security Datacenter Services |
President | Dedicated circuits 64k - 155M OC3 |
EnterZone, Inc | Virtual, Dedicated, Colocation |
http://www.enterzone.net/ | Network Consulting Services |
From pgrosset@cisco.com Sun Oct 20 11:39:39 2002
From: pgrosset@cisco.com (Patrick Grossetete)
Date: Sun, 20 Oct 2002 12:39:39 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October
2002
In-Reply-To:
References: <3DB1CC81.5080404@cisco.com>
Message-ID: <4.3.2.7.2.20021020123520.01ad1a58@europe.cisco.com>
Pekka,
Your comment on the performances is not accurate. You reported a
26 Mb/s throughput using TTCP between
2 Linux hosts, asking for feedback about that. We just tested the same
configuration but using IXIA traffic generators
and got different numbers really different from yours. I asked Theo to
provide you an official answer, so I expect
you will update the list later.
Patrick
At 02:22 AM 20-10-02 +0300, Pekka Savola wrote:
>On Sat, 19 Oct 2002, Paul Aitken wrote:
> > Nicolas DEFFAYET wrote:
> >
> > > I don't have Cisco or Juniper routers because i don't have the budget
> > > for that. You can offer me a Cisco if you want...
> >
> > http://www.cisco.com/warp/public/csc/refurb_equipment
>
>Well, to be frank, I'm not sure why anyone would want Cisco equipment for
>IPv6, old or new. They hardly seem to be able to manage 30 Mbit/s of IPv6
>traffic :-(. I guess this is enough for some, for us it isn't :-(.
>
>And no, we're not using "crap" (for some, usefull stuf for others) like
>4x00's, 2x00's etc. like many seem to be doing: rather, like 7200, 7500,
>12xxx, etc.
>
>And yes, I've tried to contact ipv6-support@cisco.com to try to find out
>whether these are really the bottlenecks, no replies.
>
>--
>Pekka Savola "Tell me of difficulties surmounted,
>Netcore Oy not those you stumble over and fall"
>Systems. Networks. Security. -- Robert Jordan: A Crown of Swords
>
>
>_______________________________________________
>6bone mailing list
>6bone@mailman.isi.edu
>http://mailman.isi.edu/mailman/listinfo/6bone
____________________________________________
Patrick Grossetete
Cisco Systems
Internet Technology Division (ITD) - Product Manager
Phone/Vmail: 33.1.58.04.61.52
Fax: 33.1.58.04.61.00
mobile: 33.6.19.98.51.31
Email:pgrosset@cisco.com
11 Rue Camille Desmoulins
92782 Issy les Moulineaux Cedex 9
France
____________________________________________
From gert@space.net Sun Oct 20 11:59:03 2002
From: gert@space.net (Gert Doering)
Date: Sun, 20 Oct 2002 12:59:03 +0200
Subject: [6bone] How i get IP V6 Addresses ?
In-Reply-To: ; from solaris@swissirc.ch on Sun, Oct 20, 2002 at 12:51:23AM +0200
References:
Message-ID: <20021020125903.Q94537@Space.Net>
hi,
On Sun, Oct 20, 2002 at 12:51:23AM +0200, Marcel Stutz wrote:
> What is the best way to get own IP V6 Addresses for a smale IRC Network in
> switzerland ?
>
> What i need to pay ?
There are some ISPs in switzerland that already offer IPv6 services. You
might want to consider just hooking up to one of them, and get addresse
from this upstream ISP.
Unless you're connected to two different ISPs (or plan to do that very
soon), it's not a good idea to get your own address block and announce
that to the whole world. Only as a last resort if you can't find a
suitable upstream ISP.
Gert Doering
-- NetMaster
--
Total number of prefixes smaller than registry allocations: 48282 (47686)
SpaceNet AG Mail: netmaster@Space.Net
Joseph-Dollinger-Bogen 14 Tel : +49-89-32356-0
80807 Muenchen Fax : +49-89-32356-299
From gert@space.net Sun Oct 20 12:01:43 2002
From: gert@space.net (Gert Doering)
Date: Sun, 20 Oct 2002 13:01:43 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <006e01c27788$de12b220$fd04a80a@alpha>; from dragon@tdoi.org on Sat, Oct 19, 2002 at 06:02:03PM +0200
References: <5.1.0.14.0.20021016061927.02891008@imap2.es.net> <006e01c27788$de12b220$fd04a80a@alpha>
Message-ID: <20021020130143.R94537@Space.Net>
Hi,
On Sat, Oct 19, 2002 at 06:02:03PM +0200, Christian Nickel wrote:
> the annoyed german ipv6 community
I just want to mention that I consider me a part of the "german IPv6
community", and I'm not annoyed.
I refused to peer with private ASNs from the very beginning, and I can
only strongly recommend this to everyone else (if you insist on peering,
at least don't give them transit). Private ASNs *hurt*.
Gert Doering
-- NetMaster
--
Total number of prefixes smaller than registry allocations: 48282 (47686)
SpaceNet AG Mail: netmaster@Space.Net
Joseph-Dollinger-Bogen 14 Tel : +49-89-32356-0
80807 Muenchen Fax : +49-89-32356-299
From gert@space.net Sun Oct 20 12:13:13 2002
From: gert@space.net (Gert Doering)
Date: Sun, 20 Oct 2002 13:13:13 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <002701c277b1$12200400$210d640a@unfix.org>; from jeroen@unfix.org on Sat, Oct 19, 2002 at 10:49:50PM +0200
References: <200210192010.g9JKAfgj024322@ludwigV.sources.org> <002701c277b1$12200400$210d640a@unfix.org>
Message-ID: <20021020131313.S94537@Space.Net>
Hi,
On Sat, Oct 19, 2002 at 10:49:50PM +0200, Jeroen Massar wrote:
> getting that space. Notez bien that even RIPE can't request space from
> thereselves as they are not a LIR. Odd rules in a odd world.
The rules are not that odd.
The RIPE NCC network is not "special" in any way requiring their own /32.
If we have rules that deny "normal companies" their /32 and urge them
to go to their upstreams, this rule has to be applied to all (!)
end sites equally. From a network perspective, the RIPE NCC is an
*end site*. They are not an ISP, they don't do *LIR* functions, so
they don't qualify for an sTLA - and technicallywise, they don't *need*
one either.
Gert Doering
-- NetMaster
--
Total number of prefixes smaller than registry allocations: 48282 (47686)
SpaceNet AG Mail: netmaster@Space.Net
Joseph-Dollinger-Bogen 14 Tel : +49-89-32356-0
80807 Muenchen Fax : +49-89-32356-299
From nicolas.deffayet@ndsoftware.net Sun Oct 20 12:52:17 2002
From: nicolas.deffayet@ndsoftware.net (Nicolas DEFFAYET)
Date: 20 Oct 2002 13:52:17 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To:
References:
Message-ID: <1035114737.4777.6.camel@wks1.fr.corp.ndsoftware.com>
On Sun, 2002-10-20 at 10:42, John Fraizer wrote:
> Now. let me pose some questions to you. Your open, honest answers to
> these questions will determine my vote (and perhaps many others) on your
> pTLA application:
>
> (1) Who are the following and what are their qualifications to be
> technical contacts? The last thing I want to hear on the other end of the
> phone line when I contact someones technical contacts is "I'm sorry.
> He's at school. I'll tell him you called." Do these people all have
> enable on your routers? Do they understand v6 routing? Would they know
> what I was talking about if I told them that you were leaking leaking
> routes or that your peering session was flapping?
There is a common phone contact for a best manegement.
I'm NOT at school because i'm NOT a kid.
I work for NDSoftware all the day.
All tech contact in NDSoftware's whois have a root access on each
routers. They understand v4/v6 routing, unix administration,...
> (2) Do you have a network plan? IE; How are your current /32's
> dispersed?
Yes we have a network plan. Our network plan is not clear for this 3
/32, but now i know my errors of IP management, and NDSoftware pTLA
address plan will be clear.
We have 3 /32, but 1 /32 is enough. We have 3 /32 for have a backup if
one of our upstream can't provide us anymore a BGP peering.
> (3) Do _you_ have a network or are you simply colocated someplace on
> someone elses network?
The both currently because our network is not finish.
> If you're colocated, do you #1 have 24hr _physical_ access to the
> equipment? Can you be onsite within a reasonable amount of time in the
> event that physical access to equipment is required to remedy a
> problem? If not, do you have a "remote hands" contract in place?
Yes, we have an 24x7 access.
> (4) If you don't have your own network, how do you propose to provide
> "production quality" 6bone backbone services?
No need to have your network for provide a production quality service...
> I submit that without your own portable v4 address space for an endpoint
> of tunnels, you're at the mercy of your upstreams. If they require you to
> renumber, every one of your peers will have to reconfigure their tunnels.
Yes, i know.
> (5) I find this strange. Can you explain it?
>
> Nice routing loop there. Have you considered: (1) Not having a v6 default on your border
> router. (2) Having a connection between your two border routers and running an IGP between them?
Ops, fixed.
I have forgot to add "ifconfig lo add 3ffe:81f1:2:1::1/64" in the init
scripts of parcr1.fr.ndsoftwarenet.net.
2 eth1-0-parcr2.fr.ndsoftwarenet.net (3ffe:81f1:12:1::1) 1.023 ms
1.068 ms 0.961 ms
3 lo0-0-parcr1.fr.ndsoftwarenet.net (3ffe:81f1:2:1::1) 189.781 ms
227.238 ms 212.632 ms
> (6) It can't be a good sign for a "production quality" network when your
> route-server can't maintain a BGP peering session with your own routers:
Yes, i know, it's because i use peer group.
This problem will be fixed when parcr1.fr.ndsoftwarenet.net will have
the new AS (i will do the migration of parcr1.fr.ndsoftwarenet.net after
the 23th October).
> (7) "a. Fully maintained, up to date, 6Bone Registry entries for their
> ipv6-site inet6num, mntner, and person objects, including each
> tunnel that the Applicant has."
>
> You've got parcr3.fr.ndsoftwarenet.net listed in your ipv6-site object but:
Removed of the whois the time that we update the DNS.
parcr3.fr.ndsoftwarenet.net is the first pre-production router on our
network and don't have IPv4 connectivity.
> (8) With regards to #7 above, I suggest that with your recent policy
> change regarding BGP peers, you remove the following line from your
> ipb6-site object:
>
> remarks: NDSoftware have an open peering policy.
We are open, why remove this ?
It's not because we have delete 5-6 BGP sessions with private ASN for
new peer with pTLA and sTLA that we aren't open...
> (9) What is your "potential user community" IE; What gap are you going to
> be filling in the service delivery arena that is not already served by
> other pTLAs?
NDSoftware operates an IPv6 network and provide a lot of IPv6 services
to many projects.
We provide to:
IPv6-FR (a non profit organisation for the developement of IPv6 in
France)
tunnel broker: 200 users, each user have a /48.
NexGentCollective (http://www.nextgencollective.net/)
tunnel broker: 150 users, each user have a /48.
ATI (A tunisian ISP, http://www.ipv6net.tn/)
and a lot of others (see our whois), this services: IPv6 connectivity
(STATIC or BGP with a IPv6 block), IPv6 newsfeeds/newsread,...
We do many actions in IPv6 research, we created FNIX6 (French
International Internet Exchange IPv6, http://www.fnix6.net/), we host
many mirrors available in IPv6, we created ftp://ftp.openipv6.com/ (a
FTP with a lot of IPv6 stuff).
> (10) What purpose will having your OWN pTLA serve that your current 3
> /32's don't already serve? Keep in mind that _wanting_ your own pTLA !=
> _NEEDING_ your own pTLA and _NEEDING_ to announce a pTLA into the DFZ
> because it's a requirement for you to have your own ASN is _not_
> sufficient justification for you to be issued a
A lot of peers filter our /32 because it's not a pTLA.
We want a pTLA for can announce without any problems our network, don't
break the IPv6 aggregation and be independant of a upstream (we don't
want be down because our upstream is down).
> (11) "d. A fully maintained, and reliable, IPv6-accessible system
> providing, at a mimimum, one or more web pages, describing the
> Applicant's IPv6 services. This server must be IPv6 pingable."
>
> Looking at http://noc.ndsoftwarenet.com, information about what
> NDSOFTWARE actually *does* is strangely absent. Your peering-policy link
> returns a 404 error. Your route-filtering link returns a 404 error. Your
> usenet-policy link returns a 404 error. Register, Login and Help all
> point to your bgp-communities page, as do your "go" button and the
> advanced-search link. Home, Products & Services, Support, Download, Buy
> and Contact links at the top page simply link the whatever page you're
> currently viewing. There is no information about what your
> "company?" actually does or offers to do even.
NDSoftware website is not ready for the moment, but the NOC website is
ready.
We will fix this 404 errors.
> (12) "b. Fully maintained, and reliable, BGP4+ peering and connectivity
> between the Applicant's boundary router and the appropriate
> connection point into the 6Bone. This router must be IPv6
> pingable. This criteria is judged by members of the 6Bone
> Operations Group at the time of the Applicant's pTLA request.
>
> We have currently 101 BGP4+ sessions."
>
> Looking at parcr1.fr.ndsoftwarenet.net, I count 87 neighbors, 15 of which
> are down, 3 of which have never established an adjacency, two of the 87
> peering sessions being yourself. (84 ?real? sessions on this router.)
>
> Looking at parcr2.fr.ndsoftwarenet.net, I count 11 neighbors, 4 of which
> are down with two of the 11 peering sessions being yourself. (9
> ?real? peering sessions.)
>
> I don't know about in France but, in the US, 84 + 9 = 93 peering sessions,
> not 101 peering sessions.
>
> Can you perhaps explain your math to us?
"We have currently 101 BGP4+ sessions."
^^^^^^^^^
We have delete many peering down since many weeks after our pTLA
request, for prepare the migration of parcr1.fr.ndsoftwarenet.net
> (13) Of those 84 peering sessions, have you verified that they have
> appropriate entries in their ipv6-site objects for the tunnel/connection
> or that they have ipv6-site objects AT-ALL? Before you answer this, take
> a look at this:
A lot don't want create an ipv6-site.
> Part of properly maintaining _YOUR_ ipv6-site object is making sure that
> you don't reference an object that doesn't exist. If someone is unable or
> unwilling to create & maintain an ipv6-site object, do you really feel
> that they are a good peering candidate? I certainly don't.
They can be a good peering candidate !
A whois updated or not don't make the quality of a peering.
> (13) Just for my own personal amusement... You have a VPI/VCI pair field
> in your list of public peering points that you participate in or plan to
> participate in on your website but, your interconnects are all listed as
> 100M FE. Um, what kind of ethernet are you using that supports VPI/VCI or
> did you just think it would look "cool" to have that field?
VPI/VCI field is for a futur use.
Why all this questions ?
I don't have asked all this questions, when you have request your
pTLA....
Best Regards,
Nicolas DEFFAYET
From cliff@oisec.net Sun Oct 20 12:56:57 2002
From: cliff@oisec.net (Cliff Albert)
Date: Sun, 20 Oct 2002 13:56:57 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <1035063885.610.2182.camel@wks1.fr.corp.ndsoftware.com>
References: <002f01c277b6$7b68e3a0$210d640a@unfix.org> <1035063885.610.2182.camel@wks1.fr.corp.ndsoftware.com>
Message-ID: <20021020115657.GA7315@oisec.net>
On Sat, Oct 19, 2002 at 11:44:45PM +0200, Nicolas DEFFAYET wrote:
> All your question have a reply in the archive of this list or in my pTLA
> request.
>
> I need a pTLA, i justify it.
>
> Why IPNg.nl have a pTLA and a sTLA ?
> Please justify the both.
IPng.nl does NOT have a pTLA or sTLA. If you would check the
corresponding whois databases you would have noticed that they are owned
by Intouch NV and not IPng. IPng just has a delegation from this space
given to them by Intouch.
ipv6-site: INTOUCH-NL
origin: AS8954
descr: Intouch NV - Amsterdam, The Netherlands
Amsterdam Exchange - Kruislaan, Watergraafsmeer
country: NL
prefix: 3FFE:8110::/28
--
Cliff Albert | RIPE: CA3348-RIPE | http://oisec.net/
cliff@oisec.net | 6BONE: CA2-6BONE |
From jeroen@unfix.org Sun Oct 20 13:19:19 2002
From: jeroen@unfix.org (Jeroen Massar)
Date: Sun, 20 Oct 2002 14:19:19 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <20021020131313.S94537@Space.Net>
Message-ID: <005901c27832$eb4ff920$210d640a@unfix.org>
Gert Doering [mailto:gert@space.net] wrote:
> On Sat, Oct 19, 2002 at 10:49:50PM +0200, Jeroen Massar wrote:
> > getting that space. Notez bien that even RIPE can't request
> space from
> > thereselves as they are not a LIR. Odd rules in a odd world.
>
> The rules are not that odd.
>
> The RIPE NCC network is not "special" in any way requiring
> their own /32.
>
> If we have rules that deny "normal companies" their /32 and urge them
> to go to their upstreams, this rule has to be applied to all (!)
> end sites equally. From a network perspective, the RIPE NCC is an
> *end site*. They are not an ISP, they don't do *LIR* functions, so
> they don't qualify for an sTLA - and technicallywise, they
> don't *need*
> one either.
That's what I meant to express. They do have political reasons though.
And as most people know politics are not nice. Also it's good
to see that RIPE doesn't bend the rules even because they deal
the space out.
Thanks for clearing me up.
Greets,
Jeroen
From gert@space.net Sun Oct 20 13:25:28 2002
From: gert@space.net (Gert Doering)
Date: Sun, 20 Oct 2002 14:25:28 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <005901c27832$eb4ff920$210d640a@unfix.org>; from jeroen@unfix.org on Sun, Oct 20, 2002 at 02:19:19PM +0200
References: <20021020131313.S94537@Space.Net> <005901c27832$eb4ff920$210d640a@unfix.org>
Message-ID: <20021020142528.X94537@Space.Net>
Hi,
On Sun, Oct 20, 2002 at 02:19:19PM +0200, Jeroen Massar wrote:
> [ RIPE rules for IPv6 address space ]
>
> That's what I meant to express. They do have political reasons though.
> And as most people know politics are not nice.
Partly political, but also partly technical - the multihoming issue
isn't really solved yet, and have every end site have their own /32
announced into the global table is not a scalable approach.
The political part is the "200 customer rule", which I personally did
not like very much (it came from ARIN and APNIC), but hey, for a serious
ISP that actually is connecting customers, it's not a major obstacle.
> Also it's good to see that RIPE doesn't bend the rules even because
> they deal the space out.
Yes.
Gert Doering
-- NetMaster
--
Total number of prefixes smaller than registry allocations: 48282 (47686)
SpaceNet AG Mail: netmaster@Space.Net
Joseph-Dollinger-Bogen 14 Tel : +49-89-32356-0
80807 Muenchen Fax : +49-89-32356-299
From tjc@ecs.soton.ac.uk Sun Oct 20 14:24:37 2002
From: tjc@ecs.soton.ac.uk (Tim Chown)
Date: Sun, 20 Oct 2002 14:24:37 +0100
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <1035114737.4777.6.camel@wks1.fr.corp.ndsoftware.com>
References: <1035114737.4777.6.camel@wks1.fr.corp.ndsoftware.com>
Message-ID: <20021020132436.GS10293@login.ecs.soton.ac.uk>
On Sun, Oct 20, 2002 at 01:52:17PM +0200, Nicolas DEFFAYET wrote:
>
> A lot of peers filter our /32 because it's not a pTLA.
> We want a pTLA for can announce without any problems our network, don't
> break the IPv6 aggregation and be independant of a upstream (we don't
> want be down because our upstream is down).
The danger is that all companies will want this independence and for the
same reason demand a pTLA/SubTLA. It's certainly true for our university,
which has a /48. Given we offer IPv6 remote access, should we be allowed
a /32 to offer static /48 "site" IPv6 prefixes to any university member
wanting connectivity?
Of course part of the problem is the lack of progress of the multi6 WG,
albeit a non-trivial problem to be working on :) The "classic" IPv6
solution for our university is to take two /48's from different providers,
and for all clients to have two global addresses, but the client-side
support for handling the multiple addressing is yet to be resolved.
Tim
From itojun@iijlab.net Sun Oct 20 14:42:56 2002
From: itojun@iijlab.net (itojun@iijlab.net)
Date: Sun, 20 Oct 2002 22:42:56 +0900
Subject: [6bone] multiple address handling
In-Reply-To: tjc's message of Sun, 20 Oct 2002 14:24:37 +0100. <20021020132436.GS10293@login.ecs.soton.ac.uk>
Message-ID: <20021020134256.D36EC4B23@coconut.itojun.org>
>Of course part of the problem is the lack of progress of the multi6 WG,
>albeit a non-trivial problem to be working on :) The "classic" IPv6
>solution for our university is to take two /48's from different providers,
>and for all clients to have two global addresses, but the client-side
>support for handling the multiple addressing is yet to be resolved.
curious: where do you see problems in multiple address handling?
i don't really see any, except the lack of ability to switching
address pair for TCP (maybe we should use SCTP?).
i'm using /48 address blocks from 4 upstreams in my home, and seeing
no problem at all.
itojun
From nicolas.deffayet@ndsoftware.net Sun Oct 20 14:43:54 2002
From: nicolas.deffayet@ndsoftware.net (Nicolas DEFFAYET)
Date: 20 Oct 2002 15:43:54 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <000001c277be$45fe9900$5c59580c@who>
References: <000001c277be$45fe9900$5c59580c@who>
Message-ID: <1035121435.4771.54.camel@wks1.fr.corp.ndsoftware.com>
On Sun, 2002-10-20 at 00:24, Gregg C Levine wrote:
Hello Gregg,
> I have been lurking on this list, for a good number of years now.
> Sometimes even posting a comment, or a gripe. This is more along the
> lines of both. I have been monitoring the traffic discussing
> NDSOFTWARE's request. Both finding the original message to Bob Fink, and
> the list, and everything. Sorry Mr. Deffayet, I disagree. For one, you
> do need to spell out who will be using the service. Is it for your
> company? Yourself? What? Who, even? Unless you can spell out neatly the
> answers to my questions, I am inclined to agree with everyone else. I am
> also agreeing with the people I have disagreed with early on. I might
> also, add, even Master Yoda's methods of speaking isn't that confusing.
Who will be using the service:
- NDSoftware (my company)
- many projects (here a list of main projects):
* IPv6-FR
A non profit organisation for the developement of IPv6 in France
IPv6-FR run a tunnel broker and have currently 200 users, each user have
a /48
=> NDSoftware provide to IPv6-FR: 1 /35 and a native IPv6 connectivity.
* NextGenCollective
IPv6 research
http://www.nextgencollective.net/
NGC[NextGen Collective] provides IPv6-over-IPv4 tunnels to people all
over the world.
ASpath-tree: http://www.nextgencollective.net/bgp4/ (AS65526 is my old
private ASN)
=> NDSoftware provide to NextGenCollective: 2 /44, 1 /40, 1 /36 and a
tunnel IPv6 over IPv4 with BGP (full transit).
* IPng.org.uk
IPv6 tunnel broker
http://www.ipng.org.uk/
ASpath-tree: http://www.ipng.org.uk/bgp/bgp-page-complete.html (AS65526
is my old private ASN)
=> NDSoftware provide to IPng.org.uk: 1 /44, 1 /40 and a tunnel IPv6
over IPv4 with BGP (full transit).
* ILS
Italian Linux Society
ILS provide IPv6 connectivity to italian user groups and organizations
experimenting with IPv6.
ILS host the IPv6 IRC server calvino.freenode.net
=> NDSoftware provide to ILS: 1 /44 and a tunnel IPv6 over IPv4 with BGP
(full transit).
* ATI
A tunisian ISP
http://www.ipv6net.tn/
http://www.ipv6net.tn/ipv6-Tunisia.pdf
=> NDSoftware provide to ATI: 1 /44 and a tunnel IPv6 over IPv4 with BGP
(full transit). NDSoftware have help ATI for the IPv6 deployement in
tunisia. ATI plan later to request a pTLA.
* FABIONNE
A projet for do IPv6 Debian package
http://debian-ipv6.fabionne.net/
=> NDSoftware provide to FABIONNE: a tunnel IPv6 over IPv4 with BGP
(full transit) and host a mirror for this projet
(http://debian-ipv6.mirrors.ndsoftwarenet.com/).
ESMT
An university in Senegal
=> NDSoftware provide to ESMT: 1 /44 and a tunnel IPv6 over IPv4.
Best regards,
Nicolas DEFFAYET
From luv@nerim.net Sun Oct 20 14:32:36 2002
From: luv@nerim.net (Ludovic Victor)
Date: Sun, 20 Oct 2002 15:32:36 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
References: <200210191954.g9JJsmgj023818@ludwigV.sources.org>
Message-ID: <000b01c2783d$27091500$bb29c6d4@ID>
> As a French LIR myself, I can say that I find annoying that cowards'
comments
> are forwarded to the 6bone list.
That's very nice to call coward peoples who wants to have their right to
stay anonymous preserved... I think Christian has done
The Right Think(tm) removing sender's name before forwarding the
comments. Next time you are doing your citizen's duty, voting at a
political election, remember you are cowardly hiding your choice by
staying anonymous.
By the way, I found the comments of that person to be very insightfull and
quite true after consideration of what has been said by that NDSoftware guy
in the internal newsgroups of its upstream provider (Nerim).
> Nicolas Deffayet is obnoxious enough with what he does publicly (such as
> pretending his toy is the "first IPv6 IXP in France"). No need to add
> anonymous accusations.
What accusations ? I read facts.
> > >time to deploy native IPv6, we are truely sad to see that a kid playing
> > >with some PCs running Zebra could annihilate all of those efforts.
>
> That b...s...t about Zebra (which we use in production) is one more thing
that
> should warn 6bone readers against that anonymous comment.
You obviously didn't get the idea. Where is it said PC running Zebra is bad
or unstable (except in Nicolas Deffayet's posts where he says that reaching
more than a hundred sessions, it starts to be unstable) ?
You could run MRTd, gated or even commercial softwares over commercial
platforms, nobody cares...
When I read this, I understand : "It is very simple nowadays to setup a
BGP router, and even more simple to do a lot of mess with it.". Didn't
you experienced it when your Zebra thingy has done weird advertisements
to Nerim (and possibily all of your peers) over the FreeIX ?
Ludo.
From nicolas.deffayet@ndsoftware.net Sun Oct 20 14:59:46 2002
From: nicolas.deffayet@ndsoftware.net (Nicolas DEFFAYET)
Date: 20 Oct 2002 15:59:46 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <20021020132436.GS10293@login.ecs.soton.ac.uk>
References:
<1035114737.4777.6.camel@wks1.fr.corp.ndsoftware.com>
<20021020132436.GS10293@login.ecs.soton.ac.uk>
Message-ID: <1035122387.4771.70.camel@wks1.fr.corp.ndsoftware.com>
On Sun, 2002-10-20 at 15:24, Tim Chown wrote:
> On Sun, Oct 20, 2002 at 01:52:17PM +0200, Nicolas DEFFAYET wrote:
> >
> > A lot of peers filter our /32 because it's not a pTLA.
> > We want a pTLA for can announce without any problems our network, don't
> > break the IPv6 aggregation and be independant of a upstream (we don't
> > want be down because our upstream is down).
>
> The danger is that all companies will want this independence and for the
> same reason demand a pTLA/SubTLA. It's certainly true for our university,
> which has a /48. Given we offer IPv6 remote access, should we be allowed
> a /32 to offer static /48 "site" IPv6 prefixes to any university member
> wanting connectivity?
All ISP/company/project who provide IPs to another ISP/company/project
and have many upstream MUST have a pTLA/sTLA.
In http://www.6bone.net/6bone_pTLA_list.html, a lot of pTLA aren't used
or are used only for a /48 and/or have only one upstream.
An exemple: MOTOROLA-LABS, have only one upstream. Do you think that
they need a pTLA ? If their upstream is down, this pTLA is not anymore
announced. MOTOROLA-LABS don't provide IPs and a /48 is enough for their
activity...
NDSoftware provide IPs to another ISP/company/project and have many
upstream...
> Of course part of the problem is the lack of progress of the multi6 WG,
> albeit a non-trivial problem to be working on :) The "classic" IPv6
> solution for our university is to take two /48's from different providers,
> and for all clients to have two global addresses, but the client-side
> support for handling the multiple addressing is yet to be resolved.
It's a lot of do-it-yourself.
It's not a good solution, because a lot of software need only an IP...
Best Regards,
Nicolas DEFFAYET
From arien+6bone@ams-ix.net Sun Oct 20 15:08:41 2002
From: arien+6bone@ams-ix.net (Arien Vijn)
Date: Sun, 20 Oct 2002 16:08:41 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October
2002
In-Reply-To: <005901c27832$eb4ff920$210d640a@unfix.org>
Message-ID:
On 20-10-2002 14:19PM, "Jeroen Massar" wrote:
> Gert Doering [mailto:gert@space.net] wrote:
>
>> On Sat, Oct 19, 2002 at 10:49:50PM +0200, Jeroen Massar wrote:
>>> getting that space. Notez bien that even RIPE can't request
>> space from
>>> thereselves as they are not a LIR. Odd rules in a odd world.
>>
>> The rules are not that odd.
>>
>> The RIPE NCC network is not "special" in any way requiring
>> their own /32.
>>
>> If we have rules that deny "normal companies" their /32 and urge them
>> to go to their upstreams, this rule has to be applied to all (!)
>> end sites equally. From a network perspective, the RIPE NCC is an
>> *end site*. They are not an ISP, they don't do *LIR* functions, so
>> they don't qualify for an sTLA - and technicallywise, they
>> don't *need*
>> one either.
>
> That's what I meant to express. They do have political reasons though.
> And as most people know politics are not nice. Also it's good
> to see that RIPE doesn't bend the rules even because they deal
> the space out.
>
In the APNIC region RIR/NIRs are "critical infrastructure" and therefore are
eligible for a /32. See:
http://www.ripe.net/ripe/meetings/archive/ripe-43/presentations/ripe43-plena
ry-apnic/sld024.html
APNIC explicitly excluded IXPs as critical infrastructure. That is
questionable. But a good argument for this exclusion is that it is pretty
easy to start an "Internet Exchange". Just put a switch in a rack and call
it an Internet Exchange and request "golden" address space.
Please note that I do not say that Nicolas is just setting up an "IXP" to
get address resources. I am curious why NDSoftware/FN6IX is not requesting
space under RIPE-256 though.
To solve the neutrality issues, AMS-IX decided to put serious efforts in a
proper multi homing solution. We think (hope) that this approach will be a
much more positive contribution to the Internet than arguing the
criticalness of an IXP.
Kind regards, Arien
--
Arien Vijn
Amsterdam Internet Exchange
http://www.ams-ix.net
From tvo@EnterZone.Net Sun Oct 20 15:24:50 2002
From: tvo@EnterZone.Net (John Fraizer)
Date: Sun, 20 Oct 2002 10:24:50 -0400 (EDT)
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <1035114737.4777.6.camel@wks1.fr.corp.ndsoftware.com>
Message-ID:
On 20 Oct 2002, Nicolas DEFFAYET wrote:
> On Sun, 2002-10-20 at 10:42, John Fraizer wrote:
>
> > (1) Who are the following and what are their qualifications to be
> > technical contacts? The last thing I want to hear on the other end of the
>
> There is a common phone contact for a best manegement.
>
OK. So, _someone_ is going to answer that phone number when we call at
zero-dark:thirty with a peering issue, right?
> I'm NOT at school because i'm NOT a kid.
> I work for NDSoftware all the day.
OK. I'm still wondering what NDSoftware does. Don't get me wrong. The
ASPath Tree is slick, the ftp site is handy for some I'm sure, providing
6bone connectivity is definately a service but, there has to be something
going on that actually generates income, otherwise, you're hemorrhaging
money on colocation and IP transit charges. Generally, when someone forms
a company, as you state you have done, it is to generate income. To do
that, you have to offer services that people will purchase. You _can't_
sell 6bone access and thus-far, every "service" you claim to provide is
6bone-centric.
I'm simply looking at the overall health of the 6bone here. If you're
issued a pTLA and start providing "services" to folks with that address
space and suddenly, your "company" goes tits-up because you're not able to
pay your colocation/transit fees (because your "company" isn't actually
SELLING anything to generate revenue) then not only have you embarrassed
yourself but, you will have inconvenienced who knows how many other
people.
>
> All tech contact in NDSoftware's whois have a root access on each
> routers. They understand v4/v6 routing, unix administration,...
>
Wow. You're a trusting soul there. SUDU is your friend, Dude. You might
want to look at the man page for it.
> > (2) Do you have a network plan? IE; How are your current /32's
> > dispersed?
>
> Yes we have a network plan. Our network plan is not clear for this 3
> /32, but now i know my errors of IP management, and NDSoftware pTLA
> address plan will be clear.
OK so, you _plan_ to have a network plan then. While not a requirement
for being issued a pTLA of v6 space, most every RIR I've looked at will
require you to show proper utilization of current upstream assigned
address space, along with appropriate SWIP or rwhois entries for your
subsequent assignment of that address space to your downstreams prior to
issuing Provider Independent v4 address space to you.
I'm simply looking for you to demonstrate that you or one of your
employees can properly maintain appropriate records for address
allocation. Just because the 6bone is experimental does not relieve you,
the administrator of a network, from the burdon of due diligence. Suppose
one of your downstreams started a SPAM campaign to v6 connected
mailservers or started trying to hack into other v6 connected
systems? How long does it take you to track down the appropriate contact
information for the source address? Do you have appropriate records to
provide to law enforcement agencies in the event that you are subpoenaed
for this type of information?
>
> We have 3 /32, but 1 /32 is enough. We have 3 /32 for have a backup if
> one of our upstream can't provide us anymore a BGP peering.
>
Ya, like if they were to say "your peering session is going to die in a
week because our routers are overloaded with BGP sessions. We've decided
to drop all of our BGP peers who are using reserved ASNs." -- Something
like that?
> > (3) Do _you_ have a network or are you simply colocated someplace on
> > someone elses network?
>
> The both currently because our network is not finish.
>
OK. What does your network consist of? Keep in mind that I went to your
website looking for this information. You don't have it listed. If I
could find it, I wouldn't be asking these questions.
> > If you're colocated, do you #1 have 24hr _physical_ access to the
> > equipment? Can you be onsite within a reasonable amount of time in the
> > event that physical access to equipment is required to remedy a
> > problem? If not, do you have a "remote hands" contract in place?
>
> Yes, we have an 24x7 access.
>
> > (4) If you don't have your own network, how do you propose to provide
> > "production quality" 6bone backbone services?
>
> No need to have your network for provide a production quality service...
Tell that to the KPNQwest folks.
> > I submit that without your own portable v4 address space for an endpoint
> > of tunnels, you're at the mercy of your upstreams. If they require you to
> > renumber, every one of your peers will have to reconfigure their tunnels.
>
> Yes, i know.
>
So, when you went after your ASN, did you try to brow-beat some v4 space
out of RIPE as well?
> > (5) I find this strange. Can you explain it?
> >
> > Nice routing loop there. Have you considered: (1) Not having a v6 default on your border
> > router. (2) Having a connection between your two border routers and running an IGP between them?
>
> Ops, fixed.
>
> I have forgot to add "ifconfig lo add 3ffe:81f1:2:1::1/64" in the init
> scripts of parcr1.fr.ndsoftwarenet.net.
Wow. I can't imagine trying to explain that to one of my customers. This
is all about attention to detail Nicolas. So, you get your own pTLA and
people start actually listening to and propagating your announcements and
you "forget" a little thing like applying an access-list or route-map to
a peering session. Guess what? Your lack of attention to detail does
more than embarrass you. It can cause service effecting outages for a
whole ton of OTHER people.
>
> 2 eth1-0-parcr2.fr.ndsoftwarenet.net (3ffe:81f1:12:1::1) 1.023 ms
> 1.068 ms 0.961 ms
> 3 lo0-0-parcr1.fr.ndsoftwarenet.net (3ffe:81f1:2:1::1) 189.781 ms
> 227.238 ms 212.632 ms
Wow! Just how much distance is between those two routers? I'm just
wondering because that's a longer RTT than from here to London.
Oh, something for you to ponder since the previous routing loop certainly
looked like you had a default route set up on that router::
>From RFC2772:
3.7 Default routes
6Bone core pTLA routers MUST be default-free.
pTLAs MAY advertise a default route to any downstream peer (non-pTLA
site). Transit pNLAs MAY advertise a default route to any of their
downstreams (other transit pNLA or leaf site).
>
> > (6) It can't be a good sign for a "production quality" network when your
> > route-server can't maintain a BGP peering session with your own routers:
>
> Yes, i know, it's because i use peer group.
> This problem will be fixed when parcr1.fr.ndsoftwarenet.net will have
> the new AS (i will do the migration of parcr1.fr.ndsoftwarenet.net after
> the 23th October).
Hrm... Guess what. You can override nearly ANY "peer group" setting with
a per-peer setting.
>
> > (7) "a. Fully maintained, up to date, 6Bone Registry entries for their
> > ipv6-site inet6num, mntner, and person objects, including each
> > tunnel that the Applicant has."
> >
> > You've got parcr3.fr.ndsoftwarenet.net listed in your ipv6-site object but:
>
> Removed of the whois the time that we update the DNS.
>
> parcr3.fr.ndsoftwarenet.net is the first pre-production router on our
> network and don't have IPv4 connectivity.
OK. If you say so. Just wondering. How long does it take to update a
zone file and type "kill -HUP [pid of named]"?
>
> > (8) With regards to #7 above, I suggest that with your recent policy
> > change regarding BGP peers, you remove the following line from your
> > ipb6-site object:
> >
> > remarks: NDSoftware have an open peering policy.
>
> We are open, why remove this ?
>
> It's not because we have delete 5-6 BGP sessions with private ASN for
> new peer with pTLA and sTLA that we aren't open...
>
Well, if you're full, as in, you're having to remove current peers to
bring on NEW peers, I wouldn't consider you to be "open."
> > (9) What is your "potential user community" IE; What gap are you going to
> > be filling in the service delivery arena that is not already served by
> > other pTLAs?
>
> NDSoftware operates an IPv6 network and provide a lot of IPv6 services
> to many projects.
>
Inquiring minds would like to see a network map for your network. I'm
serious. What geographic points does it connect? What media? Is it a
meshed network? 227ms from one NDSoftware router in France to another
NDSoftware router in France doesn't exactly scream "I've got a production
quality network!" if you know what I mean.
> We provide to:
>
> IPv6-FR (a non profit organisation for the developement of IPv6 in
> France)
> tunnel broker: 200 users, each user have a /48.
>
Hrm...
role: IPv6-FR NOC
address: IPv6-FR
address: 57 rue du president Wilson
address: 92300 Levallois-Perret
address: France
phone: +33 671887502
role: NDSoftware NOC
address: NDSoftware
address: 57 rue du president Wilson
address: 92300 Levallois-Perret
address: France
phone: +33 671887502
I'm sorry Nicolas. Providing address space to YOURSELF doesn't
count! Sheesh!
>
> NexGentCollective (http://www.nextgencollective.net/)
> tunnel broker: 150 users, each user have a /48.
ipv6-site: NEXTGENCOLLECTIVE
origin: AS65055
descr: NextGenCollective IPv6 Research Organization
country: US
prefix: 3FFE:8271:A020::/44
prefix: 3FFE:8271:A030::/44
prefix: 3FFE:8271:B000::/40
prefix: 3FFE:2C01:1000::/36
tunnel: IPv6 in IPv4 wireless.cs.twsu.edu ->
parcr1.fr.ndsoftwarenet.net NDSOFTWARE BGP4+
contact: AB12-6BONE
contact: BP6-6BONE
remarks: Report abuses at abuse@nextgencollective.net
url: http://www.nextgencollective.net
mnt-by: MNT-NEXTGENCOLLECTIVE
changed: basit@nextgencollective.net 20020819
source: 6BONE
person: Bryce PORTER
address: Peoria, IL
phone: +1 555-555-5555
e-mail: x86@nextgencollective.net
nic-hdl: BP6-6BONE
mnt-by: MNT-NEXTGENCOLLECTIVE
changed: x86@nextgencollective.net 20020405
source: 6BONE
person: Abdul Basit
address: 3116 E.18th Street
address: Wichita , KS 67214 USA
phone: +1 316 978-3729
e-mail: basit@nextgencollective.net
e-mail: basit@basit.cc
nic-hdl: AB12-6BONE
url: http://basit.cc
notify: basit@basit.cc
mnt-by: MNT-NEXTGENCOLLECTIVE
changed: basit@basit.cc 20020220
source: 6BONE
Don't you think that a tunnel-broker housed in Wichita, KS, USA would be
better served by a 6bone pTLA *IN* the USA? Also, with your current
peering policy change, isn't this site going to get NIXED? I note their
use of a Reserved ASN.
>
> ATI (A tunisian ISP, http://www.ipv6net.tn/)
>
$ traceroute6 3FFE:8271:A010::1
traceroute to 3FFE:8271:A010::1 (3ffe:8271:a010::1) from
3ffe:4010:ff09::1, 30 hops max, 16 byte packets
1 enterzone-ndsoftware-gw.paris.ipv6.enterzone.net (3ffe:4010:ff09::2) 762.094 ms 797.236 ms 815.096 ms
2 feth0-1-parcr1.fr.ndsoftwarenet.net (3ffe:81f1:0:1::1) 1286.27 ms 1567.95 ms 1651.54 ms
3 3ffe:8270:0:1::64 (3ffe:8270:0:1::64) 1021.78 ms 1088.46 ms 1132.56 ms
I do believe that they would also be better served by someone
geographically closer to them. Note: Our peering session with you is up
at YOUR request. RTTs to you are over 6X longer than any other peer we
have. Also note: reverse for 3ffe:8270:0:1::64 is broken or non-existant.
> We do many actions in IPv6 research, we created FNIX6 (French
> International Internet Exchange IPv6, http://www.fnix6.net/), we host
ipv6-site: FNIX6
origin: AS25358
descr: French National Internet eXchange IPv6
country: FR
contact: NDS1-6BONE
url: http://www.fnix6.net/
notify: notify@ndsoftwarenet.com
mnt-by: MNT-NDSOFTWARE
changed: nicolas.deffayet@ndsoftware.net 20021018
source: 6BONE
Which is it? French National Internet eXchange or the French
International Internet Exchange?
2002-10-08 NDSoftware launch FNIX6 - I get it. You haven't decided. It's
not even two weeks old! I must say that you've got more information on
that website than you do on the NDSoftware site though.
> many mirrors available in IPv6, we created ftp://ftp.openipv6.com/ (a
> FTP with a lot of IPv6 stuff).
ncftp ...patched/eggdrop/1.6.12 > ls
eggdrop1.6.12.ipv6.precomp.linux.tgz info.txt
Now there's a serious service. Robbie Pointer will be
proud. (Sidenote: I went to school with Robbie. I hope he's changed
since then!)
> > (10) What purpose will having your OWN pTLA serve that your current 3
> > /32's don't already serve? Keep in mind that _wanting_ your own pTLA !=
> > _NEEDING_ your own pTLA and _NEEDING_ to announce a pTLA into the DFZ
> > because it's a requirement for you to have your own ASN is _not_
> > sufficient justification for you to be issued a
>
> A lot of peers filter our /32 because it's not a pTLA.
> We want a pTLA for can announce without any problems our network, don't
> break the IPv6 aggregation and be independant of a upstream (we don't
> want be down because our upstream is down).
>
Well, that's a valid wish and I can understand your point. Didn't you say
earlier though that you had 3 /32's for redundancy already?
> > (11) "d. A fully maintained, and reliable, IPv6-accessible system
> > providing, at a mimimum, one or more web pages, describing the
> > Applicant's IPv6 services. This server must be IPv6 pingable."
> >
> > Looking at http://noc.ndsoftwarenet.com, information about what
> > NDSOFTWARE actually *does* is strangely absent. Your peering-policy link
> > returns a 404 error. Your route-filtering link returns a 404 error. Your
> > usenet-policy link returns a 404 error. Register, Login and Help all
> > point to your bgp-communities page, as do your "go" button and the
> > advanced-search link. Home, Products & Services, Support, Download, Buy
> > and Contact links at the top page simply link the whatever page you're
> > currently viewing. There is no information about what your
> > "company?" actually does or offers to do even.
>
> NDSoftware website is not ready for the moment, but the NOC website is
> ready.
>
> We will fix this 404 errors.
>
That *was* the NOC website I was talking about. Again, I'm simply
pointing out attention to detail flaws here Nicolas. I'll take 10 seconds
of attention to detail over 10 hours of the best intentions EVERY TIME.
> > I don't know about in France but, in the US, 84 + 9 = 93 peering sessions,
> > not 101 peering sessions.
> >
> > Can you perhaps explain your math to us?
>
> "We have currently 101 BGP4+ sessions."
> ^^^^^^^^^
> We have delete many peering down since many weeks after our pTLA
> request, for prepare the migration of parcr1.fr.ndsoftwarenet.net
So, that should have read "when I wrote this, we had 101 BGP4+ peering
sessions but, I'm getting ready to axe a bunch of them."
>
> > (13) Of those 84 peering sessions, have you verified that they have
> > appropriate entries in their ipv6-site objects for the tunnel/connection
> > or that they have ipv6-site objects AT-ALL? Before you answer this, take
> > a look at this:
>
> A lot don't want create an ipv6-site.
>
> > Part of properly maintaining _YOUR_ ipv6-site object is making sure that
> > you don't reference an object that doesn't exist. If someone is unable or
> > unwilling to create & maintain an ipv6-site object, do you really feel
> > that they are a good peering candidate? I certainly don't.
>
> They can be a good peering candidate !
>
> A whois updated or not don't make the quality of a peering.
I SERIOUSLY BEG TO DIFFER! If someone is too damned lazy to create and
maintain an ipv6-site object, how on earth can you expect them to maintain
appropriate BGP filters, allocation records, etc, etc, etc? Man, it is
_OBVIOUS_ that this is a *toy* to you.
By virtue of your ipv6-site object referencing tunnel endpoints that have
no corresponding ipv6-site object, it is NOT accurate and you (and your
sites with nonexistant or invalid ipv6-site objects) are in violation of
RFC2772:
5. The 6Bone Registry
The 6Bone registry is a RIPE-181 database with IPv6 extensions used
to store information about the 6Bone, and its sites. The 6bone is
accessible at:
)
Each 6Bone site MUST maintain the relevant entries in the 6Bone
registry. In particular, the following object MUST be present for all
6Bone leaf sites, pNLAs and pTLAs:
- IPv6-site: site description
- Inet6num: prefix delegation (one record MUST exist for each
delegation)
- Mntner: contact info for site maintance/administration staff.
Other object MAY be maintained at the discretion of the sites such as
routing policy descriptors, person, or role objects. The Mntner
object MUST make reference to a role or person object, but those MAY
NOT necessarily reside in the 6Bone registry. They can be stored
within any of the Internet registry databases (ARIN, APNIC, RIPE-NCC,
etc.)
6. Guidelines for new sites joining the 6Bone
New sites joining the 6Bone should seek to connect to a transit pNLA
or a pTLA within their region, and preferably as close as possible to
their existing IPv4 physical and routing path for Internet service.
The 6Bone web site at has various information
and tools to help find candidate 6bone networks.
Any site connected to the 6Bone MUST maintain a DNS server for
forward name lookups and reverse address lookups. The joining site
MUST maintain the 6Bone objects relative to its site, as describe in
section 5.
The upstream provider MUST delegate the reverse address translation
zone in DNS to the joining site, or have an agreement in place to
perform primary DNS for that downstream. The provider MUST also
create the 6Bone registry inet6num object reflecting the delegated
address space.
Now, from section 7 of RFC2772, a bit more for you to ponder:
During the entire qualifying period the Applicant must be
operational providing the following:
a. Fully maintained, up to date, 6Bone Registry entries for their
ipv6-site inet6num, mntner, and person objects, including each
tunnel that the Applicant has.
4. The pTLA Applicant MUST commit to abide by the current 6Bone
operational rules and policies as they exist at time of its
application, and agree to abide by future 6Bone backbone
operational rules and policies as they evolve by consensus of the
6Bone backbone and user community.
Now, since you obviously don't care if your peers maintain their ipv6-site
objects or even HAVE them for that matter, how is it that you are abiding
by RFC2772, Section 5?
>
> Why all this questions ?
> I don't have asked all this questions, when you have request your
> pTLA....
Nicolas, NOBODY asked any questions when I requested a pTLA for EnterZone.
We already held an sTLA, our website provided accurate, up-to-date
information about our service offerings as well as our company and we had,
at the time of the request, been in business for nearly 7 years providing
IP transit and datacenter services. A large portion of the v6 community
is running the looking-glass code that I wrote, I wrote the first exchange
point route-server hack (which has become the "transparent" features) for
the Zebra code and am VERY active in the Zebra mailing list, NANOG, and am
one of the moderators for the Linux-ATM project. Suffice it to say that
people knew who I was, who EnterZone, Inc was and they didn't have any
question about our ability to provide "production quality" services, or if
we had a potential "user community."
Had anyone had questions for us, I would have personally sat down and in
great detail answered any concern that they had. Your defensive posture
is doing nothing to ease any anxieties people may have over your becoming
a pTLA holder and participating the Default Free Zone.
This is _not_ anything personal Nicolas. It's all about due
diligence. Any BUSINESSMAN should understand that.
---
John Fraizer | High-Security Datacenter Services |
President | Dedicated circuits 64k - 155M OC3 |
EnterZone, Inc | Virtual, Dedicated, Colocation |
http://www.enterzone.net/ | Network Consulting Services |
From pekkas@netcore.fi Sun Oct 20 15:36:10 2002
From: pekkas@netcore.fi (Pekka Savola)
Date: Sun, 20 Oct 2002 17:36:10 +0300 (EEST)
Subject: sTLA alloc policies [Re: [6bone] pTLA request NDSOFTWARE - review
closes 23 October 2002]
In-Reply-To: <20021020142528.X94537@Space.Net>
Message-ID:
On Sun, 20 Oct 2002, Gert Doering wrote:
> On Sun, Oct 20, 2002 at 02:19:19PM +0200, Jeroen Massar wrote:
> > [ RIPE rules for IPv6 address space ]
> >
> > That's what I meant to express. They do have political reasons though.
> > And as most people know politics are not nice.
>
> Partly political, but also partly technical - the multihoming issue
> isn't really solved yet, and have every end site have their own /32
> announced into the global table is not a scalable approach.
>
> The political part is the "200 customer rule", which I personally did
> not like very much (it came from ARIN and APNIC), but hey, for a serious
> ISP that actually is connecting customers, it's not a major obstacle.
Speaking of which, I'd be really interested in knowing how Internet
Software Consortium is going to fill the "200 customer rule":
http://ws.arin.net/cgi-bin/whois.pl?queryinput=ISC6-1
I never thought they'd be in the ISP business..
Oh, Nokia must also have colored the truth slightly..
--
Pekka Savola "Tell me of difficulties surmounted,
Netcore Oy not those you stumble over and fall"
Systems. Networks. Security. -- Robert Jordan: A Crown of Swords
From arien+6bone@ams-ix.net Sun Oct 20 16:19:05 2002
From: arien+6bone@ams-ix.net (Arien Vijn)
Date: Sun, 20 Oct 2002 17:19:05 +0200
Subject: [6bone] multiple address handling
In-Reply-To: <20021020134256.D36EC4B23@coconut.itojun.org>
Message-ID:
On 20-10-2002 15:42PM, "itojun@iijlab.net" wrote:
>> Of course part of the problem is the lack of progress of the multi6 WG,
>> albeit a non-trivial problem to be working on :) The "classic" IPv6
>> solution for our university is to take two /48's from different providers,
>> and for all clients to have two global addresses, but the client-side
>> support for handling the multiple addressing is yet to be resolved.
>
> curious: where do you see problems in multiple address handling?
> i don't really see any, except the lack of ability to switching
> address pair for TCP (maybe we should use SCTP?).
>
This is the major problem. Increased reliability is certainly a factor why
one wants to be multi homed.
Address selection is another issue.
> i'm using /48 address blocks from 4 upstreams in my home, and seeing
> no problem at all.
>
Glad to see someone is actually happily doing this. For short sessions it
should not be a problem indeed.
Do you monitor how the distribution over the 4 upstream goes? How does your
client select which source address it will use.
Are you running multi homed servers or do you just use it as a client?
Kind regards, Arien
From tjc@ecs.soton.ac.uk Sun Oct 20 16:35:05 2002
From: tjc@ecs.soton.ac.uk (Tim Chown)
Date: Sun, 20 Oct 2002 16:35:05 +0100
Subject: sTLA alloc policies [Re: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002]
In-Reply-To:
References: <20021020142528.X94537@Space.Net>
Message-ID: <20021020153504.GV10293@login.ecs.soton.ac.uk>
On Sun, Oct 20, 2002 at 05:36:10PM +0300, Pekka Savola wrote:
>
> Speaking of which, I'd be really interested in knowing how Internet
> Software Consortium is going to fill the "200 customer rule":
Or small to medium sized NRENs in Europe. I spoke last week to someone from
one of the smallest European research networks who felt he couldn't get a
SubTLA for that reason (not 200 universities in the country). This isn't
(I hope) the type of organisation the new rules are trying to exclude. (in
other ways the new rules are very open, which may lead to (increased)
6bonisation of the 2001: space...
Tim
From gert@space.net Sun Oct 20 16:39:17 2002
From: gert@space.net (Gert Doering)
Date: Sun, 20 Oct 2002 17:39:17 +0200
Subject: sTLA alloc policies [Re: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002]
In-Reply-To: ; from pekkas@netcore.fi on Sun, Oct 20, 2002 at 05:36:10PM +0300
References: <20021020142528.X94537@Space.Net>
Message-ID: <20021020173917.Z94537@Space.Net>
Hi,
On Sun, Oct 20, 2002 at 05:36:10PM +0300, Pekka Savola wrote:
> > The political part is the "200 customer rule", which I personally did
> > not like very much (it came from ARIN and APNIC), but hey, for a serious
> > ISP that actually is connecting customers, it's not a major obstacle.
>
> Speaking of which, I'd be really interested in knowing how Internet
> Software Consortium is going to fill the "200 customer rule":
>
> http://ws.arin.net/cgi-bin/whois.pl?queryinput=ISC6-1
>
> I never thought they'd be in the ISP business..
>
> Oh, Nokia must also have colored the truth slightly..
Well, Nokia (and Cisco and such) might actually be able to justify it
- by declaring all their national offices as customers. Legally, they
are likely to be separate entities anyway.
*I* personally don't really mind if "the 100 biggest companies in the
world" get a sTLA - if they can provide decent network documentation,
fine with me.
As for ISC, hmmm, interesting question. Being a RIPE person, I can't
comment on what's going on in ARIN land...
Gert Doering
-- NetMaster
--
Total number of prefixes smaller than registry allocations: 48282 (47686)
SpaceNet AG Mail: netmaster@Space.Net
Joseph-Dollinger-Bogen 14 Tel : +49-89-32356-0
80807 Muenchen Fax : +49-89-32356-299
From gert@space.net Sun Oct 20 16:43:48 2002
From: gert@space.net (Gert Doering)
Date: Sun, 20 Oct 2002 17:43:48 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: ; from arien+6bone@ams-ix.net on Sun, Oct 20, 2002 at 04:08:41PM +0200
References: <005901c27832$eb4ff920$210d640a@unfix.org>
Message-ID: <20021020174348.A94537@Space.Net>
Hi,
On Sun, Oct 20, 2002 at 04:08:41PM +0200, Arien Vijn wrote:
> In the APNIC region RIR/NIRs are "critical infrastructure" and therefore are
> eligible for a /32. See:
>
> http://www.ripe.net/ripe/meetings/archive/ripe-43/presentations/ripe43-plena
> ry-apnic/sld024.html
Yes, I've seen that presentation. I consider that whole thing hilarious -
in my opinion, the only thing that is "criticial infrastructure" enough
to warrant fixed, permanent, "personal" IP addresses are root DNS servers.
The RIPE members have voted against introducing "golden space" in the v4
world a couple of times, and (IIRC) also did so for v6.
What's so special (technically) about a RIR/NIR's network? Yes, there's
a whois server, and a web server. But why is it more critical for the
whole internet than, say, "www.google.com"? I'd consider the latter far
more important for the majority of internet users.
(As for the IXP address space: tricky. I'm sure we've had this discussion
in person before, and I just have a different opinion - the DECIX runs
well with non-routeable IXP space, while the INXS runs well on C&W space)
Gert Doering
-- NetMaster
--
Total number of prefixes smaller than registry allocations: 48282 (47686)
SpaceNet AG Mail: netmaster@Space.Net
Joseph-Dollinger-Bogen 14 Tel : +49-89-32356-0
80807 Muenchen Fax : +49-89-32356-299
From hansolofalcon@worldnet.att.net Sun Oct 20 16:53:52 2002
From: hansolofalcon@worldnet.att.net (Gregg C Levine)
Date: Sun, 20 Oct 2002 11:53:52 -0400
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <1035121435.4771.54.camel@wks1.fr.corp.ndsoftware.com>
Message-ID: <000f01c27850$e3fcaf60$f356580c@who>
Hello from Gregg C Levine
All very nice. But it still doesn't answer any of John Fraize's many
questions. In fact, I'd say it looks slick. But it won't work, nor will
it stack up. Have you studied completely the man pages behind your
server's software, on each component? Some of them are relevant, and
aren't just put there for the machine's sake. I'm sorry Nicholas, but I
still agree with John Fraize, and everyone else, and disagree with you.
-------------------
Gregg C Levine hansolofalcon@worldnet.att.net
------------------------------------------------------------
"The Force will be with you...Always." Obi-Wan Kenobi
"Use the Force, Luke." Obi-Wan Kenobi
(This company dedicates this E-Mail to General Obi-Wan Kenobi )
(This company dedicates this E-Mail to Master Yoda )
> -----Original Message-----
> From: 6bone-admin@mailman.isi.edu [mailto:6bone-admin@mailman.isi.edu]
On
> Behalf Of Nicolas DEFFAYET
> Sent: Sunday, October 20, 2002 9:44 AM
> To: Gregg C Levine
> Cc: 6bone Mail List
> Subject: RE: [6bone] pTLA request NDSOFTWARE - review closes 23
October
> 2002
>
> On Sun, 2002-10-20 at 00:24, Gregg C Levine wrote:
>
> Hello Gregg,
>
> > I have been lurking on this list, for a good number of years now.
> > Sometimes even posting a comment, or a gripe. This is more along the
> > lines of both. I have been monitoring the traffic discussing
> > NDSOFTWARE's request. Both finding the original message to Bob Fink,
and
> > the list, and everything. Sorry Mr. Deffayet, I disagree. For one,
you
> > do need to spell out who will be using the service. Is it for your
> > company? Yourself? What? Who, even? Unless you can spell out neatly
the
> > answers to my questions, I am inclined to agree with everyone else.
I am
> > also agreeing with the people I have disagreed with early on. I
might
> > also, add, even Master Yoda's methods of speaking isn't that
confusing.
>
> Who will be using the service:
>
> - NDSoftware (my company)
>
> - many projects (here a list of main projects):
>
> * IPv6-FR
> A non profit organisation for the developement of IPv6 in France
> IPv6-FR run a tunnel broker and have currently 200 users, each user
have
> a /48
> => NDSoftware provide to IPv6-FR: 1 /35 and a native IPv6
connectivity.
>
> * NextGenCollective
> IPv6 research
> http://www.nextgencollective.net/
> NGC[NextGen Collective] provides IPv6-over-IPv4 tunnels to people all
> over the world.
> ASpath-tree: http://www.nextgencollective.net/bgp4/ (AS65526 is my old
> private ASN)
> => NDSoftware provide to NextGenCollective: 2 /44, 1 /40, 1 /36 and a
> tunnel IPv6 over IPv4 with BGP (full transit).
>
> * IPng.org.uk
> IPv6 tunnel broker
> http://www.ipng.org.uk/
> ASpath-tree: http://www.ipng.org.uk/bgp/bgp-page-complete.html
(AS65526
> is my old private ASN)
> => NDSoftware provide to IPng.org.uk: 1 /44, 1 /40 and a tunnel IPv6
> over IPv4 with BGP (full transit).
>
> * ILS
> Italian Linux Society
> ILS provide IPv6 connectivity to italian user groups and organizations
> experimenting with IPv6.
> ILS host the IPv6 IRC server calvino.freenode.net
> => NDSoftware provide to ILS: 1 /44 and a tunnel IPv6 over IPv4 with
BGP
> (full transit).
>
> * ATI
> A tunisian ISP
> http://www.ipv6net.tn/
> http://www.ipv6net.tn/ipv6-Tunisia.pdf
> => NDSoftware provide to ATI: 1 /44 and a tunnel IPv6 over IPv4 with
BGP
> (full transit). NDSoftware have help ATI for the IPv6 deployement in
> tunisia. ATI plan later to request a pTLA.
>
> * FABIONNE
> A projet for do IPv6 Debian package
> http://debian-ipv6.fabionne.net/
> => NDSoftware provide to FABIONNE: a tunnel IPv6 over IPv4 with BGP
> (full transit) and host a mirror for this projet
> (http://debian-ipv6.mirrors.ndsoftwarenet.com/).
>
> ESMT
> An university in Senegal
> => NDSoftware provide to ESMT: 1 /44 and a tunnel IPv6 over IPv4.
>
>
> Best regards,
>
> Nicolas DEFFAYET
>
> _______________________________________________
> 6bone mailing list
> 6bone@mailman.isi.edu
> http://mailman.isi.edu/mailman/listinfo/6bone
From jeroen@unfix.org Sun Oct 20 16:55:31 2002
From: jeroen@unfix.org (Jeroen Massar)
Date: Sun, 20 Oct 2002 17:55:31 +0200
Subject: Multihoming (WAS: RE: [6bone] pTLA request NDSOFTWARE - review closes 23 October2002)
In-Reply-To:
Message-ID: <006301c27851$20ec5b00$210d640a@unfix.org>
Arien Vijn [mailto:arien+6bone@ams-ix.net] wrote:
> On 20-10-2002 14:19PM, "Jeroen Massar" wrote:
>
> > Gert Doering [mailto:gert@space.net] wrote:
> To solve the neutrality issues, AMS-IX decided to put serious efforts
in a
> proper multi homing solution. We think (hope) that this approach will
be a
> much more positive contribution to the Internet than arguing the
> criticalness of an IXP.
Perfectly put.
People interrested in this subject could/should have a look at:
http://arneill-py.sacramento.ca.us/ipv6mh/
Which is the site of the: fearless IPv6 multihoming solutions developers
"mulsixhomers"
Aka ipv6mh.
They are currently doing very good work in the multihoming field.
Greets,
Jeroen
From tjc@ecs.soton.ac.uk Sun Oct 20 16:59:55 2002
From: tjc@ecs.soton.ac.uk (Tim Chown)
Date: Sun, 20 Oct 2002 16:59:55 +0100
Subject: [6bone] multiple address handling
In-Reply-To:
References: <20021020134256.D36EC4B23@coconut.itojun.org>
Message-ID: <20021020155955.GW10293@login.ecs.soton.ac.uk>
On Sun, Oct 20, 2002 at 05:19:05PM +0200, Arien Vijn wrote:
>
> On 20-10-2002 15:42PM, "itojun@iijlab.net" wrote:
>
> >> Of course part of the problem is the lack of progress of the multi6 WG,
> >> albeit a non-trivial problem to be working on :) The "classic" IPv6
> >> solution for our university is to take two /48's from different providers,
> >> and for all clients to have two global addresses, but the client-side
> >> support for handling the multiple addressing is yet to be resolved.
> >
> > curious: where do you see problems in multiple address handling?
> > i don't really see any, except the lack of ability to switching
> > address pair for TCP (maybe we should use SCTP?).
>
> This is the major problem. Increased reliability is certainly a factor why
> one wants to be multi homed.
This was the major "problem" I had in mind, yes. The issues include:
(1) Configuration of internal site router infrastructure with new prefixes
as new external links are administratively added.
(2) Removal of configuration on site infrastructure for prefixes for links
that get administratively removed (e.g. a change of main provider, both
prefixes are used for a rollover period).
(3) Deprecation for links that are unavailable. The internal site routing
would presumably remain, but the end hosts should be signalled that the
prefix is deprecated. A temporary case of (2).
(4) Renewed availability of a temporarily deprecated prefix.
[Some may say (3) and (4) are the same as (1) and (2)]
(5) Source host src/dst address selection. There has been good work on this.
I'm not clear on how this ties in to default router selection with router
precedences, or the (relatively) new load balancing proposal.
(6) Sustaining TCP connection when (2) or (3) occur. This is something of
a void that SCTP could fill, or maybe some Mobility-like solution. Of
course it only matters for some applications, but it has to be solved.
In theory IPv6 router renumber handles cases (1) and (2). Cases (3) and (4)
could also use router renumbering and RAs combined.
It's (6) that's the fuzzy area.
I think the standard renumbering scenario that Christian Huitema(?) presented
a while ago assumed orderly renumbering with provider change, rather than
sudden link loss.
Then of course there's the side issues of DNS, firewalls, etc for new
prefixes introduced.
I'd be interested in seeing a case study of a site that has all the above
working and 3 or more layers of internal routers (which we have), not
just a single multihomed router. I had assumed multi6 was stalled because
this hadn't been shown, not because it has :)
Tim
From hansolofalcon@worldnet.att.net Sun Oct 20 17:49:30 2002
From: hansolofalcon@worldnet.att.net (Gregg C Levine)
Date: Sun, 20 Oct 2002 12:49:30 -0400
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <000f01c27850$e3fcaf60$f356580c@who>
Message-ID: <000001c27858$a99589c0$b159580c@who>
Hello from Gregg C Levine
Excuse me. I meant to say, John Fraizer, as the name that I agreed with.
My e-mail client, yes Outlook, goofed again. Everything else in that
message is still a valid point. John, don't be insulted by my program's
goofs.
-------------------
Gregg C Levine hansolofalcon@worldnet.att.net
------------------------------------------------------------
"The Force will be with you...Always." Obi-Wan Kenobi
"Use the Force, Luke." Obi-Wan Kenobi
(This company dedicates this E-Mail to General Obi-Wan Kenobi )
(This company dedicates this E-Mail to Master Yoda )
> -----Original Message-----
> From: 6bone-admin@mailman.isi.edu [mailto:6bone-admin@mailman.isi.edu]
On
> Behalf Of Gregg C Levine
> Sent: Sunday, October 20, 2002 11:54 AM
> To: 'Nicolas DEFFAYET'
> Cc: '6bone Mail List'
> Subject: RE: [6bone] pTLA request NDSOFTWARE - review closes 23
October
> 2002
>
> Hello from Gregg C Levine
> All very nice. But it still doesn't answer any of John Fraize's many
> questions. In fact, I'd say it looks slick. But it won't work, nor
will
> it stack up. Have you studied completely the man pages behind your
> server's software, on each component? Some of them are relevant, and
> aren't just put there for the machine's sake. I'm sorry Nicholas, but
I
> still agree with John Fraize, and everyone else, and disagree with
you.
> -------------------
> Gregg C Levine hansolofalcon@worldnet.att.net
> ------------------------------------------------------------
> "The Force will be with you...Always." Obi-Wan Kenobi
> "Use the Force, Luke." Obi-Wan Kenobi
> (This company dedicates this E-Mail to General Obi-Wan Kenobi )
> (This company dedicates this E-Mail to Master Yoda )
>
>
>
> > -----Original Message-----
> > From: 6bone-admin@mailman.isi.edu
[mailto:6bone-admin@mailman.isi.edu]
> On
> > Behalf Of Nicolas DEFFAYET
> > Sent: Sunday, October 20, 2002 9:44 AM
> > To: Gregg C Levine
> > Cc: 6bone Mail List
> > Subject: RE: [6bone] pTLA request NDSOFTWARE - review closes 23
> October
> > 2002
> >
> > On Sun, 2002-10-20 at 00:24, Gregg C Levine wrote:
> >
> > Hello Gregg,
> >
> > > I have been lurking on this list, for a good number of years now.
> > > Sometimes even posting a comment, or a gripe. This is more along
the
> > > lines of both. I have been monitoring the traffic discussing
> > > NDSOFTWARE's request. Both finding the original message to Bob
Fink,
> and
> > > the list, and everything. Sorry Mr. Deffayet, I disagree. For one,
> you
> > > do need to spell out who will be using the service. Is it for your
> > > company? Yourself? What? Who, even? Unless you can spell out
neatly
> the
> > > answers to my questions, I am inclined to agree with everyone
else.
> I am
> > > also agreeing with the people I have disagreed with early on. I
> might
> > > also, add, even Master Yoda's methods of speaking isn't that
> confusing.
> >
> > Who will be using the service:
> >
> > - NDSoftware (my company)
> >
> > - many projects (here a list of main projects):
> >
> > * IPv6-FR
> > A non profit organisation for the developement of IPv6 in France
> > IPv6-FR run a tunnel broker and have currently 200 users, each user
> have
> > a /48
> > => NDSoftware provide to IPv6-FR: 1 /35 and a native IPv6
> connectivity.
> >
> > * NextGenCollective
> > IPv6 research
> > http://www.nextgencollective.net/
> > NGC[NextGen Collective] provides IPv6-over-IPv4 tunnels to people
all
> > over the world.
> > ASpath-tree: http://www.nextgencollective.net/bgp4/ (AS65526 is my
old
> > private ASN)
> > => NDSoftware provide to NextGenCollective: 2 /44, 1 /40, 1 /36 and
a
> > tunnel IPv6 over IPv4 with BGP (full transit).
> >
> > * IPng.org.uk
> > IPv6 tunnel broker
> > http://www.ipng.org.uk/
> > ASpath-tree: http://www.ipng.org.uk/bgp/bgp-page-complete.html
> (AS65526
> > is my old private ASN)
> > => NDSoftware provide to IPng.org.uk: 1 /44, 1 /40 and a tunnel IPv6
> > over IPv4 with BGP (full transit).
> >
> > * ILS
> > Italian Linux Society
> > ILS provide IPv6 connectivity to italian user groups and
organizations
> > experimenting with IPv6.
> > ILS host the IPv6 IRC server calvino.freenode.net
> > => NDSoftware provide to ILS: 1 /44 and a tunnel IPv6 over IPv4 with
> BGP
> > (full transit).
> >
> > * ATI
> > A tunisian ISP
> > http://www.ipv6net.tn/
> > http://www.ipv6net.tn/ipv6-Tunisia.pdf
> > => NDSoftware provide to ATI: 1 /44 and a tunnel IPv6 over IPv4 with
> BGP
> > (full transit). NDSoftware have help ATI for the IPv6 deployement in
> > tunisia. ATI plan later to request a pTLA.
> >
> > * FABIONNE
> > A projet for do IPv6 Debian package
> > http://debian-ipv6.fabionne.net/
> > => NDSoftware provide to FABIONNE: a tunnel IPv6 over IPv4 with BGP
> > (full transit) and host a mirror for this projet
> > (http://debian-ipv6.mirrors.ndsoftwarenet.com/).
> >
> > ESMT
> > An university in Senegal
> > => NDSoftware provide to ESMT: 1 /44 and a tunnel IPv6 over IPv4.
> >
> >
> > Best regards,
> >
> > Nicolas DEFFAYET
> >
> > _______________________________________________
> > 6bone mailing list
> > 6bone@mailman.isi.edu
> > http://mailman.isi.edu/mailman/listinfo/6bone
>
> _______________________________________________
> 6bone mailing list
> 6bone@mailman.isi.edu
> http://mailman.isi.edu/mailman/listinfo/6bone
From pekkas@netcore.fi Sun Oct 20 17:58:12 2002
From: pekkas@netcore.fi (Pekka Savola)
Date: Sun, 20 Oct 2002 19:58:12 +0300 (EEST)
Subject: sTLA alloc policies [Re: [6bone] pTLA request NDSOFTWARE - review
closes 23 October 2002]
In-Reply-To: <20021020153504.GV10293@login.ecs.soton.ac.uk>
Message-ID:
On Sun, 20 Oct 2002, Tim Chown wrote:
> On Sun, Oct 20, 2002 at 05:36:10PM +0300, Pekka Savola wrote:
> >
> > Speaking of which, I'd be really interested in knowing how Internet
> > Software Consortium is going to fill the "200 customer rule":
>
> Or small to medium sized NRENs in Europe. I spoke last week to someone from
> one of the smallest European research networks who felt he couldn't get a
> SubTLA for that reason (not 200 universities in the country).
Yep, with strict interpretation this is a huge problem. But not in
practise, there's flexibility for those that need it; for example,
NORDUnet, the research network transit for Nordics, has about 5 customers
and no hope of getting more (or having to give addresses to any of these).
And it just got an sTLA a week or two ago :-).
But even if you don't have customers, you have to have these addresses if
you participate (_really_) in the DFZ.
> This isn't
> (I hope) the type of organisation the new rules are trying to exclude. (in
> other ways the new rules are very open, which may lead to (increased)
> 6bonisation of the 2001: space...
My worry as well.. I'm really curious about the ISC stuff, and I really
hope ARIN will not just de-facto ignore the particular point of the rules.
--
Pekka Savola "Tell me of difficulties surmounted,
Netcore Oy not those you stumble over and fall"
Systems. Networks. Security. -- Robert Jordan: A Crown of Swords
From nicolas.deffayet@ndsoftware.net Sun Oct 20 18:07:57 2002
From: nicolas.deffayet@ndsoftware.net (Nicolas DEFFAYET)
Date: 20 Oct 2002 19:07:57 +0200
Subject: [6bone] FAQ about pTLA request NDSOFTWARE - review closes 23 October 2002
Message-ID: <1035133678.4729.343.camel@wks1.fr.corp.ndsoftware.com>
Here a FAQ, no more reply on this list about "pTLA request NDSOFTWARE -
review closes 23 October 2002".
NDSoftware pTLA request is fully compliant with RFC2772.
http://mailman.isi.edu/pipermail/6bone/2002-October/006364.html
Don't try to find a bug, there is no bugs.
If you are jealous, search another victim.
Please respect the Bob Fink's email:
----------------------------------------------------------------------->
6bone Folk,
Please keep the discussion from getting personal, or becoming defamatory
or
using swear words.
Many folks watch how we carry out our 6bone business; it is important
that
we remain open, objective and willing to hear all sides.
Thanks,
Bob
----------------------------------------------------------------------->
---
What's NDSoftware ?
Founded in 2000 by Nicolas DEFFAYET, NDSoftware at the base software
publisher offers a wide range of products and services for personal and
business.
We work a lot in the creation of software, consulting and creation of
website.
We develop an ISP activity for offer to our customers a complete
solution (if a customer order the creation of a software for its
customers management, we want offer additional services like the
creation of the website linked with the software of customers
management, the hosting of this website,...)
---
Why NDSoftware don't have a website ?
For the moment, we don't need a website for find new customers and do
our business. Do marketing stuff is not a priority for us, because we
get new customers with our actual customers.
The NDSoftware website is under developpement since many mounths. Our
website will be not a vulgar website with 20 static pages about our
activites. We will manage our client with it, automatize all
administrative tasks (invoice for exemple),... Our aim is develop more
services without more staff.
---
What about NDSoftware & IPv6 ?
NDSoftware work on 6bone since january 2001. We do a lot of tests for
prepare our ISP activity and at the same time we help the IPv6
community. We provide IPv6 connectivity and our help to many projects.
---
Does NDSoftware plan do commercial activities on 6bone ?
No, of course !
All our IPv6 services are free in production quality but without any
guarantee.
We will do commercial activities only with a sTLA from the RIPE.
---
Is Nicolas DEFFAYET is a kid without brain who destroy 6bone ?
I'm Nicolas DEFFAYET, i have 4 years, i got every day to school, and i
play with my FisherPrice routers and i destroy all 6bone.
---
Why NDSoftware collecting tunnels and BGP sessions ?
It's for do a lot of tests.
---
Does NDSoftware have a bad routing ?
No, we use MED.
Our MED for tunnels:
500: - 10 ms
510: 10 - 25 ms
520: 25 - 50 ms
530: 50 - 100 ms
540: + 100 ms
A lot of pTLA and sTLA don't use MED, we use MED for have a good
quality.
We use too "bgp always-compare-med" of course.
---
Does NDSoftware is the source of "ghost AS_pathes" ?
No, read archive of the 6bone mailing-list, you will find all details
about this problem.
---
Why there is the same phone number on all contact of NDSoftware's whois
?
There is a common phone contact for a best manegement.
---
Does NDSoftware staff have 24x7 physical access to the equipment ?
Yes of course !
---
NDSoftware routers have many BGP peer down, why ?
We do a migration (for the moment only parcr2.fr.ndsoftwarenet.net and
route-server.ndsoftwarenet.net), we change AS65526 to AS25358.
---
What's NDSoftware "potential user community" ?
- NDSoftware (my company)
- many projects (here a list of main projects):
* IPv6-FR
A non profit organisation for the developement of IPv6 in France
IPv6-FR run a tunnel broker and have currently 200 users, each user have
a /48
=> NDSoftware provide to IPv6-FR: 1 /35 and a native IPv6 connectivity.
* NextGenCollective
IPv6 research
http://www.nextgencollective.net/
NGC[NextGen Collective] provides IPv6-over-IPv4 tunnels to people all
over the world.
ASpath-tree: http://www.nextgencollective.net/bgp4/ (AS65526 is my old
private ASN)
=> NDSoftware provide to NextGenCollective: 2 /44, 1 /40, 1 /36 and a
tunnel IPv6 over IPv4 with BGP (full transit).
* IPng.org.uk
IPv6 tunnel broker
http://www.ipng.org.uk/
ASpath-tree: http://www.ipng.org.uk/bgp/bgp-page-complete.html (AS65526
is my old private ASN)
=> NDSoftware provide to IPng.org.uk: 1 /44, 1 /40 and a tunnel IPv6
over IPv4 with BGP (full transit).
* ILS
Italian Linux Society
ILS provide IPv6 connectivity to italian user groups and organizations
experimenting with IPv6.
ILS host the IPv6 IRC server calvino.freenode.net
=> NDSoftware provide to ILS: 1 /44 and a tunnel IPv6 over IPv4 with BGP
(full transit).
* ATI
A tunisian ISP
http://www.ipv6net.tn/
http://www.ipv6net.tn/ipv6-Tunisia.pdf
=> NDSoftware provide to ATI: 1 /44 and a tunnel IPv6 over IPv4 with BGP
(full transit). NDSoftware have help ATI for the IPv6 deployement in
tunisia. ATI plan later to request a pTLA.
* FABIONNE
A projet for do IPv6 Debian package
http://debian-ipv6.fabionne.net/
=> NDSoftware provide to FABIONNE: a tunnel IPv6 over IPv4 with BGP
(full transit) and host a mirror for this projet
(http://debian-ipv6.mirrors.ndsoftwarenet.com/).
ESMT
An university in Senegal
=> NDSoftware provide to ESMT: 1 /44 and a tunnel IPv6 over IPv4.
We do many actions in IPv6 research, we created FNIX6 (French
International Internet Exchange IPv6, http://www.fnix6.net/), we host
many mirrors available in IPv6, we created ftp://ftp.openipv6.com/ (a
FTP with a lot of IPv6 stuff).
---
Why NDSoftware need a pTLA ?
A lot of peers filter our /32 because it's not a pTLA.
We want a pTLA for can announce without any problems our network, don't
break the IPv6 aggregation and be independant of a upstream (we don't
want be down because our upstream is down).
All ISP/company/project who provide IPs to another ISP/company/project
and have many upstream MUST have a pTLA/sTLA.
In http://www.6bone.net/6bone_pTLA_list.html, a lot of pTLA aren't used
or are used only for a /48 and/or have only one upstream.
An exemple: MOTOROLA-LABS, have only one upstream. Do you think that
they need a pTLA ? If their upstream is down, this pTLA is not anymore
announced. MOTOROLA-LABS don't provide IPs and a /48 is enough for their
activity...
NDSoftware provide IPs to another ISP/company/project and have many
upstream...
---
http://noc.ndsoftwarenet.com have many 404 errors pages in "Documents"
submenu...
Yes, we know, it will be fixed soon.
---
Does NDSoftware plan to be LIR at the RIPE ?
Yes, it's planned.
We will be LIR when our ISP activity will be stable.
---
Why NDSoftware will close peering with private ASN ?
We have 101 BGP4+ peers, our current routers are full (zebra is very
unstable if we add new peer) and we want get new peer with other
pTLA/sTLA that we can't get with our old private ASN.
We have choose to delete all peers with private ASN for free BGP session
on our routers for this new peers.
We don't delete peers with private ASN because "private ASN sucks", we
keep peering with important private ASN like NextGenCollective or
IPNG-UK (this 2 projects projet provide a lot of tunnels to users). I
understand their status, it's why i keep peering with them.
We will try to find a solution before the 23th October for keep peer
with all private ASN.
---
What's the address 57 rue du president Wilson, 92300 Levallois-Perret,
France ?
It's the postal address of NDSoftware and IPv6-FR.
---
What's IPv6-FR ?
IPv6-FR is a non profit organisation for the developement of IPv6 in
France.
---
What's FNIX6 ?
FNIX6 (French National Internet Exchange IPv6) is an IPv6 Internet
Exchange.
---
I have an anoter question...
Ask to nicolas.deffayet@ndsoftware.net
with a CC to fink@es.net
---
From tvo@EnterZone.Net Sun Oct 20 18:07:35 2002
From: tvo@EnterZone.Net (John Fraizer)
Date: Sun, 20 Oct 2002 13:07:35 -0400 (EDT)
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <000001c27858$a99589c0$b159580c@who>
Message-ID:
On Sun, 20 Oct 2002, Gregg C Levine wrote:
> Hello from Gregg C Levine
> Excuse me. I meant to say, John Fraizer, as the name that I agreed with.
> My e-mail client, yes Outlook, goofed again. Everything else in that
> message is still a valid point. John, don't be insulted by my program's
> goofs.
> -------------------
> Gregg C Levine hansolofalcon@worldnet.att.net
> ------------------------------------------------------------
> "The Force will be with you...Always." Obi-Wan Kenobi
> "Use the Force, Luke." Obi-Wan Kenobi
> (This company dedicates this E-Mail to General Obi-Wan Kenobi )
> (This company dedicates this E-Mail to Master Yoda )
>
No problem Gregg. Everyone hoses the spelling of my name at least
once. ;-) BTW: What a fine example of attention to detail you just made.
---
John Fraizer | High-Security Datacenter Services |
President | Dedicated circuits 64k - 155M OC3 |
EnterZone, Inc | Virtual, Dedicated, Colocation |
http://www.enterzone.net/ | Network Consulting Services |
From hansolofalcon@worldnet.att.net Sun Oct 20 18:11:28 2002
From: hansolofalcon@worldnet.att.net (Gregg C Levine)
Date: Sun, 20 Oct 2002 13:11:28 -0400
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To:
Message-ID: <000201c2785b$bb64d5e0$b159580c@who>
Hello again from Gregg C Levine
Right, thanks. But can I help it, if that guy over there thinks his
"company" is more real then mine, simply because some ISP decided to
register his domain name? I think his pTLA should be denied based on
that suggestion.
-------------------
Gregg C Levine hansolofalcon@worldnet.att.net
------------------------------------------------------------
"The Force will be with you...Always." Obi-Wan Kenobi
"Use the Force, Luke." Obi-Wan Kenobi
(This company dedicates this E-Mail to General Obi-Wan Kenobi )
(This company dedicates this E-Mail to Master Yoda )
> -----Original Message-----
> From: John Fraizer [mailto:tvo@EnterZone.Net]
> Sent: Sunday, October 20, 2002 1:08 PM
> To: Gregg C Levine
> Cc: '6bone Mail List'
> Subject: RE: [6bone] pTLA request NDSOFTWARE - review closes 23
October
> 2002
>
>
>
> On Sun, 20 Oct 2002, Gregg C Levine wrote:
>
> > Hello from Gregg C Levine
> > Excuse me. I meant to say, John Fraizer, as the name that I agreed
with.
> > My e-mail client, yes Outlook, goofed again. Everything else in that
> > message is still a valid point. John, don't be insulted by my
program's
> > goofs.
> > -------------------
> > Gregg C Levine hansolofalcon@worldnet.att.net
> > ------------------------------------------------------------
> > "The Force will be with you...Always." Obi-Wan Kenobi
> > "Use the Force, Luke." Obi-Wan Kenobi
> > (This company dedicates this E-Mail to General Obi-Wan Kenobi )
> > (This company dedicates this E-Mail to Master Yoda )
> >
>
>
> No problem Gregg. Everyone hoses the spelling of my name at least
> once. ;-) BTW: What a fine example of attention to detail you just
made.
>
>
> ---
> John Fraizer | High-Security Datacenter Services |
> President | Dedicated circuits 64k - 155M OC3 |
> EnterZone, Inc | Virtual, Dedicated, Colocation |
> http://www.enterzone.net/ | Network Consulting Services |
>
From nicolas.deffayet@ndsoftware.net Sun Oct 20 18:36:52 2002
From: nicolas.deffayet@ndsoftware.net (Nicolas DEFFAYET)
Date: 20 Oct 2002 19:36:52 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To:
References:
Message-ID: <1035135413.4779.402.camel@wks1.fr.corp.ndsoftware.com>
On Sun, 2002-10-20 at 16:24, John Fraizer wrote:
> OK. I'm still wondering what NDSoftware does. Don't get me wrong. The
> ASPath Tree is slick, the ftp site is handy for some I'm sure, providing
> 6bone connectivity is definately a service but, there has to be something
> going on that actually generates income, otherwise, you're hemorrhaging
> money on colocation and IP transit charges. Generally, when someone forms
> a company, as you state you have done, it is to generate income. To do
> that, you have to offer services that people will purchase. You _can't_
> sell 6bone access and thus-far, every "service" you claim to provide is
> 6bone-centric.
6bone is a research activity for NDSoftware, see my FAQ
(http://mailman.isi.edu/pipermail/6bone/2002-October/006462.html).
> I'm simply looking at the overall health of the 6bone here. If you're
> issued a pTLA and start providing "services" to folks with that address
> space and suddenly, your "company" goes tits-up because you're not able to
> pay your colocation/transit fees (because your "company" isn't actually
> SELLING anything to generate revenue) then not only have you embarrassed
> yourself but, you will have inconvenienced who knows how many other
> people.
We generate revenue, see my FAQ
(http://mailman.isi.edu/pipermail/6bone/2002-October/006462.html).
We have an operational bussiness.
> > All tech contact in NDSoftware's whois have a root access on each
> > routers. They understand v4/v6 routing, unix administration,...
> >
>
> Wow. You're a trusting soul there. SUDU is your friend, Dude. You might
> want to look at the man page for it.
Sudo is very limited.
We trust our technical staff.
> I'm simply looking for you to demonstrate that you or one of your
> employees can properly maintain appropriate records for address
> allocation. Just because the 6bone is experimental does not relieve you,
> the administrator of a network, from the burdon of due diligence. Suppose
> one of your downstreams started a SPAM campaign to v6 connected
> mailservers or started trying to hack into other v6 connected
> systems? How long does it take you to track down the appropriate contact
> information for the source address? Do you have appropriate records to
> provide to law enforcement agencies in the event that you are subpoenaed
> for this type of information?
We have a database with all informations about peering and IPv6
connectivity provided.
> > We have 3 /32, but 1 /32 is enough. We have 3 /32 for have a backup if
> > one of our upstream can't provide us anymore a BGP peering.
> >
>
> Ya, like if they were to say "your peering session is going to die in a
> week because our routers are overloaded with BGP sessions. We've decided
> to drop all of our BGP peers who are using reserved ASNs." -- Something
> like that?
Now, please forgot the problem of delete of peer with private ASN on our
routers.
It's not a world and public problem.
> > > I submit that without your own portable v4 address space for an endpoint
> > > of tunnels, you're at the mercy of your upstreams. If they require you to
> > > renumber, every one of your peers will have to reconfigure their tunnels.
> >
> > Yes, i know.
> >
>
> So, when you went after your ASN, did you try to brow-beat some v4 space
> out of RIPE as well?
Yes, if we start in production NDSoftware Hosting project.
> > > (5) I find this strange. Can you explain it?
> > >
> > > Nice routing loop there. Have you considered: (1) Not having a v6 default on your border
> > > router. (2) Having a connection between your two border routers and running an IGP between them?
> >
> > Ops, fixed.
> >
> > I have forgot to add "ifconfig lo add 3ffe:81f1:2:1::1/64" in the init
> > scripts of parcr1.fr.ndsoftwarenet.net.
>
> Wow. I can't imagine trying to explain that to one of my customers. This
> is all about attention to detail Nicolas. So, you get your own pTLA and
> people start actually listening to and propagating your announcements and
> you "forget" a little thing like applying an access-list or route-map to
> a peering session. Guess what? Your lack of attention to detail does
> more than embarrass you. It can cause service effecting outages for a
> whole ton of OTHER people.
All humains do errors.
You have do too many errors...
> role: IPv6-FR NOC
> address: IPv6-FR
> address: 57 rue du president Wilson
> address: 92300 Levallois-Perret
> address: France
> phone: +33 671887502
>
> role: NDSoftware NOC
> address: NDSoftware
> address: 57 rue du president Wilson
> address: 92300 Levallois-Perret
> address: France
> phone: +33 671887502
>
>
> I'm sorry Nicolas. Providing address space to YOURSELF doesn't
> count! Sheesh!
NDSoftware and IPv6-FR are in the same building but aren't the same
legal organization.
I'm a network administrator for the both.
> > NexGentCollective (http://www.nextgencollective.net/)
> > tunnel broker: 150 users, each user have a /48.
>
> ipv6-site: NEXTGENCOLLECTIVE
> origin: AS65055
> descr: NextGenCollective IPv6 Research Organization
> country: US
> prefix: 3FFE:8271:A020::/44
> prefix: 3FFE:8271:A030::/44
> prefix: 3FFE:8271:B000::/40
> prefix: 3FFE:2C01:1000::/36
>
> Don't you think that a tunnel-broker housed in Wichita, KS, USA would be
> better served by a 6bone pTLA *IN* the USA? Also, with your current
> peering policy change, isn't this site going to get NIXED? I note their
> use of a Reserved ASN.
They don't find any help somewhere...
We don't only provide a block, we provide a small tech support, help in
tunnel and zebra configuration,....
> > > Part of properly maintaining _YOUR_ ipv6-site object is making sure that
> > > you don't reference an object that doesn't exist. If someone is unable or
> > > unwilling to create & maintain an ipv6-site object, do you really feel
> > > that they are a good peering candidate? I certainly don't.
> >
> > They can be a good peering candidate !
> >
> > A whois updated or not don't make the quality of a peering.
>
>
> I SERIOUSLY BEG TO DIFFER! If someone is too damned lazy to create and
> maintain an ipv6-site object, how on earth can you expect them to maintain
> appropriate BGP filters, allocation records, etc, etc, etc? Man, it is
> _OBVIOUS_ that this is a *toy* to you.
>
> By virtue of your ipv6-site object referencing tunnel endpoints that have
> no corresponding ipv6-site object, it is NOT accurate and you (and your
> sites with nonexistant or invalid ipv6-site objects) are in violation of
> RFC2772:
>
> 5. The 6Bone Registry
>
> The 6Bone registry is a RIPE-181 database with IPv6 extensions used
> to store information about the 6Bone, and its sites. The 6bone is
> accessible at:
>
> )
>
> Each 6Bone site MUST maintain the relevant entries in the 6Bone
> registry. In particular, the following object MUST be present for all
> 6Bone leaf sites, pNLAs and pTLAs:
>
> - IPv6-site: site description
>
> - Inet6num: prefix delegation (one record MUST exist for each
> delegation)
>
> - Mntner: contact info for site maintance/administration staff.
>
> Other object MAY be maintained at the discretion of the sites such as
> routing policy descriptors, person, or role objects. The Mntner
> object MUST make reference to a role or person object, but those MAY
> NOT necessarily reside in the 6Bone registry. They can be stored
> within any of the Internet registry databases (ARIN, APNIC, RIPE-NCC,
> etc.)
>
> 6. Guidelines for new sites joining the 6Bone
>
> New sites joining the 6Bone should seek to connect to a transit pNLA
> or a pTLA within their region, and preferably as close as possible to
> their existing IPv4 physical and routing path for Internet service.
> The 6Bone web site at has various information
> and tools to help find candidate 6bone networks.
>
> Any site connected to the 6Bone MUST maintain a DNS server for
> forward name lookups and reverse address lookups. The joining site
> MUST maintain the 6Bone objects relative to its site, as describe in
> section 5.
>
> The upstream provider MUST delegate the reverse address translation
> zone in DNS to the joining site, or have an agreement in place to
> perform primary DNS for that downstream. The provider MUST also
> create the 6Bone registry inet6num object reflecting the delegated
> address space.
>
>
>
>
>
>
> Now, from section 7 of RFC2772, a bit more for you to ponder:
>
> During the entire qualifying period the Applicant must be
> operational providing the following:
>
> a. Fully maintained, up to date, 6Bone Registry entries for their
> ipv6-site inet6num, mntner, and person objects, including each
> tunnel that the Applicant has.
>
>
>
>
> 4. The pTLA Applicant MUST commit to abide by the current 6Bone
> operational rules and policies as they exist at time of its
> application, and agree to abide by future 6Bone backbone
> operational rules and policies as they evolve by consensus of the
> 6Bone backbone and user community.
>
>
>
> Now, since you obviously don't care if your peers maintain their ipv6-site
> objects or even HAVE them for that matter, how is it that you are abiding
> by RFC2772, Section 5?
>
Check all others pTLA request, you have the same problem.
You can't force a peer to register a whois entry...
Our whois is always updated.
From jeroen@unfix.org Sun Oct 20 19:18:49 2002
From: jeroen@unfix.org (Jeroen Massar)
Date: Sun, 20 Oct 2002 20:18:49 +0200
Subject: [6bone] FAQ about pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <1035133678.4729.343.camel@wks1.fr.corp.ndsoftware.com>
Message-ID: <008f01c27865$238df0d0$210d640a@unfix.org>
Nicolas DEFFAYET wrote:
> Here a FAQ, no more reply on this list about "pTLA request
> NDSOFTWARE -
> review closes 23 October 2002".
>
> NDSoftware pTLA request is fully compliant with RFC2772.
> http://mailman.isi.edu/pipermail/6bone/2002-October/006364.html
>
> Don't try to find a bug, there is no bugs.
> If you are jealous, search another victim.
As I said before, I *CANNOT* be jealous at you.
Actually I am pitying you a bit.
You are not a victim, you are now a perfect example
why not everybody should be able to get a TLA.
6bone Community is a PUBLIC place. Not private.
A pTLA request is a PUBLIC thing. Not private.
Thats why I will reply to your "FAQ" in private.
I also wonder why this FAQ isn't on your "noc" website.
> What's NDSoftware ?
Aha software company, just like any other person having a webdesign
company.
And nobody sees them requesting a huge amount of IP space.
Also these 'services' are they going to run from your pTLA -> nopes
because then they would be commercial.
> ---
>
> Is Nicolas DEFFAYET is a kid without brain who destroy 6bone ?
>
> I'm Nicolas DEFFAYET, i have 4 years, i got every day to school, and i
> play with my FisherPrice routers and i destroy all 6bone.
Thank you for answering this question, many people where wondering about
this.
I hope you take 6bone and your 'company' as serious as this.
> Does NDSoftware is the source of "ghost AS_pathes" ?
>
> No, read archive of the 6bone mailing-list, you will find all details
> about this problem.
Please quote the solution of the ghost paths, URL's will do to.
If you know the answer please present it as many people are looking for
it.
> What's NDSoftware "potential user community" ?
>
> - NDSoftware (my company)
And your company needs a /40, nice 'numberplan'.
How many employees does your company have to justify this?
> We do many actions in IPv6 research, we created FNIX6 (French
> International Internet Exchange IPv6, http://www.fnix6.net/), we host
> many mirrors available in IPv6, we created ftp://ftp.openipv6.com/ (a
> FTP with a lot of IPv6 stuff).
Those "ISPs" should request their own TLA.
Your "mirrors" can run out of ONE /48 too.
> A lot of peers filter our /32 because it's not a pTLA.
That's the general rule of the internet.
> We want a pTLA for can announce without any problems our
> network, don't
> break the IPv6 aggregation and be independant of a upstream (we don't
> want be down because our upstream is down).
You might ask your IPv4 uplink to get some IPv6 space for you and use
that.
As you have no own infrastructure you are still depending on your
upstream
even if you do get a TLA.
> In http://www.6bone.net/6bone_pTLA_list.html, a lot of pTLA
> aren't used
> or are used only for a /48 and/or have only one upstream.
>
> An exemple: MOTOROLA-LABS, have only one upstream. Do you think that
> they need a pTLA ? If their upstream is down, this pTLA is not anymore
> announced. MOTOROLA-LABS don't provide IPs and a /48 is
> enough for their
> activity...
What has that to do with anything?
Do you work at MOTOROLA-LABS that you know their internal structure ?
> NDSoftware provide IPs to another ISP/company/project and have many
> upstream...
You got TUNNELS, no native upstreams.
> Why NDSoftware will close peering with private ASN ?
>
> We have 101 BGP4+ peers, our current routers are full (zebra is very
> unstable if we add new peer) and we want get new peer with other
> pTLA/sTLA that we can't get with our old private ASN.
>
> We have choose to delete all peers with private ASN for free
> BGP session
> on our routers for this new peers.
>
> We don't delete peers with private ASN because "private ASN sucks", we
> keep peering with important private ASN like NextGenCollective or
> IPNG-UK (this 2 projects projet provide a lot of tunnels to users). I
> understand their status, it's why i keep peering with them.
>
> We will try to find a solution before the 23th October for keep peer
> with all private ASN.
> FNIX6 (French National Internet Exchange IPv6) is an IPv6 Internet
> Exchange.
National, International? Checking the website it doesn't exist.
And I hope that your personal IX is not experimental.
If it isn't you can't use a pTLA for it.
> I have an anoter question...
>
> Ask to nicolas.deffayet@ndsoftware.net
> with a CC to fink@es.net
See at the top, 6bone is PUBLIC, Internet is PUBLIC.
Also, reread RFC2772.
Greets,
Jeroen
From bortzmeyer@gitoyen.net Sun Oct 20 20:24:35 2002
From: bortzmeyer@gitoyen.net (Stephane Bortzmeyer)
Date: Sun, 20 Oct 2002 21:24:35 +0200
Subject: [6bone] Re: IPv6-only IXP's are absolutely wrong
In-Reply-To: <20021019214631.GE18889@login.ecs.soton.ac.uk>
References: <1035056975.634.2109.camel@wks1.fr.corp.ndsoftware.com> <200210192010.g9JKAfgj024322@ludwigV.sources.org> <20021019214631.GE18889@login.ecs.soton.ac.uk>
Message-ID: <20021020192435.GC15861@nic.fr>
On Sat, Oct 19, 2002 at 10:46:31PM +0100,
Tim Chown wrote
a message of 15 lines which said:
> > You can always go through an existing LIR. Gitoyen could certainly do it, even if I find the idea of an IPv6-only IXP absolutely wrong.
...
> I'm curious as to why you think this?
Because of two reasons, one temporary and the other more permanent.
1) There is very few actual IPv6 traffic and it is mostly
ICMP. Setting up an IXP and making people pay to connect
(unless it is located in a well-populated data center), just for this
small amount of traffic, seems a financial mistake.
2) Most IXP work at level 2. They are a (several) switch(es) and a set of IP
addresses. Since IPv4 and IPv6 can happily coexist on the same network
(like we all do on our links), I see no reason to set up two different IXP.
> Is the UK6X heading down the wrong path?
I believe so, but I will ask the same question: why do you set up an
IPv6-only a real IPv6 IXP exchange point?
From pim@ipng.nl Sun Oct 20 20:40:02 2002
From: pim@ipng.nl (Pim van Pelt)
Date: Sun, 20 Oct 2002 21:40:02 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <1035064533.606.2187.camel@wks1.fr.corp.ndsoftware.com>
References: <003701c277b9$465d4d10$210d640a@unfix.org> <1035064533.606.2187.camel@wks1.fr.corp.ndsoftware.com>
Message-ID: <20021020194002.GA17735@bfib.colo.bit.nl>
| > You are trying and wanting to do constructive things for IPv6.
| > But you really really have to realize that NOT everybody can have a
| > pTLA.
|
| Why IPNG.NL have a pTLA and a sTLA ?
This is quite amazing. You are mistaken, IPng is not even an ISP. It's a
personal hobby project of mine that is running at Intouch.
| Please justify this.
IPng does not have a pTLA.
The operating ISP, called Intouch (NV) in The Netherlands (AS8954),
requested a pTLA several years ago. IPng is a project that is running
within Intouch which requires IPv6 connectivity and they get this
statically routed from biscuit.intouch.net (connected to the AMS-IX
shared medium with 10baseT).
IPng does not have an sTLA.
The operating ISP, Intouch NV, after some 9 months of operational
experience with the pTLA, thought themselves ready to offer commercial
services with IPv6. They then requested (and were allocated) the second
IPv6 allocation for .nl (the first being our NREN).
Note that there is a large difference in the way Intouch handled things
at the time, and the way you are proceding today.
groet,
Pim
--
---------- - - - - -+- - - - - ----------
Pim van Pelt Email: pim@ipng.nl
http://www.ipng.nl/ IPv6 Deployment
-----------------------------------------------
From pim@ipng.nl Sun Oct 20 21:14:45 2002
From: pim@ipng.nl (Pim van Pelt)
Date: Sun, 20 Oct 2002 22:14:45 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <5.1.0.14.0.20021016061927.02891008@imap2.es.net>
References: <5.1.0.14.0.20021016061927.02891008@imap2.es.net>
Message-ID: <20021020201445.GB17735@bfib.colo.bit.nl>
On Wed, Oct 16, 2002 at 06:25:47AM -0700, Bob Fink wrote:
| 6bone Folk,
Dear Bob,
| NDSOFTWARE has requested a pTLA allocation and I find their request fully
| compliant with RFC2772. The open review period for this will close 23
| October 2002. Please send your comments to me or the list.
In short: I oppose to this request also. Please read on.
It's Sun Oct 20 when I'm writing this mail. I have read some 60 mails,
most of which stick to the topic of DEFFAYETs request. Reading through
them, I find it refreshing to see that many of the regular posters of
this mailinglist seem to agree that it is not yet time to allocate a
pTLA to NDSoftware or Nicolas.
Last month, I saw a small thread on the lir-wg@ripe.net mailinglist,
where Nicolas complained in this public forum about the fact that the
NCC did not grant him an AS number request. NCC rectified their prior
decision and allocated an AS number to NDSoftware. I still had my doubts.
An autonomous system, in my book, is a set of routers which have the
same routing policies implemented. Somehow, the routers in NDSoftware
are not interconnected with private circuits (sdh/atm/ethernet). This
means the IPv4 cloud at NDSoftware consists of one or more IPv4
cloudlets (consisting of one machine each).
I myself did check out the websites (fnix6 and ndsoftware) and did see
mutitudes of 404's. I thought them to be typical of Nicolas' methodology.
Connecting them together with IPv6 tunnels does not seem like a big
deal. The current customer base is predominantly non-french/parisian.
MEDs are multiple exit discriminators, used to steer traffic into some
specific router if you have multiple BGP sessions between your AS and
theirs.
IPng does not have a pTLA nor an sTLA. They are simply a project running
at the commercial ISP Intouch NV (AS8954) and they have a statically
(and natively) routed /32 out of the Intouch pTLA. It is irrelevant and
does not have to be dragged into the discussion just because Jeroen
helps administer that project.
I would like to personally thank John F (razier, look I spelled it
correctly :) for his long list of terribly useful questions (posted in
his mail with id g9K8gLo08725). I find DEFFAYETs replies to this mail
defensive and evasive to say the least.
I am not supportive of this request and protest against it because ever
since I've seen Nicolas join 'the scene', I have seen him push his work
forward using unorthodox methods. The most important one, is insisting
on using AS65526 because -- as he said so himself -- RIPE refused to
give him his own AS to run out of. During this phase, DEFFAYET kept on
introducing more instabilities into the 'Net, as also made public at
RIPE42 by Gert Doering.
The reflection of the current setup at NDSoftware in the whois database
is crappy. It seems like DEFFAYET was collecting /32s from other pTLA
holders. We (the 6bone community) have requested him to clean up his act
on numerous occasions and he simply refused.
DEFFAYET has his own set of rules that he wishes to play by. I don't
think (and urge you to note this), that the last point of rfc2772 states
that the pTLA requestor will obide by the best common practice and
cooperate with other 6BONE members. I don't believe that this was the
case in the past, nor that it will be the case in the future.
Please reconsider the pTLA allocation to NDSoftware.
--
---------- - - - - -+- - - - - ----------
Pim van Pelt Email: pim@ipng.nl
http://www.ipng.nl/ IPv6 Deployment
-----------------------------------------------
From Daniel Austin"
Message-ID: <009901c27885$1e2945c0$611c08d9@kewlio.net>
Hi Marcel,
One of our peers, Dolphins (Switzerland) may be able to provide you with a
block of IP's
http://www.ipv6.as8758.net
With Thanks,
Daniel Austin,
Managing Director,
kewlio.net Limited.
----- Original Message -----
From: "Marcel Stutz"
To: <6bone@mailman.isi.edu>
Sent: Saturday, October 19, 2002 11:51 PM
Subject: [6bone] How i get IP V6 Addresses ?
> What is the best way to get own IP V6 Addresses for a smale IRC Network in
> switzerland ?
>
> What i need to pay ?
>
> Thanks Marcel
>
>
>
>
> _______________________________________________
> 6bone mailing list
> 6bone@mailman.isi.edu
> http://mailman.isi.edu/mailman/listinfo/6bone
>
From fink@es.net Sun Oct 20 23:58:30 2002
From: fink@es.net (Bob Fink)
Date: Sun, 20 Oct 2002 15:58:30 -0700
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October
2002
In-Reply-To: <20021020201445.GB17735@bfib.colo.bit.nl>
References: <5.1.0.14.0.20021016061927.02891008@imap2.es.net>
<5.1.0.14.0.20021016061927.02891008@imap2.es.net>
Message-ID: <5.1.0.14.0.20021020155559.030c0c50@imap2.es.net>
Pim,
Thanks for your comments. They (and others) will be taken into account
before any pTLA is allocated.
I specifically try to stay quiet during the review phase to let folks make
their cases. Comments and reasoned arguments, on both sides, do make a
difference in the outcome. It is a collective process.
Thanks,
Bob
===
At 10:14 PM 10/20/2002 +0200, Pim van Pelt wrote:
>On Wed, Oct 16, 2002 at 06:25:47AM -0700, Bob Fink wrote:
>| 6bone Folk,
>Dear Bob,
>
>| NDSOFTWARE has requested a pTLA allocation and I find their request fully
>| compliant with RFC2772. The open review period for this will close 23
>| October 2002. Please send your comments to me or the list.
>
>In short: I oppose to this request also. Please read on.
>
>It's Sun Oct 20 when I'm writing this mail. I have read some 60 mails,
>most of which stick to the topic of DEFFAYETs request. Reading through
>them, I find it refreshing to see that many of the regular posters of
>this mailinglist seem to agree that it is not yet time to allocate a
>pTLA to NDSoftware or Nicolas.
>
>Last month, I saw a small thread on the lir-wg@ripe.net mailinglist,
>where Nicolas complained in this public forum about the fact that the
>NCC did not grant him an AS number request. NCC rectified their prior
>decision and allocated an AS number to NDSoftware. I still had my doubts.
>
>An autonomous system, in my book, is a set of routers which have the
>same routing policies implemented. Somehow, the routers in NDSoftware
>are not interconnected with private circuits (sdh/atm/ethernet). This
>means the IPv4 cloud at NDSoftware consists of one or more IPv4
>cloudlets (consisting of one machine each).
>
>I myself did check out the websites (fnix6 and ndsoftware) and did see
>mutitudes of 404's. I thought them to be typical of Nicolas' methodology.
>Connecting them together with IPv6 tunnels does not seem like a big
>deal. The current customer base is predominantly non-french/parisian.
>
>MEDs are multiple exit discriminators, used to steer traffic into some
>specific router if you have multiple BGP sessions between your AS and
>theirs.
>
>IPng does not have a pTLA nor an sTLA. They are simply a project running
>at the commercial ISP Intouch NV (AS8954) and they have a statically
>(and natively) routed /32 out of the Intouch pTLA. It is irrelevant and
>does not have to be dragged into the discussion just because Jeroen
>helps administer that project.
>
>I would like to personally thank John F (razier, look I spelled it
>correctly :) for his long list of terribly useful questions (posted in
>his mail with id g9K8gLo08725). I find DEFFAYETs replies to this mail
>defensive and evasive to say the least.
>
>I am not supportive of this request and protest against it because ever
>since I've seen Nicolas join 'the scene', I have seen him push his work
>forward using unorthodox methods. The most important one, is insisting
>on using AS65526 because -- as he said so himself -- RIPE refused to
>give him his own AS to run out of. During this phase, DEFFAYET kept on
>introducing more instabilities into the 'Net, as also made public at
>RIPE42 by Gert Doering.
>
>The reflection of the current setup at NDSoftware in the whois database
>is crappy. It seems like DEFFAYET was collecting /32s from other pTLA
>holders. We (the 6bone community) have requested him to clean up his act
>on numerous occasions and he simply refused.
>
>DEFFAYET has his own set of rules that he wishes to play by. I don't
>think (and urge you to note this), that the last point of rfc2772 states
>that the pTLA requestor will obide by the best common practice and
>cooperate with other 6BONE members. I don't believe that this was the
>case in the past, nor that it will be the case in the future.
>
>Please reconsider the pTLA allocation to NDSoftware.
>--
>---------- - - - - -+- - - - - ----------
>Pim van Pelt Email: pim@ipng.nl
>http://www.ipng.nl/ IPv6 Deployment
>-----------------------------------------------
From michel@arneill-py.sacramento.ca.us Mon Oct 21 00:02:08 2002
From: michel@arneill-py.sacramento.ca.us (Michel Py)
Date: Sun, 20 Oct 2002 16:02:08 -0700
Subject: [6bone] multiple address handling
Message-ID: <2B81403386729140A3A899A8B39B046405E399@server2000>
> Tim Chown wrote:
> The danger is that all companies will want this independence
> and for the same reason demand a pTLA/SubTLA. It's certainly
> true for our university, which has a /48. Given we offer
> IPv6 remote access, should we be allowed a /32 to offer static
> /48 "site" IPv6 prefixes to any university member wanting
> connectivity?
I all depends if you consider these customers or not. Some do, some
don't, some have or will request a pTLA or become LIRs by setting up a
company that functions as an ISP.
> Of course part of the problem is the lack of progress of the
> multi6 WG, albeit a non-trivial problem to be working on :)
This is purely a political issue. If the IETF wanted multi6 to produce a
solution, multi6 would have produced one by now. Multi6 simply is in the
same batch as ngtrans, the 6bone and other stuff that has been closed,
is being closed or is slated to be closed soon.
> The "classic" IPv6 solution for our university is to take two
> /48's from different providers, and for all clients to have
> two global addresses, but the client-side support for handling
> the multiple addressing is yet to be resolved.
There are multiple issues associated with doing this. It is acceptable
for a home/soho setup, but there are not too many people that are
willing to run a large setup with this, especially in the total absence
of standards.
Michel.
From bmanning@ISI.EDU Mon Oct 21 01:13:47 2002
From: bmanning@ISI.EDU (Bill Manning)
Date: Sun, 20 Oct 2002 17:13:47 -0700 (PDT)
Subject: [6bone] Re: IPv6-only IXP's are absolutely wrong
In-Reply-To: <20021020192435.GC15861@nic.fr> from Stephane Bortzmeyer at "Oct 20, 2 09:24:35 pm"
Message-ID: <200210210013.g9L0Dlt25633@boreas.isi.edu>
% 2) Most IXP work at level 2. They are a (several) switch(es) and a set of IP
% addresses. Since IPv4 and IPv6 can happily coexist on the same network
% (like we all do on our links), I see no reason to set up two different IXP.
%
% > Is the UK6X heading down the wrong path?
%
% I believe so, but I will ask the same question:
it is true that for L2 fabric, it can be hard to prevent IPv6 or IPv4
from using the fabric. There are reasons to try and split them however.
This is not the right thread to take on that topic though.
If this is of interest, send a note and I'll forward on the rational.
--
--bill
From tvo@EnterZone.Net Mon Oct 21 03:14:20 2002
From: tvo@EnterZone.Net (John Fraizer)
Date: Sun, 20 Oct 2002 22:14:20 -0400 (EDT)
Subject: [6bone] FAQ about pTLA request NDSOFTWARE - review closes 23
October 2002
In-Reply-To: <1035133678.4729.343.camel@wks1.fr.corp.ndsoftware.com>
Message-ID:
On 20 Oct 2002, Nicolas DEFFAYET wrote:
> Here a FAQ, no more reply on this list about "pTLA request NDSOFTWARE -
> review closes 23 October 2002".
>
> NDSoftware pTLA request is fully compliant with RFC2772.
> http://mailman.isi.edu/pipermail/6bone/2002-October/006364.html
>
> Don't try to find a bug, there is no bugs.
> If you are jealous, search another victim.
>
Sorry Nicolas. It doesn't work that way. You don't get to say "Don't
discuss my application any more."
I don't know about "bugs" in your application but, there are several
compliance issues, which I have already pointed out in previous emails.
As for being jealous, you have got to be smoking crack if you think that
I'm jealous. We are _NOT_ trying to make you a victim but rather prevent
the participants in the DFZ from becoming a victim of yours.
>
> Please respect the Bob Fink's email:
>
> ----------------------------------------------------------------------->
>
> 6bone Folk,
>
> Please keep the discussion from getting personal, or becoming defamatory
> or
> using swear words.
>
> Many folks watch how we carry out our 6bone business; it is important
> that
> we remain open, objective and willing to hear all sides.
>
>
> Thanks,
>
> Bob
>
Nicolas, Bob is a big boy. I'm sure that if he feels that someone is not
respecting his wishes and following the guidance he provides to the list,
he will deal with it himself. He does not need for you to play "Master at
Arms" from France.
> We develop an ISP activity for offer to our customers a complete
> solution (if a customer order the creation of a software for its
> customers management, we want offer additional services like the
> creation of the website linked with the software of customers
> management, the hosting of this website,...)
Nicolas, this flies in the face of what you posted previously.
In a previous email, listed at
http://mailman.isi.edu/pipermail/6bone/2002-October/006348.html you write:
"=> How get an ASN for our new IPv6 network ?
(we don't want do IPv4 network)"
Which one is it Nicolas? Do you want to offer a complete solution, or do
you want to do v6 only with no v4?
> ---
>
> Why NDSoftware don't have a website ?
>
> For the moment, we don't need a website for find new customers and do
> our business. Do marketing stuff is not a priority for us, because we
> get new customers with our actual customers.
OK. If you insist that you actually have customers, you have
customers. Since you use word-of-mouth as your sole marketing technique,
I'm sure that you have at least ONE customer who has given you permission
to use them as a reference. Who would they be?
> The NDSoftware website is under developpement since many mounths. Our
> website will be not a vulgar website with 20 static pages about our
> activites. We will manage our client with it, automatize all
> administrative tasks (invoice for exemple),... Our aim is develop more
> services without more staff.
OK. So this is a matter of the cobblers children going barefoot
huh? Your "company," as one of it's commercial, revenue generating
activities, to quote you, "creation of the website." Your customers
simply take your word for you that you can complete a project though and
none of them express any concern or pay any attention to the fact that
"since many months" your own ?corporate? website is defunct.
I was born at night Nicolas but, it wasn't LAST NIGHT.
> ---
>
> Does NDSoftware plan do commercial activities on 6bone ?
>
> No, of course !
>
> All our IPv6 services are free in production quality but without any
> guarantee.
> We will do commercial activities only with a sTLA from the RIPE.
OK. And how do you suppose that you're going to qualify for an sTLA?
>
> Is Nicolas DEFFAYET is a kid without brain who destroy 6bone ?
>
> I'm Nicolas DEFFAYET, i have 4 years, i got every day to school, and i
> play with my FisherPrice routers and i destroy all 6bone.
>
I hope you don't mind but, I'm going to use that in my signature line from
now on. It's so funny!
> ---
>
> Why NDSoftware collecting tunnels and BGP sessions ?
>
> It's for do a lot of tests.
>
OK. You've got our attention. I'm always interested in gadgets, tests,
new innovations. What kind of tests are you performing? Have you
documented any of them? Have you or do you plan to publish your results
to the public?
> ---
>
> Does NDSoftware have a bad routing ?
>
> No, we use MED.
>
> Our MED for tunnels:
>
> 500: - 10 ms
> 510: 10 - 25 ms
> 520: 25 - 50 ms
> 530: 50 - 100 ms
> 540: + 100 ms
>
Nicolas,
The use of Multi Exit Discriminator does not in and of itself all of the
sudden mean your routing != BAD or sub-optimal.
I'm not saying that your routing IS sub-optimal but, I will certainly say
that I believe you have gone WAY overboard with the number of _UPSTREAM_
tunnels you have. If you had 101 downstream peers, we'd all be jumping
for joy at your request for a pTLA.
> A lot of pTLA and sTLA don't use MED, we use MED for have a good
> quality.
>
> We use too "bgp always-compare-med" of course.
>
The use or lack of use of the MED attribute does NOT in and of itself
equal good routing.
> ---
>
> NDSoftware routers have many BGP peer down, why ?
>
> We do a migration (for the moment only parcr2.fr.ndsoftwarenet.net and
> route-server.ndsoftwarenet.net), we change AS65526 to AS25358.
>
You know Nicolas, the last time I would start doing service effecting
maintenance is right before I knew someone was going to start looking at
our network under a magnifying glass. There are graceful ways to achieve
exactly what you are trying to achieve. You simply have opted not to do
so.
> ---
>
> Why NDSoftware need a pTLA ?
>
> A lot of peers filter our /32 because it's not a pTLA.
> We want a pTLA for can announce without any problems our network, don't
> break the IPv6 aggregation and be independant of a upstream (we don't
> want be down because our upstream is down).
I understand that you _WANT_ a pTLA.
>
> All ISP/company/project who provide IPs to another ISP/company/project
> and have many upstream MUST have a pTLA/sTLA.
Wow. That is a pretty broad statement and does not hold true in
practice.
>
> In http://www.6bone.net/6bone_pTLA_list.html, a lot of pTLA aren't used
> or are used only for a /48 and/or have only one upstream.
>
> An exemple: MOTOROLA-LABS, have only one upstream. Do you think that
> they need a pTLA ? If their upstream is down, this pTLA is not anymore
> announced. MOTOROLA-LABS don't provide IPs and a /48 is enough for their
> activity...
As has already been stated, Motorola-Labs has many, many, MANY
"internal" customers. With several THOUSAND locations, world-wide, I am
quite certain that once their v6 network is fully deployed, they will have
made appropriate use of their pTLA.
There is no requirement that pTLAs have more than one connection into the
backbone. And just so you know, at the pTLA/sTLA - pTLA/sTLA peering
level, we peer in the horizontal plane, not the vertical plane.
>
> NDSoftware provide IPs to another ISP/company/project and have many
> upstream...
>
OK. And this is appropriate justification for a pTLA?
> Does NDSoftware plan to be LIR at the RIPE ?
>
> Yes, it's planned.
> We will be LIR when our ISP activity will be stable.
>
Again, I reference your previous email at
http://mailman.isi.edu/pipermail/6bone/2002-October/006348.html
"=> How get an ASN for our new IPv6 network ?
(we don't want do IPv4 network)"
The lack of a v4 network is going to put a serious hurt on any plans you
have to become an LIR or for that matter, an ISP by many definitions.
> ---
>
> Why NDSoftware will close peering with private ASN ?
>
> We have 101 BGP4+ peers, our current routers are full (zebra is very
> unstable if we add new peer) and we want get new peer with other
> pTLA/sTLA that we can't get with our old private ASN.
>
> We have choose to delete all peers with private ASN for free BGP session
> on our routers for this new peers.
Nicolas, didn't we already point out that you don't have 101 BGP4+
sessions? You _had_ 93 sessions when we counted them this
morning. Believe me. The number is high enough with artificial inflation
on your part.
Now that we have that said, I submit again, that if you are going to be a
viable pTLA, you'll build another router. Good grief! You can build a
KICK-BUTT tunnel endpoint router for $200USD! If you're already too full
to maintain your current peering obligations while bringing on requisite
lateral peers, and you won't build a new router to accomodate additional
peers, just how is it that you're going to make use of your pTLA?
>
> We don't delete peers with private ASN because "private ASN sucks", we
> keep peering with important private ASN like NextGenCollective or
> IPNG-UK (this 2 projects projet provide a lot of tunnels to users). I
> understand their status, it's why i keep peering with them.
>
> We will try to find a solution before the 23th October for keep peer
> with all private ASN.
>
Simple solution. If your router is overloaded, for whatever reason, you
build another one. That's how you build a _network_ of routers. Do you
see a cause and effect relationship here?
>
> I have an anoter question...
>
> Ask to nicolas.deffayet@ndsoftware.net
> with a CC to fink@es.net
>
If I had to venture a guess, I would say that Bob would rather have any
discussions on the List and as such, until instructed otherwise by Bob, I
will continue to post my questions to the list.
---
John Fraizer
EnterZone, Inc
"I'm Nicolas DEFFAYET, i have 4 years, i got every day to school, and i
play with my FisherPrice routers and i destroy all 6bone." - Nicolas
DEFAYET, 20 October 2002.
From tvo@EnterZone.Net Mon Oct 21 03:54:58 2002
From: tvo@EnterZone.Net (John Fraizer)
Date: Sun, 20 Oct 2002 22:54:58 -0400 (EDT)
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <1035135413.4779.402.camel@wks1.fr.corp.ndsoftware.com>
Message-ID:
On 20 Oct 2002, Nicolas DEFFAYET wrote:
> > > All tech contact in NDSoftware's whois have a root access on each
> > > routers. They understand v4/v6 routing, unix administration,...
> > >
> >
> > Wow. You're a trusting soul there. SUDU is your friend, Dude. You might
> > want to look at the man page for it.
>
> Sudo is very limited.
> We trust our technical staff.
Sudu will do anything you tell it to do. That is moot though. I didn't
expect you to take advise from those who have been running production
networks for a decade.
> > week because our routers are overloaded with BGP sessions. We've decided
> > to drop all of our BGP peers who are using reserved ASNs." -- Something
> > like that?
>
> Now, please forgot the problem of delete of peer with private ASN on our
> routers.
>
> It's not a world and public problem.
Nicolas, welcome to the real world. You desire to become an equal peer
with other pTLA/sTLA entities. You want to join the elite Default Free
Zone club. I am happy to say that we _do_ have some standards and those
who don't meet those standards are, in my humble opinion, not welcome.
Your treatment of your CURRENT peers is the only guideline we have to go
on to determine how tactfully you may interact with other DFZ peers.
> > > Ops, fixed.
> > >
> > > I have forgot to add "ifconfig lo add 3ffe:81f1:2:1::1/64" in the init
> > > scripts of parcr1.fr.ndsoftwarenet.net.
> >
> > Wow. I can't imagine trying to explain that to one of my customers. This
> > is all about attention to detail Nicolas. So, you get your own pTLA and
> > people start actually listening to and propagating your announcements and
> > you "forget" a little thing like applying an access-list or route-map to
> > a peering session. Guess what? Your lack of attention to detail does
> > more than embarrass you. It can cause service effecting outages for a
> > whole ton of OTHER people.
>
>
> All humains do errors.
> You have do too many errors...
Sorry Nicolas. I hate to break it to you. You're not going to find an
instance where my lack of attention to detail has caused routing
instability for ANYONE. Yes. I have made errors in my lifetime but, when
it comes to routing, _REAL_ engineers check their work.
>
> > role: IPv6-FR NOC
> > address: IPv6-FR
> > address: 57 rue du president Wilson
> > address: 92300 Levallois-Perret
> > address: France
> > phone: +33 671887502
> >
> > role: NDSoftware NOC
> > address: NDSoftware
> > address: 57 rue du president Wilson
> > address: 92300 Levallois-Perret
> > address: France
> > phone: +33 671887502
> >
> >
> > I'm sorry Nicolas. Providing address space to YOURSELF doesn't
> > count! Sheesh!
>
>
> NDSoftware and IPv6-FR are in the same building but aren't the same
> legal organization.
> I'm a network administrator for the both.
And would this be your home by some strange coincidence?
> > Don't you think that a tunnel-broker housed in Wichita, KS, USA would be
> > better served by a 6bone pTLA *IN* the USA? Also, with your current
> > peering policy change, isn't this site going to get NIXED? I note their
> > use of a Reserved ASN.
>
> They don't find any help somewhere...
>
> We don't only provide a block, we provide a small tech support, help in
> tunnel and zebra configuration,....
They didn't look very hard.
> >
> > Now, since you obviously don't care if your peers maintain their ipv6-site
> > objects or even HAVE them for that matter, how is it that you are abiding
> > by RFC2772, Section 5?
> >
>
> Check all others pTLA request, you have the same problem.
>
> You can't force a peer to register a whois entry...
>
> Our whois is always updated.
(1) We're not talking about other peoples pTLA request. We're talking
about YOURS.
(2) We do NOT have the same problem with the ipv6-site object. There may
be discrepancy of listed routing protocols or the other site may not have
UPDATED _THEIR_ ipv6-site object but, the ENTERZONE ipv6-site object does
NOT reference NON-EXISTENT objects as yours does. Referencing
non-existent objects simply pollutes the 6bone database with inaccurate
information.
(3) You are correct. You can't force a peer to register a whois
entry. You *CAN* refuse to peer with entities who refuse to register, at
minimum, IPv6-site, Inet6num and Mntner objects as specifically _REQUIRED_
by section 5 of RFC2772. If they have a problem with this policy, you can
simply point them to the RFC and tell them that if you peer with them, you
have to create an entry in your ipv6-site object referencing
them. Without their having _REAL_ objects for you to reference, and
especially if you just make up ipv6-site objects to reference in yours,
you are in violation of RFC2772, section 7, subsection 1, paragraph A.
Let me lead you through Section 9 of RFC2772:
9. Common rules enforcement for the 6bone
Participation in the 6Bone is a voluntary and benevolent undertaking.
However, participating sites are expected to adhere to the rules and
policies described in this document in order to maintain the 6Bone as
a quality tool for the deployment of, and transition to, IPv6
protocols and the products implementing them.
The following is in support of policing adherence to 6Bone rules and
policies:
1. Each pTLA site has committed to implement the 6Bone's rules and
policies, and SHOULD try to ensure they are adhered to by sites
within their administrative control, i.e. those to who prefixes
under their respective pTLA prefix have been delegated.
Hrm... I'm betting that this applies to your peers who haven't registered
the appropriate objects.
2. When a site detects an issue, it SHOULD first use the 6Bone
registry to contact the site maintainer and work the issue.
Now, how do you suppose that I should go about contacting an entity whom I
can NOT look up in the 6bone registry?
3. If nothing happens, or there is disagreement on what the right
solution is, the issue SHOULD be brought to the 6Bone Operations
Group.
Guess what Nicolas. I detected problems with your application for a
pTLA. I detected problems with your ipv6-site object referencing
NON-EXISTENT objects. I brought these issues to your attention. Nothing
has been done to correct these problems and as a matter of fact, instead
of attending to the problems, you start pointing the finger at everyone
else and then go on to say "you can't force a peer to register a whois
entry."
---
John Fraizer
EnterZone, Inc
"I'm Nicolas DEFFAYET, i have 4 years, i got every day to school, and i
play with my FisherPrice routers and i destroy all 6bone." - Nicolas
DEFFAYET, 20 October 2002.
From pekkas@netcore.fi Mon Oct 21 08:30:40 2002
From: pekkas@netcore.fi (Pekka Savola)
Date: Mon, 21 Oct 2002 10:30:40 +0300 (EEST)
Subject: Cisco performance [Re: [6bone] pTLA request NDSOFTWARE - review
closes 23 October 2002]
In-Reply-To: <4.3.2.7.2.20021020123520.01ad1a58@europe.cisco.com>
Message-ID:
Hello,
To clarify on the list:
I redid the testing w/ ttcp a bit more carefully and I observed that of
7200, 7500 and GSR, the GSR is the slowest platform, capable of only the
stated 26 Mbit/s (in practise, a bit less due to packet drops: we are able
to get a constant 13 Mbit/s TCP stream through it). This should be fixed
with EFT code using CEFv6, but we weren't able to test that. 7500 with
CEFv6 is able to perform at full rate on FastEthernet, and on production
software a bit less than that: at least 50 Mbit/s should still yield
acceptable packet drop levels.
So, the situation is not as bad as first observed (and in all honesty, not
everybody needs to have 10+ Mbit/s grade throughput yet :-), even though
GSR performing worst with production-level features was a bit surprising.
(With "production software" here I mean any publicly available software...
whether it's usable in "production environment" is always a judgment call
:-)
Pekka
On Sun, 20 Oct 2002, Patrick Grossetete wrote:
> Your comment on the performances is not accurate. You reported a
> 26 Mb/s throughput using TTCP between
> 2 Linux hosts, asking for feedback about that. We just tested the same
> configuration but using IXIA traffic generators
> and got different numbers really different from yours. I asked Theo to
> provide you an official answer, so I expect
> you will update the list later.
>
> Patrick
>
> At 02:22 AM 20-10-02 +0300, Pekka Savola wrote:
> >On Sat, 19 Oct 2002, Paul Aitken wrote:
> > > Nicolas DEFFAYET wrote:
> > >
> > > > I don't have Cisco or Juniper routers because i don't have the budget
> > > > for that. You can offer me a Cisco if you want...
> > >
> > > http://www.cisco.com/warp/public/csc/refurb_equipment
> >
> >Well, to be frank, I'm not sure why anyone would want Cisco equipment for
> >IPv6, old or new. They hardly seem to be able to manage 30 Mbit/s of IPv6
> >traffic :-(. I guess this is enough for some, for us it isn't :-(.
> >
> >And no, we're not using "crap" (for some, usefull stuf for others) like
> >4x00's, 2x00's etc. like many seem to be doing: rather, like 7200, 7500,
> >12xxx, etc.
> >
> >And yes, I've tried to contact ipv6-support@cisco.com to try to find out
> >whether these are really the bottlenecks, no replies.
> >
> >--
> >Pekka Savola "Tell me of difficulties surmounted,
> >Netcore Oy not those you stumble over and fall"
> >Systems. Networks. Security. -- Robert Jordan: A Crown of Swords
> >
> >
> >_______________________________________________
> >6bone mailing list
> >6bone@mailman.isi.edu
> >http://mailman.isi.edu/mailman/listinfo/6bone
>
> ____________________________________________
> Patrick Grossetete
> Cisco Systems
> Internet Technology Division (ITD) - Product Manager
>
> Phone/Vmail: 33.1.58.04.61.52
> Fax: 33.1.58.04.61.00
> mobile: 33.6.19.98.51.31
> Email:pgrosset@cisco.com
> 11 Rue Camille Desmoulins
> 92782 Issy les Moulineaux Cedex 9
> France
> ____________________________________________
>
> _______________________________________________
> 6bone mailing list
> 6bone@mailman.isi.edu
> http://mailman.isi.edu/mailman/listinfo/6bone
>
--
Pekka Savola "Tell me of difficulties surmounted,
Netcore Oy not those you stumble over and fall"
Systems. Networks. Security. -- Robert Jordan: A Crown of Swords
From A.Weinberger@ebv.com Mon Oct 21 08:40:31 2002
From: A.Weinberger@ebv.com (Weinberger Andreas)
Date: Mon, 21 Oct 2002 09:40:31 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
Message-ID: <51420C40A341F14499B8C5A488BEDA5808EE78@EXSRV02.int.ebv.com>
hi there,
as pim wrote:
> In short: I oppose to this request also. Please read on.
i oppose to this request, too.
there is imho no need for a pTLA (someone said private tla,
perhabs the better description ;)) or ndsoftware, they aren't
any kind of an isp nor they have a lot of ipv6 customers.
for playing around i think, the /32 are enough (rir space is
also "only" /32, and lot of isp can live quite good from it...).
if any ix (btw, most ix are v6 only, they use a seperate switch)
needs ip addresses, go to the rir and request a /48 for it like
the others.
also, it looks like that ndsoftware wants to collect a lot of
bgp peers to be "kewl". they dont think about routing instability
or bgp flapping - and the effects to the whole ipv6 network
(both 6bone and "rir").
have a nice day,
bye,
::::
:: andreas 'randy' weinberger
:: networking group
:: ebv elektronik gmbh&co. kg
:: mail: a.weinberger@ebv.com
:: phone: +49 (0)8121 774-508
From pgrosset@cisco.com Mon Oct 21 08:47:27 2002
From: pgrosset@cisco.com (Patrick Grossetete)
Date: Mon, 21 Oct 2002 09:47:27 +0200
Subject: Cisco performance [Re: [6bone] pTLA request NDSOFTWARE -
review closes 23 October 2002]
In-Reply-To:
References: <4.3.2.7.2.20021020123520.01ad1a58@europe.cisco.com>
Message-ID: <4.3.2.7.2.20021021094109.01ae11e0@europe.cisco.com>
Thanks Pekka for this update. Slightly correction, on Cisco 12000,
you effectively need an upcoming
release [IOS 12.0(23)S] or its current EFT to take benefit of IPv6 Hardware
Forwading for the Engine 3. I acknowledge
you didn't get it for your tests, sorry for that.
Regards
Patrick
At 10:30 AM 21-10-02 +0300, Pekka Savola wrote:
>Hello,
>
>To clarify on the list:
>
>I redid the testing w/ ttcp a bit more carefully and I observed that of
>7200, 7500 and GSR, the GSR is the slowest platform, capable of only the
>stated 26 Mbit/s (in practise, a bit less due to packet drops: we are able
>to get a constant 13 Mbit/s TCP stream through it). This should be fixed
>with EFT code using CEFv6, but we weren't able to test that. 7500 with
>CEFv6 is able to perform at full rate on FastEthernet, and on production
>software a bit less than that: at least 50 Mbit/s should still yield
>acceptable packet drop levels.
>
>So, the situation is not as bad as first observed (and in all honesty, not
>everybody needs to have 10+ Mbit/s grade throughput yet :-), even though
>GSR performing worst with production-level features was a bit surprising.
>
>(With "production software" here I mean any publicly available software...
>whether it's usable in "production environment" is always a judgment call
>:-)
>
>Pekka
>
>On Sun, 20 Oct 2002, Patrick Grossetete wrote:
> > Your comment on the performances is not accurate. You reported a
> > 26 Mb/s throughput using TTCP between
> > 2 Linux hosts, asking for feedback about that. We just tested the same
> > configuration but using IXIA traffic generators
> > and got different numbers really different from yours. I asked Theo to
> > provide you an official answer, so I expect
> > you will update the list later.
> >
> > Patrick
> >
> > At 02:22 AM 20-10-02 +0300, Pekka Savola wrote:
> > >On Sat, 19 Oct 2002, Paul Aitken wrote:
> > > > Nicolas DEFFAYET wrote:
> > > >
> > > > > I don't have Cisco or Juniper routers because i don't have the budget
> > > > > for that. You can offer me a Cisco if you want...
> > > >
> > > > http://www.cisco.com/warp/public/csc/refurb_equipment
> > >
> > >Well, to be frank, I'm not sure why anyone would want Cisco equipment for
> > >IPv6, old or new. They hardly seem to be able to manage 30 Mbit/s of IPv6
> > >traffic :-(. I guess this is enough for some, for us it isn't :-(.
> > >
> > >And no, we're not using "crap" (for some, usefull stuf for others) like
> > >4x00's, 2x00's etc. like many seem to be doing: rather, like 7200, 7500,
> > >12xxx, etc.
> > >
> > >And yes, I've tried to contact ipv6-support@cisco.com to try to find out
> > >whether these are really the bottlenecks, no replies.
> > >
> > >--
> > >Pekka Savola "Tell me of difficulties surmounted,
> > >Netcore Oy not those you stumble over and fall"
> > >Systems. Networks. Security. -- Robert Jordan: A Crown of Swords
> > >
> > >
> > >_______________________________________________
> > >6bone mailing list
> > >6bone@mailman.isi.edu
> > >http://mailman.isi.edu/mailman/listinfo/6bone
> >
> > ____________________________________________
> > Patrick Grossetete
> > Cisco Systems
> > Internet Technology Division (ITD) - Product Manager
> >
> > Phone/Vmail: 33.1.58.04.61.52
> > Fax: 33.1.58.04.61.00
> > mobile: 33.6.19.98.51.31
> > Email:pgrosset@cisco.com
> > 11 Rue Camille Desmoulins
> > 92782 Issy les Moulineaux Cedex 9
> > France
> > ____________________________________________
> >
> > _______________________________________________
> > 6bone mailing list
> > 6bone@mailman.isi.edu
> > http://mailman.isi.edu/mailman/listinfo/6bone
> >
>
>--
>Pekka Savola "Tell me of difficulties surmounted,
>Netcore Oy not those you stumble over and fall"
>Systems. Networks. Security. -- Robert Jordan: A Crown of Swords
____________________________________________
Patrick Grossetete
Cisco Systems
Internet Technology Division (ITD) - Product Manager
Phone/Vmail: 33.1.58.04.61.52
Fax: 33.1.58.04.61.00
mobile: 33.6.19.98.51.31
Email:pgrosset@cisco.com
11 Rue Camille Desmoulins
92782 Issy les Moulineaux Cedex 9
France
____________________________________________
From rjorgensen@upctechnology.com Mon Oct 21 09:05:54 2002
From: rjorgensen@upctechnology.com (Roger Jorgensen)
Date: Mon, 21 Oct 2002 10:05:54 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October
2002
In-Reply-To: <5.1.0.14.0.20021016061927.02891008@imap2.es.net>
Message-ID: <5.1.0.14.0.20021021100345.02bdf460@213.46.233.213>
FYI,
NDSOFTWARE only have 99 peers, both of their peers to us,
AS6830/chello/aorta) and AS8733/TVD are being shutdown shortly.
Due to conflicts with our internal use of private ASN we have decided to no
longer
support private ASN peerings with external parties.
At 06:25 AM 10/16/2002 -0700, Bob Fink wrote:
>6bone Folk,
>
>NDSOFTWARE has requested a pTLA allocation and I find their request fully
>compliant with RFC2772. The open review period for this will close 23
>October 2002. Please send your comments to me or the list.
>
>
>
>
>
>
>Thanks,
>
>Bob
>
>=====
>>Hello,
>>
>>On behalf of NDSoftware, I would like to submit our application for a
>>pTLA.
>>
>>Best Regards,
>>
>>Nicolas DEFFAYET
>>
>>------------------------------------------------------------------------
>>
>> From RFC 2772
>>
>>
>>7. Guidelines for 6Bone pTLA sites
>>
>>
>> The following rules apply to qualify for a 6Bone pTLA allocation. It
>> should be recognized that holders of 6Bone pTLA allocations are
>> expected to provide production quality backbone network services for
>> the 6Bone.
>>
>>
>> 1. The pTLA Applicant must have a minimum of three (3) months
>> qualifying experience as a 6Bone end-site or pNLA transit.
>>During
>> the entire qualifying period the Applicant must be operationally
>> providing the following:
>>
>>Our ipv6-site is operational since 17 january 2001 on 6bone.
>>
>> a. Fully maintained, up to date, 6Bone Registry entries for their
>> ipv6-site inet6num, mntner, and person objects, including each
>> tunnel that the Applicant has.
>>
>>http://whois.6bone.net/cgi-bin/whois?NDSOFTWARE
>>
>>
>> b. Fully maintained, and reliable, BGP4+ peering and connectivity
>> between the Applicant's boundary router and the appropriate
>> connection point into the 6Bone. This router must be IPv6
>> pingable. This criteria is judged by members of the 6Bone
>> Operations Group at the time of the Applicant's pTLA request.
>>
>>We have currently 101 BGP4+ sessions.
>>
>>Our ASN is AS25358:
>>aut-num: AS25358
>>as-name: NDSOFTWARE-AS
>>descr: NDSoftware IP Network
>>
>>We use 2 routers:
>> - parcr1.fr.ndsoftwarenet.net
>> - parcr2.fr.ndsoftwarenet.net
>>Looking Glass: http://noc.ndsoftwarenet.com/lg/
>>
>>
>> c. Fully maintained DNS forward (AAAA) and reverse (ip6.int)
>> entries for the Applicant's router(s) and at least one host
>> system.
>>
>>We have 3 nameservers:
>> - ns1.ndsoftwarenet.com
>> - ns2.ndsoftwarenet.com
>> - ns3.ndsoftwarenet.com
>>
>> d. A fully maintained, and reliable, IPv6-accessible system
>> providing, at a mimimum, one or more web pages, describing the
>> Applicant's IPv6 services. This server must be IPv6 pingable.
>>
>>http://noc.ndsoftwarenet.com/
>>
>> 2. The pTLA Applicant MUST have the ability and intent to provide
>> "production-quality" 6Bone backbone service. Applicants must
>> provide a statement and information in support of this claim.
>> This MUST include the following:
>>
>>
>> a. A support staff of two persons minimum, three preferable, with
>> person attributes registered for each in the ipv6-site object
>> for the pTLA applicant.
>>
>>NDN1-6BONE
>>CB2-6BONE
>>BN3-6BONE
>>MM14-6BONE
>>MC7-6BONE
>>
>> b. A common mailbox for support contact purposes that all support
>> staff have acess to, pointed to with a notify attribute in the
>> ipv6-site object for the pTLA Applicant.
>>
>>ipmaster@ndsoftwarenet.com
>>
>> 3. The pTLA Applicant MUST have a potential "user community" that
>> would be served by its becoming a pTLA, e.g., the Applicant is a
>> major provider of Internet service in a region, country, or focus
>> of interest. Applicant must provide a statement and information
>> in support this claim.
>>
>>NDSoftware operates an IPv6 network and provide a lot of IPv6 services
>>to many projects.
>>
>>We provide to:
>>
>>IPv6-FR (a non profit organisation for the developement of IPv6 in France
>> 200 users, each user have a /48.
>>
>>NexGenCollective (http://www.nexgencollective.net/)
>> 150 users, each user have a /48.
>>
>>ATI (A tunisian ISP, http://www.ipv6net.tn/)
>>
>>and a lot of others (see our whois), this services: IPv6 connectivity
>>(STATIC or BGP with a IPv6 block), IPv6 newsfeeds/newsread,...
>>
>>We do many actions in IPv6 research, we created FNIX6 (French
>>International Internet Exchange IPv6, http://www.fnix6.net/), we host
>>many mirrors
>>available in IPv6, we created ftp://ftp.openipv6.com/ (a FTP with a lot
>>of IPv6 stuff).
>>
>> 4. The pTLA Applicant MUST commit to abide by the current 6Bone
>> operational rules and policies as they exist at time of its
>> application, and agree to abide by future 6Bone backbone
>> operational rules and policies as they evolve by consensus of the
>> 6Bone backbone and user community.
>>
>>
>>We agree to all current and future rules and policies.
>>
>>----
>
>-end
>
>_______________________________________________
>6bone mailing list
>6bone@mailman.isi.edu
>http://mailman.isi.edu/mailman/listinfo/6bone
---
Roger Jorgensen (rjorgensen@upctechnology.com)
System Engineer @ UPC Technology / IP engineering
handles: ROJO1-6BONE ROJO9-RIPE RJC10-NORID
From gert@space.net Mon Oct 21 09:18:13 2002
From: gert@space.net (Gert Doering)
Date: Mon, 21 Oct 2002 10:18:13 +0200
Subject: [6bone] Re: IPv6-only IXP's are absolutely wrong
In-Reply-To: <200210210013.g9L0Dlt25633@boreas.isi.edu>; from bmanning@ISI.EDU on Sun, Oct 20, 2002 at 05:13:47PM -0700
References: <20021020192435.GC15861@nic.fr> <200210210013.g9L0Dlt25633@boreas.isi.edu>
Message-ID: <20021021101812.D94537@Space.Net>
Hi,
On Sun, Oct 20, 2002 at 05:13:47PM -0700, Bill Manning wrote:
> it is true that for L2 fabric, it can be hard to prevent IPv6 or IPv4
> from using the fabric. There are reasons to try and split them however.
> This is not the right thread to take on that topic though.
> If this is of interest, send a note and I'll forward on the rational.
I'd like to hear more about that. At the german DE-CIX, we do v6 and v4
on the same mesh, just to make it easy to get v6 native peerings deployed,
and of course I'd like to hear all possible "why" and "why not"'s...
thanks,
Gert Doering
-- NetMaster
--
Total number of prefixes smaller than registry allocations: 48282 (47686)
SpaceNet AG Mail: netmaster@Space.Net
Joseph-Dollinger-Bogen 14 Tel : +49-89-32356-0
80807 Muenchen Fax : +49-89-32356-299
From bortzmeyer@gitoyen.net Mon Oct 21 09:21:33 2002
From: bortzmeyer@gitoyen.net (Stephane Bortzmeyer)
Date: Mon, 21 Oct 2002 10:21:33 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <20021020201445.GB17735@bfib.colo.bit.nl>
References: <5.1.0.14.0.20021016061927.02891008@imap2.es.net> <20021020201445.GB17735@bfib.colo.bit.nl>
Message-ID: <20021021082133.GA20569@nic.fr>
On Sun, Oct 20, 2002 at 10:14:45PM +0200,
Pim van Pelt wrote
a message of 76 lines which said:
> Last month, I saw a small thread on the lir-wg@ripe.net mailinglist,
> where Nicolas complained in this public forum about the fact that the
> NCC did not grant him an AS number request.
RIPE-NCC gave a stupid reply ("you do not need an AS number for IPv6")
and fixed the mistake afterwards with apologies. In that case, it was
not Nicolas' fault.
From pim@ipng.nl Mon Oct 21 09:27:34 2002
From: pim@ipng.nl (Pim van Pelt)
Date: Mon, 21 Oct 2002 10:27:34 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <20021021082133.GA20569@nic.fr>
References: <5.1.0.14.0.20021016061927.02891008@imap2.es.net> <20021020201445.GB17735@bfib.colo.bit.nl> <20021021082133.GA20569@nic.fr>
Message-ID: <20021021082734.GA14333@bfib.colo.bit.nl>
On Mon, Oct 21, 2002 at 10:21:33AM +0200, Stephane Bortzmeyer wrote:
| On Sun, Oct 20, 2002 at 10:14:45PM +0200,
| Pim van Pelt wrote
| a message of 76 lines which said:
|
| > Last month, I saw a small thread on the lir-wg@ripe.net mailinglist,
| > where Nicolas complained in this public forum about the fact that the
| > NCC did not grant him an AS number request.
|
| RIPE-NCC gave a stupid reply ("you do not need an AS number for IPv6")
| and fixed the mistake afterwards with apologies. In that case, it was
| not Nicolas' fault.
Stephane,
To be frank: the answer was not stupid, nor was there a need to
appologise for it. To run IPv6, one does not need an AS number. One
needs a machine which supports the protocol.
I know of several hundreds of companies and individuals, even ISPs, who
are connected to the internet in another autonomous system. I am not
quite sure on the policies of AS number allocation at the RIPE NCC, but
I am quite sure that to be eligible for an AS number, one must operate
an autonomous system.
Nevertheless, DEFFAYET is (still) and enduser. Complaining in public
fora is his right (perogative :), but it is not his place to deal with
these matters: his LIR should act on his behalf. I think RIPE NCC should
stick to dealing with their members, not their members' customers.
Anyway, I wish DEFFAYET good luck with deploying his ISP solution and
with his software and webdesign company as well (no pun intended).
groet,
Pim
--
---------- - - - - -+- - - - - ----------
Pim van Pelt Email: pim@ipng.nl
http://www.ipng.nl/ IPv6 Deployment
-----------------------------------------------
From mohacsi@niif.hu Mon Oct 21 13:31:21 2002
From: mohacsi@niif.hu (Janos Mohacsi)
Date: Mon, 21 Oct 2002 14:31:21 +0200 (CEST)
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <1035053895.634.1978.camel@wks1.fr.corp.ndsoftware.com>
Message-ID: <20021021142805.Y614-100000@evil.ki.iif.hu>
On Sat, 19 Oct 2002, Nicolas DEFFAYET wrote:
> On Sat, 2002-10-19 at 20:38, Rico Gloeckner wrote:
>
> > Ok, let me rephrase you:
> > | I close tunnels to small local Peers because i want to peer with Large
> > | Peers around the world.
> >
> > Did I understand you correctly?
> >
> > If so, i hope all your Peers will decrease the Priority to you to a very
> > minimum, because this is the best Way to fuck up IPv6-Routing.
>
> I use MED, i don't have a bad routing, you can check:
>
> http://noc.ndsoftwarenet.com/stats/aspath-tree/bgp-page-complete.php
>
> I see all european pTLA/sTLA by European peer,...
>
> My MED for tunnels:
>
> 500: - 10 ms
> 510: 10 - 25 ms
> 520: 25 - 50 ms
> 530: 50 - 100 ms
> 540: + 100 ms
>
> A lot of pTLA and sTLA don't use MED, i use MED for have a good quality.
>
MED is only useful for the same autonomous systems! As its name is
describing: Multiple Exit Discriminator. Do you have more than one tunnel
to the same autonomous systems? I strongly support to have answer to Tim
Chown's questions...
Janos Mohacsi
From tvo@EnterZone.Net Mon Oct 21 14:01:29 2002
From: tvo@EnterZone.Net (John Fraizer)
Date: Mon, 21 Oct 2002 09:01:29 -0400 (EDT)
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: <20021021101624.W43272-100000@doos.cluecentral.net>
Message-ID:
On Mon, 21 Oct 2002, Sabri Berisha wrote:
> On 19 Oct 2002, Nicolas DEFFAYET wrote:
>
> > > - Can't he get some IPv6 space from his upstreams?
> >
> > No, i want announce a pTLA.
>
> I want a pony. But unless I give a really good reason for it, my
> girlfriend will never allow me to get me one.
And I want a 4-bay vertical array for 160m, and two 96-ft freestanding
towers (you have to have redundancy ya know) each with a FluidMotion
SteppIR 3-element 20m-6m yagi, an M2 80M3LLA 3-element 80m yagi and an M2
40M4LLDD 4-element 40m yagi, all fed with 1-7/8 hardline, positioned by an
M2 OR2800P-DC prop-pitch positioner with with the set capable of being
operated as a phased array.
You arrange that and I'll support the NDSoftware pTLA request - screw the
rest of you! I want my antenna farm!
Bwahahahahah!
-.- -.-. ....- -.- --. ..-
From d.alligand@pobox.com Mon Oct 21 15:11:32 2002
From: d.alligand@pobox.com (Denis Alligand)
Date: Mon, 21 Oct 2002 16:11:32 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October
2002
In-Reply-To:
References: <1035135413.4779.402.camel@wks1.fr.corp.ndsoftware.com>
Message-ID: <5.1.0.14.2.20021021155755.01bcbe48@mail.phpfreelance.com>
>
> >
> > > role: IPv6-FR NOC
> > > address: IPv6-FR
> > > address: 57 rue du president Wilson
> > > address: 92300 Levallois-Perret
> > > address: France
> > > phone: +33 671887502
> > >
> > > role: NDSoftware NOC
> > > address: NDSoftware
> > > address: 57 rue du president Wilson
> > > address: 92300 Levallois-Perret
> > > address: France
> > > phone: +33 671887502
> > >
> > >
> > > I'm sorry Nicolas. Providing address space to YOURSELF doesn't
> > > count! Sheesh!
> >
> >
> > NDSoftware and IPv6-FR are in the same building but aren't the same
> > legal organization.
> > I'm a network administrator for the both.
>
>And would this be your home by some strange coincidence?
Strange ain't it ;-)
Ndsofware does not seems to exist in the French registration company, and
when i have a look at this address i can only find out: Deffayet Jean-Yves
(nicolas'father?)
no one else, and no ndsoftware at all. So something i don't really
understand about that:
What is NDSoftware?
The phone number is a mobile phone, and i can't find any regular phone line
for this company.
I had a look about the IPv6 peering point, it is supposed to be located in
telecity in paris, i would like to know where is the rack.
weird, who said this word? ;-)
Denis
From d.alligand@pobox.com Mon Oct 21 15:14:06 2002
From: d.alligand@pobox.com (Denis Alligand)
Date: Mon, 21 Oct 2002 16:14:06 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October
2002
Message-ID: <5.1.0.14.2.20021021161353.034721a0@pop.nwc.fr>
>
> >
> > > role: IPv6-FR NOC
> > > address: IPv6-FR
> > > address: 57 rue du president Wilson
> > > address: 92300 Levallois-Perret
> > > address: France
> > > phone: +33 671887502
> > >
> > > role: NDSoftware NOC
> > > address: NDSoftware
> > > address: 57 rue du president Wilson
> > > address: 92300 Levallois-Perret
> > > address: France
> > > phone: +33 671887502
> > >
> > >
> > > I'm sorry Nicolas. Providing address space to YOURSELF doesn't
> > > count! Sheesh!
> >
> >
> > NDSoftware and IPv6-FR are in the same building but aren't the same
> > legal organization.
> > I'm a network administrator for the both.
>
>And would this be your home by some strange coincidence?
Strange ain't it ;-)
Ndsofware does not seems to exist in the French registration company, and
when i have a look at this address i can only find out: Deffayet Jean-Yves
(nicolas'father?)
no one else, and no ndsoftware at all. So something i don't really
understand about that:
What is NDSoftware?
The phone number is a mobile phone, and i can't find any regular phone line
for this company.
I had a look about the IPv6 peering point, it is supposed to be located in
telecity in paris, i would like to know where is the rack.
weird, who said this word? ;-)
Denis
From eric@roxanne.org Mon Oct 21 15:18:42 2002
From: eric@roxanne.org (Eric Gauthier)
Date: Mon, 21 Oct 2002 10:18:42 -0400
Subject: [6bone] Re: IPv6-only IXP's are absolutely wrong
In-Reply-To: <20021021101812.D94537@Space.Net>; from gert@space.net on Mon, Oct 21, 2002 at 10:18:13AM +0200
References: <20021020192435.GC15861@nic.fr> <200210210013.g9L0Dlt25633@boreas.isi.edu> <20021021101812.D94537@Space.Net>
Message-ID: <20021021101842.A9700@roxanne.org>
> On Sun, Oct 20, 2002 at 05:13:47PM -0700, Bill Manning wrote:
> > it is true that for L2 fabric, it can be hard to prevent IPv6 or IPv4
> > from using the fabric. There are reasons to try and split them however.
> > This is not the right thread to take on that topic though.
> > If this is of interest, send a note and I'll forward on the rational.
>
> I'd like to hear more about that. At the german DE-CIX, we do v6 and v4
> on the same mesh, just to make it easy to get v6 native peerings deployed,
> and of course I'd like to hear all possible "why" and "why not"'s...
I don't know about the various IX's, but the I2 gigapop that our University
uses is concerned about this. If I remember correctly (not that the exact
numbers are important), but IPv4 is in an ethernet frame with type 0x0800 and
IPv6 is in an ethernet frame with type 0x86dd, so layer 3/4 aware switches
will likely handle these frames differently. In our case, the Cisco 12,000
and 6500's that we're using are great for IPv4 packets (they handle them
in hardware), but IPv6 packets are handled in software so we don't expect
nearly the same type of performance. I'd imagine that something like this
is what they're alluding to.
Eric :)
From gert@space.net Mon Oct 21 15:24:49 2002
From: gert@space.net (Gert Doering)
Date: Mon, 21 Oct 2002 16:24:49 +0200
Subject: [6bone] Re: IPv6-only IXP's are absolutely wrong
In-Reply-To: <20021021101842.A9700@roxanne.org>; from eric@roxanne.org on Mon, Oct 21, 2002 at 10:18:42AM -0400
References: <20021020192435.GC15861@nic.fr> <200210210013.g9L0Dlt25633@boreas.isi.edu> <20021021101812.D94537@Space.Net> <20021021101842.A9700@roxanne.org>
Message-ID: <20021021162449.U94537@Space.Net>
Hi,
On Mon, Oct 21, 2002 at 10:18:42AM -0400, Eric Gauthier wrote:
> I don't know about the various IX's, but the I2 gigapop that our University
> uses is concerned about this. If I remember correctly (not that the exact
> numbers are important), but IPv4 is in an ethernet frame with type 0x0800 and
> IPv6 is in an ethernet frame with type 0x86dd, so layer 3/4 aware switches
> will likely handle these frames differently. In our case, the Cisco 12,000
> and 6500's that we're using are great for IPv4 packets (they handle them
> in hardware), but IPv6 packets are handled in software so we don't expect
> nearly the same type of performance. I'd imagine that something like this
> is what they're alluding to.
This is definitely relevant for the individual participants - but for
the IXP switch (which is strictly layer 2 *only*, at least for all IXes
that I know), L3/4 forwarding performance should not be an issue.
One issue that I see is multicast (neighbor discovery etc) which isn't
seen on an IPv4 unicast exchange switch.
Gert Doering
-- NetMaster
--
Total number of prefixes smaller than registry allocations: 48282 (47686)
SpaceNet AG Mail: netmaster@Space.Net
Joseph-Dollinger-Bogen 14 Tel : +49-89-32356-0
80807 Muenchen Fax : +49-89-32356-299
From bmanning@ISI.EDU Mon Oct 21 16:50:05 2002
From: bmanning@ISI.EDU (Bill Manning)
Date: Mon, 21 Oct 2002 08:50:05 -0700 (PDT)
Subject: [6bone] Re: IPv6-only IXP's are absolutely wrong
In-Reply-To: <20021021162449.U94537@Space.Net> from Gert Doering at "Oct 21, 2 04:24:49 pm"
Message-ID: <200210211550.g9LFo5N11129@boreas.isi.edu>
% On Mon, Oct 21, 2002 at 10:18:42AM -0400, Eric Gauthier wrote:
% > I don't know about the various IX's, but the I2 gigapop that our University
% > uses is concerned about this. If I remember correctly (not that the exact
% > numbers are important), but IPv4 is in an ethernet frame with type 0x0800 and
% > IPv6 is in an ethernet frame with type 0x86dd, so layer 3/4 aware switches
% > will likely handle these frames differently. In our case, the Cisco 12,000
% > and 6500's that we're using are great for IPv4 packets (they handle them
% > in hardware), but IPv6 packets are handled in software so we don't expect
% > nearly the same type of performance. I'd imagine that something like this
% > is what they're alluding to.
%
% This is definitely relevant for the individual participants - but for
% the IXP switch (which is strictly layer 2 *only*, at least for all IXes
% that I know), L3/4 forwarding performance should not be an issue.
%
% One issue that I see is multicast (neighbor discovery etc) which isn't
% seen on an IPv4 unicast exchange switch.
%
% Gert Doering
reasons to split v4 and v6:
multicast - v4 and v6 treat this differently.
fabric "optimizations" - framing support, buffer sizing, MTU, etc.
a decent L2 fabric will be able to accomodate the larger MTUs
of native v6 and won't complain about divergent framing.
"Smart" fabrics tend to be tuned to IPv4.
then there are issues wrt RA/ND on an exchange... having all the
participants trying to "stamp" the fabric with their version of
which prefix to use is noisy at best and an effective DOS at worst.
then there are mgmt issues. most L2 fabrics do not have up to date
v6 mib support, so the stats/traffic collection is not as accurate
as it should be.
these things, in addition to the issues with the connecting gear
(v4 in HW, v6 in SW), tend to argue that a mixed-mode exchange may
be less stable / harder to troubleshoot than an single use
facility. The offset is the capex/opex costs of connecting to two
facilities, one for each protocol.
--
--bill
From pekkas@netcore.fi Mon Oct 21 17:11:16 2002
From: pekkas@netcore.fi (Pekka Savola)
Date: Mon, 21 Oct 2002 19:11:16 +0300 (EEST)
Subject: [6bone] very drafty draft on 6bone routing mess
In-Reply-To:
Message-ID:
This is getting a bit less drafty now, but a much expanded version is
available at:
http://www.netcore.fi/pekkas/ietf/6bone-messv2.txt
Comments are welcome, as always. I'll send this to internet-drafts@ in a
couple of days in any case.
Perhaps I should summarize (Braveheart voice): "We need the transits". And
what is more, we need transits that have good policies, and will be
significant enough to force the pTLA/sTLA's they give service to into
something sane (to drop 95% of those tunnels etc.).
On Sun, 20 Oct 2002, Pekka Savola wrote:
> Hello,
>
> The so-called 6bone-mess has been discussed here back and forth, with no
> apparent result or success.
>
> Based on experiences gained in 6NET and seeing how others are doing, I got
> motivated enough to write something down on the subject.
>
> A very drafty draft (result of 3 hours of torturing the keyboard in middle
> of the night :-) on 6bone routing policy issues which I believe are
> causing current problems is available at:
>
> http://www.netcore.fi/pekkas/ietf/6bone-mess.txt
>
> There are also some ideas, but nothing specific, how one could get around
> those.
>
> Before I want to go make this something a bit less drafty I'd like to get
> opinions and thoughts on this: should we try to do something to try to
> ensure IPv6 Internet would actually get usable one of these years (it sure
> ain't now!)
>
> Comments, please.
>
> (Anyone with more than 5 peers with transit should feel the sting of guilt
> now. :-)
>
>
--
Pekka Savola "Tell me of difficulties surmounted,
Netcore Oy not those you stumble over and fall"
Systems. Networks. Security. -- Robert Jordan: A Crown of Swords
From chbm@cprm.net Mon Oct 21 19:01:07 2002
From: chbm@cprm.net (Carlos Morgado)
Date: Mon, 21 Oct 2002 19:01:07 +0100
Subject: sTLA alloc policies [Re: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002]
In-Reply-To: ; from pekkas@netcore.fi on Sun, Oct 20, 2002 at 05:36:10PM +0300
References: <20021020142528.X94537@Space.Net>
Message-ID: <20021021190107.B4699@cprm.net>
On Sun, Oct 20, 2002 at 05:36:10PM +0300, Pekka Savola wrote:
> On Sun, 20 Oct 2002, Gert Doering wrote:
> > On Sun, Oct 20, 2002 at 02:19:19PM +0200, Jeroen Massar wrote:
> > > [ RIPE rules for IPv6 address space ]
> > >
> > > That's what I meant to express. They do have political reasons though.
> > > And as most people know politics are not nice.
> >
> > Partly political, but also partly technical - the multihoming issue
> > isn't really solved yet, and have every end site have their own /32
> > announced into the global table is not a scalable approach.
> >
> > The political part is the "200 customer rule", which I personally did
> > not like very much (it came from ARIN and APNIC), but hey, for a serious
> > ISP that actually is connecting customers, it's not a major obstacle.
>
> Speaking of which, I'd be really interested in knowing how Internet
> Software Consortium is going to fill the "200 customer rule":
>
CPR Marconi (PTComunicações now) routes about 80% of the portuguese commercial
internet traffic. We have an IPv4 /19 and are pretty much multihomed in
IPv4 as any self respecting internet whole saler should be. However, after
reading RIPE's IPv6 policies I came to the conclusion we can't request a
block from them. "Get it from your upstream" is pretty much useless for
multihomed nets so we're pretty much stuck.
All our customers however can get /32s from RIPE as they can fill a plan
saying "we have 250 PoPs". Soooo, our larger upstreams have IPv6 blocks,
our *client* ISPs have IPv6 blocks but we, *their upstream*, can't get a
block.
Pretty much laughable eh ?
--
Carlos Morgado - Internet Engineering - Phone +351 214146594
GPG key: 0x75E451E2 FP: B98B 222B F276 18C0 266B 599D 93A1 A3FB 75E4 51E2
The views expressed above do not bind my employer.
From pekkas@netcore.fi Mon Oct 21 19:38:07 2002
From: pekkas@netcore.fi (Pekka Savola)
Date: Mon, 21 Oct 2002 21:38:07 +0300 (EEST)
Subject: [6bone] Re: IPv6-only IXP's are absolutely wrong
In-Reply-To: <200210211550.g9LFo5N11129@boreas.isi.edu>
Message-ID:
On Mon, 21 Oct 2002, Bill Manning wrote:
> % On Mon, Oct 21, 2002 at 10:18:42AM -0400, Eric Gauthier wrote:
> % > I don't know about the various IX's, but the I2 gigapop that our University
> % > uses is concerned about this. If I remember correctly (not that the exact
> % > numbers are important), but IPv4 is in an ethernet frame with type 0x0800 and
> % > IPv6 is in an ethernet frame with type 0x86dd, so layer 3/4 aware switches
> % > will likely handle these frames differently. In our case, the Cisco 12,000
> % > and 6500's that we're using are great for IPv4 packets (they handle them
> % > in hardware), but IPv6 packets are handled in software so we don't expect
> % > nearly the same type of performance. I'd imagine that something like this
> % > is what they're alluding to.
> %
> % This is definitely relevant for the individual participants - but for
> % the IXP switch (which is strictly layer 2 *only*, at least for all IXes
> % that I know), L3/4 forwarding performance should not be an issue.
> %
> % One issue that I see is multicast (neighbor discovery etc) which isn't
> % seen on an IPv4 unicast exchange switch.
> %
> % Gert Doering
>
> reasons to split v4 and v6:
>
> multicast - v4 and v6 treat this differently.
> fabric "optimizations" - framing support, buffer sizing, MTU, etc.
> a decent L2 fabric will be able to accomodate the larger MTUs
> of native v6 and won't complain about divergent framing.
> "Smart" fabrics tend to be tuned to IPv4.
>
> then there are issues wrt RA/ND on an exchange... having all the
> participants trying to "stamp" the fabric with their version of
> which prefix to use is noisy at best and an effective DOS at worst.
>
> then there are mgmt issues. most L2 fabrics do not have up to date
> v6 mib support, so the stats/traffic collection is not as accurate
> as it should be.
>
> these things, in addition to the issues with the connecting gear
> (v4 in HW, v6 in SW), tend to argue that a mixed-mode exchange may
> be less stable / harder to troubleshoot than an single use
> facility. The offset is the capex/opex costs of connecting to two
> facilities, one for each protocol.
I believe most of your points would be satisfied by "run different vlan's
for v4 and v6'. Seems sane to me, and works quite fine for L2 exchanges.
If you need L3 for some reason the issue of HW/SW forwarding may become
important.
On smaller exchanges, building dedicated peer-to-peer VLAN's is IMO also
an option worth considering.
--
Pekka Savola "Tell me of difficulties surmounted,
Netcore Oy not those you stumble over and fall"
Systems. Networks. Security. -- Robert Jordan: A Crown of Swords
From david@IPRG.nokia.com Mon Oct 21 19:46:57 2002
From: david@IPRG.nokia.com (David Kessens)
Date: Mon, 21 Oct 2002 11:46:57 -0700
Subject: sTLA alloc policies [Re: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002]
In-Reply-To: <20021021190107.B4699@cprm.net>; from chbm@cprm.net on Mon, Oct 21, 2002 at 07:01:07PM +0100
References: <20021020142528.X94537@Space.Net> <20021021190107.B4699@cprm.net>
Message-ID: <20021021114657.C28213@iprg.nokia.com>
Carlos,
On Mon, Oct 21, 2002 at 07:01:07PM +0100, Carlos Morgado wrote:
>
> CPR Marconi (PTComunicagues now) routes about 80% of the portuguese commercial
> internet traffic. We have an IPv4 /19 and are pretty much multihomed in
> IPv4 as any self respecting internet whole saler should be. However, after
> reading RIPE's IPv6 policies I came to the conclusion we can't request a
> block from them. "Get it from your upstream" is pretty much useless for
> multihomed nets so we're pretty much stuck.
>
> All our customers however can get /32s from RIPE as they can fill a plan
> saying "we have 250 PoPs". Soooo, our larger upstreams have IPv6 blocks,
> our *client* ISPs have IPv6 blocks but we, *their upstream*, can't get a
> block.
> Pretty much laughable eh ?
Before actually stating that you cannot get addresses, did you
actually filled out an apllication and try to get them ?!? The new
rules are really not as strict as many people believe they are.
Also, there is not much point in moaning about ipv6 allocation
policies on this list. This list doesn't decide about ipv6 policies -
the RIRs communities do.
As for the policy itself: the current policy is, like most policies, a
compromise. It certainly doesn't addresses every single parties needs.
The policy was adopted because most people felt that it was a step
forward from the previous policy, not because all parties felt this
was the final and best policy we have ever achieved.
Finally, the policy can be changed in the future and you are
encouraged to speak up in the public RIR policy fora on how you want
to have the policy changed in such a manner that it still achieves the
goals as set forward in the policy document.
David K.
---
From bmanning@ISI.EDU Mon Oct 21 19:49:30 2002
From: bmanning@ISI.EDU (Bill Manning)
Date: Mon, 21 Oct 2002 11:49:30 -0700 (PDT)
Subject: [6bone] Re: IPv6-only IXP's are absolutely wrong
In-Reply-To: from Pekka Savola at "Oct 21, 2 09:38:07 pm"
Message-ID: <200210211849.g9LInUp22799@boreas.isi.edu>
% > fabric "optimizations" - framing support, buffer sizing, MTU, etc.
% > a decent L2 fabric will be able to accomodate the larger MTUs
% > of native v6 and won't complain about divergent framing.
% > "Smart" fabrics tend to be tuned to IPv4.
% >
% > then there are mgmt issues. most L2 fabrics do not have up to date
% > v6 mib support, so the stats/traffic collection is not as accurate
% > as it should be.
% >
% I believe most of your points would be satisfied by "run different vlan's
% for v4 and v6'. Seems sane to me, and works quite fine for L2 exchanges.
% If you need L3 for some reason the issue of HW/SW forwarding may become
% important.
%
% On smaller exchanges, building dedicated peer-to-peer VLAN's is IMO also
% an option worth considering.
%
presuming vlan support on the fabric
presuming accurate auditing/monitoring capabilities for v6
presuming a switch that won't eat your lunch (performance
hits) with larger packets, different framing, MTU than "normal"
IPv4.
-but- given the level of v6 native traffic, esp. in relationship
to the level of v4 traffic, the arguments may be moot.
These were simply points as to why one might consider
running v6 only exchanges. I think that the issues are more
pronounced with transit providers infrastructrure, which, (IMHO)
is one reason that v6 is having a more gradual rampup than
some would have liked to see.
--
--bill
From chbm@cprm.net Mon Oct 21 20:07:15 2002
From: chbm@cprm.net (Carlos Morgado)
Date: Mon, 21 Oct 2002 20:07:15 +0100
Subject: sTLA alloc policies [Re: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002]
In-Reply-To: <20021021114657.C28213@iprg.nokia.com>; from david@iprg.nokia.com on Mon, Oct 21, 2002 at 11:46:57AM -0700
References: <20021020142528.X94537@Space.Net> <20021021190107.B4699@cprm.net> <20021021114657.C28213@iprg.nokia.com>
Message-ID: <20021021200715.A4844@cprm.net>
On Mon, Oct 21, 2002 at 11:46:57AM -0700, David Kessens wrote:
>
> Carlos,
>
> On Mon, Oct 21, 2002 at 07:01:07PM +0100, Carlos Morgado wrote:
> >
> > CPR Marconi (PTComunicagues now) routes about 80% of the portuguese commercial
> > internet traffic. We have an IPv4 /19 and are pretty much multihomed in
> > IPv4 as any self respecting internet whole saler should be. However, after
> > reading RIPE's IPv6 policies I came to the conclusion we can't request a
> > block from them. "Get it from your upstream" is pretty much useless for
> > multihomed nets so we're pretty much stuck.
> >
> > All our customers however can get /32s from RIPE as they can fill a plan
> > saying "we have 250 PoPs". Soooo, our larger upstreams have IPv6 blocks,
> > our *client* ISPs have IPv6 blocks but we, *their upstream*, can't get a
> > block.
> > Pretty much laughable eh ?
>
> Before actually stating that you cannot get addresses, did you
> actually filled out an apllication and try to get them ?!? The new
> rules are really not as strict as many people believe they are.
>
Yes, I said we can't fulfill
d) have a plan for making at least 200 /48 assignments
to other organisations within two years.
as part of the problem presentation.
They said "our policies are at ....".
True, this isn't the correct place to discuss RIPE policy, I was just
ilustrating Pekka's point.
--
Carlos Morgado - Internet Engineering - Phone +351 214146594
GPG key: 0x75E451E2 FP: B98B 222B F276 18C0 266B 599D 93A1 A3FB 75E4 51E2
The views expressed above do not bind my employer.
From bortzmeyer@gitoyen.net Mon Oct 21 20:05:51 2002
From: bortzmeyer@gitoyen.net (Stephane Bortzmeyer)
Date: Mon, 21 Oct 2002 21:05:51 +0200
Subject: [6bone] Re: IPv6-only IXP's are absolutely wrong
In-Reply-To: <200210211550.g9LFo5N11129@boreas.isi.edu>
(Bill Manning 's message of
Mon, 21 Oct 2002 08:50:05 PDT)
Message-ID: <200210211905.g9LJ5ptJ007617@ludwigV.sources.org>
On Monday 21 October 2002, at 8 h 50,
Bill Manning wrote:
> then there are issues wrt RA/ND on an exchange... having all the
> participants trying to "stamp" the fabric with their version of
> which prefix to use is noisy at best and an effective DOS at worst.
It's exactly like people having a DHCP server which actually replies to request broadcasted on the exchange (I've seen that). By definition, an exchange is shared and people can do funny things on it. IMHO, this problem should be solved by documentation/education/monitoring/ass-kicking, not by separating the traffics.
From bmanning@ISI.EDU Mon Oct 21 20:10:58 2002
From: bmanning@ISI.EDU (Bill Manning)
Date: Mon, 21 Oct 2002 12:10:58 -0700 (PDT)
Subject: [6bone] Re: IPv6-only IXP's are absolutely wrong
In-Reply-To: <200210211905.g9LJ5ptJ007617@ludwigV.sources.org> from Stephane Bortzmeyer at "Oct 21, 2 09:05:51 pm"
Message-ID: <200210211910.g9LJAwJ08978@boreas.isi.edu>
% On Monday 21 October 2002, at 8 h 50,
% Bill Manning wrote:
%
% > then there are issues wrt RA/ND on an exchange... having all the
% > participants trying to "stamp" the fabric with their version of
% > which prefix to use is noisy at best and an effective DOS at worst.
%
% It's exactly like people having a DHCP server which actually replies to request broadcasted on the exchange (I've seen that). By definition, an exchange is shared and people can do funny things on it. IMHO, this problem should be solved by documentation/education/monitoring/ass-kicking, not by separating the traffics.
%
Yes. but the impact of "default" behaviours on a mixed mode
facility may be more than some of the participants may wish
to absorb.
--
--bill
From ipv6@worrells.org Mon Oct 21 20:35:12 2002
From: ipv6@worrells.org (Jeremy Worrells)
Date: Mon, 21 Oct 2002 13:35:12 -0600
Subject: [6bone] IPv6 via NAT - problems
Message-ID: <005001c27938$faffe900$020010ac@worrells.org>
Good afternoon, 6bone!
I am a career sysadmin with a great interest in IPv6. I am trying to get a
small IPv6 network running at home, using Freenet6 and a FreeBSD 4.6.2
server. I have the software installed, and the interface comes up. However,
I cannot get a ping6 or traceroute6 to supply any answers. My Cisco 678 DSL
router is set up to NAT all protocol 41 and TCP port 4343 traffic to my
FreeBSD server. Here is the output from a tcpdump on another machine on the
LAN:
11:33:50.929054 worrells.dsl.xmission.com > www.freenet6.net:
jworrells.tsps1.freenet6.net > 3ffe:505:2008:1:2e0:18ff:fea8:4533: icmp6:
echo request (encap)
11:33:51.358566 www.freenet6.net > 172.16.0.4:
3ffe:505:2008:1:2e0:18ff:fea8:4533 > jworrells.tsps1.freenet6.net: icmp6:
echo reply (encap)
It appears that the packets are going to the right places, but are for some
reason not getting to the IPv6 stack. Here are my interface details:
fxp0: flags=8843 mtu 1500
inet 172.16.0.4 netmask 0xffffff00 broadcast 172.16.0.255
inet6 fe80::203:47ff:fe98:96f5%fxp0 prefixlen 64 scopeid 0x1
ether 00:03:47:98:96:f5
media: Ethernet autoselect (10baseT/UTP)
status: active
gif0: flags=8051 mtu 1280
tunnel inet 166.70.27.6 --> 206.123.31.114
inet6 fe80::203:47ff:fe98:96f5%gif0 prefixlen 64 scopeid 0x7
inet6 3ffe:b80:2:f1a6::2 --> 3ffe:b80:2:f1a6::1 prefixlen 128
Any ideas?
Jeremy
From pekkas@netcore.fi Mon Oct 21 20:37:05 2002
From: pekkas@netcore.fi (Pekka Savola)
Date: Mon, 21 Oct 2002 22:37:05 +0300 (EEST)
Subject: sTLA alloc policies [Re: [6bone] pTLA request NDSOFTWARE - review
closes 23 October 2002]
In-Reply-To: <20021021190107.B4699@cprm.net>
Message-ID:
On Mon, 21 Oct 2002, Carlos Morgado wrote:
> On Sun, Oct 20, 2002 at 05:36:10PM +0300, Pekka Savola wrote:
> > On Sun, 20 Oct 2002, Gert Doering wrote:
> > > On Sun, Oct 20, 2002 at 02:19:19PM +0200, Jeroen Massar wrote:
> > > > [ RIPE rules for IPv6 address space ]
> > > >
> > > > That's what I meant to express. They do have political reasons though.
> > > > And as most people know politics are not nice.
> > >
> > > Partly political, but also partly technical - the multihoming issue
> > > isn't really solved yet, and have every end site have their own /32
> > > announced into the global table is not a scalable approach.
> > >
> > > The political part is the "200 customer rule", which I personally did
> > > not like very much (it came from ARIN and APNIC), but hey, for a serious
> > > ISP that actually is connecting customers, it's not a major obstacle.
> >
> > Speaking of which, I'd be really interested in knowing how Internet
> > Software Consortium is going to fill the "200 customer rule":
> >
>
> CPR Marconi (PTComunicações now) routes about 80% of the portuguese commercial
> internet traffic. We have an IPv4 /19 and are pretty much multihomed in
> IPv4 as any self respecting internet whole saler should be. However, after
> reading RIPE's IPv6 policies I came to the conclusion we can't request a
> block from them. "Get it from your upstream" is pretty much useless for
> multihomed nets so we're pretty much stuck.
Looking at a looking glass, you do indeed seem to be as big as one can get
in the country; pretty impressive.
Don't you really have (about) 200 customers, or do you only provide
transit service?
The rules are a bit flexible..
> All our customers however can get /32s from RIPE as they can fill a plan
> saying "we have 250 PoPs". Soooo, our larger upstreams have IPv6 blocks,
> our *client* ISPs have IPv6 blocks but we, *their upstream*, can't get a
> block.
> Pretty much laughable eh ?
PoP's are not important: one can have 200 customers (e.g. DSL users) in a
single router.
--
Pekka Savola "Tell me of difficulties surmounted,
Netcore Oy not those you stumble over and fall"
Systems. Networks. Security. -- Robert Jordan: A Crown of Swords
From jeroen@unfix.org Mon Oct 21 21:31:01 2002
From: jeroen@unfix.org (Jeroen Massar)
Date: Mon, 21 Oct 2002 22:31:01 +0200
Subject: [6bone] IPv6 via NAT - problems
In-Reply-To: <005001c27938$faffe900$020010ac@worrells.org>
Message-ID: <001801c27940$c93e8220$210d640a@unfix.org>
Jeremy Worrells wrote:
> 11:33:50.929054 worrells.dsl.xmission.com > www.freenet6.net:
> jworrells.tsps1.freenet6.net >
> 3ffe:505:2008:1:2e0:18ff:fea8:4533: icmp6:
> echo request (encap)
> 11:33:51.358566 www.freenet6.net > 172.16.0.4:
> 3ffe:505:2008:1:2e0:18ff:fea8:4533 >
> jworrells.tsps1.freenet6.net: icmp6:
> echo reply (encap)
Tunnel endpoint is 172.16.0.4 but:
> gif0: flags=8051 mtu 1280
> tunnel inet 166.70.27.6 --> 206.123.31.114
> inet6 fe80::203:47ff:fe98:96f5%gif0 prefixlen 64 scopeid 0x7
> inet6 3ffe:b80:2:f1a6::2 --> 3ffe:b80:2:f1a6::1 prefixlen 128
It's configured as 166.70.27.6.
Change that and it will work:
ifconfig gif0 tunnel 172.16.0.4 206.123.31.114
The kernel throws away your packets as the local endpoint doesn't match.
Greets,
Jeroen
From arien+6bone@ams-ix.net Mon Oct 21 21:45:03 2002
From: arien+6bone@ams-ix.net (Arien Vijn)
Date: Mon, 21 Oct 2002 22:45:03 +0200
Subject: [6bone] Re: IPv6-only IXP's are absolutely wrong
In-Reply-To: <200210211550.g9LFo5N11129@boreas.isi.edu>
Message-ID:
On 21-10-2002 17:50PM, "Bill Manning" wrote:
[..]
>
> reasons to split v4 and v6:
>
> multicast - v4 and v6 treat this differently.
> fabric "optimizations" - framing support, buffer sizing, MTU, etc.
> a decent L2 fabric will be able to accomodate the larger MTUs
> of native v6 and won't complain about divergent framing.
> "Smart" fabrics tend to be tuned to IPv4.
>
Certainly things to consider when choosing your equipment.
> then there are issues wrt RA/ND on an exchange... having all the
> participants trying to "stamp" the fabric with their version of
> which prefix to use is noisy at best and an effective DOS at worst.
>
In practice this is not an issue, at least in my experience. We explicitly
tell everyone that RA messages are not wanted on the shared medium. If you
have a Cisco you have to be suppress them explicitly (we now that). Junipers
you have to switch it on so no problem, same goes for most Zebra boxes.
RA messages are easy to monitor. Once they occur we politely ask the
technical contacts to stop these messages. Up to now we never had a problem
with that. Most are techies quite happy with the attention they get :-)
Even when they occur it doesn't seem a big issue since most routers do not
seem listen to them at all.
> then there are mgmt issues. most L2 fabrics do not have up to date
> v6 mib support, so the stats/traffic collection is not as accurate
> as it should be.
>
This is a quite annoying problem indeed. Not only limited to v6 MIB support
BTW. However I do not see how a separate v6 exchange will solve that.
Except that one just can count the port statistics which is exactly what you
can do when everyone hooks up a separate IPv6 router.
> these things, in addition to the issues with the connecting gear
> (v4 in HW, v6 in SW), tend to argue that a mixed-mode exchange may
> be less stable / harder to troubleshoot than an single use
> facility.
We actually had a separate v6 VLAN for a number of years and concluded that
it was not a problem at all to use both protocols in the same VLAN. The
benefit of doing that is that members can use dual stack routers which is
exactly what most v6 enabled IXPs do.
Kind regards, Arien
--
Arien Vijn
Amsterdam Internet Exchange
http://www.ams-ix.net
From ipv6@worrells.org Mon Oct 21 22:02:04 2002
From: ipv6@worrells.org (Jeremy Worrells)
Date: Mon, 21 Oct 2002 15:02:04 -0600
Subject: [6bone] IPv6 via NAT - problems
References: <001801c27940$c93e8220$210d640a@unfix.org>
Message-ID: <006a01c27945$1ce577e0$020010ac@worrells.org>
Unfortunately, 172.16.0.4 is on a non-routeable subnet that I am using. I
tried using 172.16.0.4 as my address, but freenet6 throws away any request
using 10/8, 172.16/12, or 192.168/16 addresses, as it's not routable. I
thought that using my DSL router's address would let the packets be NATted
and work just fine.
Jeremy
----- Original Message -----
From: "Jeroen Massar"
To: "'Jeremy Worrells'" ; <6bone@ISI.EDU>
Sent: Monday, October 21, 2002 2:31 PM
Subject: RE: [6bone] IPv6 via NAT - problems
> Jeremy Worrells wrote:
>
>
> > 11:33:50.929054 worrells.dsl.xmission.com > www.freenet6.net:
> > jworrells.tsps1.freenet6.net >
> > 3ffe:505:2008:1:2e0:18ff:fea8:4533: icmp6:
> > echo request (encap)
> > 11:33:51.358566 www.freenet6.net > 172.16.0.4:
> > 3ffe:505:2008:1:2e0:18ff:fea8:4533 >
> > jworrells.tsps1.freenet6.net: icmp6:
> > echo reply (encap)
>
> Tunnel endpoint is 172.16.0.4 but:
>
>
>
> > gif0: flags=8051 mtu 1280
> > tunnel inet 166.70.27.6 --> 206.123.31.114
> > inet6 fe80::203:47ff:fe98:96f5%gif0 prefixlen 64 scopeid 0x7
> > inet6 3ffe:b80:2:f1a6::2 --> 3ffe:b80:2:f1a6::1 prefixlen 128
>
> It's configured as 166.70.27.6.
>
> Change that and it will work:
> ifconfig gif0 tunnel 172.16.0.4 206.123.31.114
>
> The kernel throws away your packets as the local endpoint doesn't match.
>
> Greets,
> Jeroen
>
>
> _______________________________________________
> 6bone mailing list
> 6bone@mailman.isi.edu
> http://mailman.isi.edu/mailman/listinfo/6bone
From bmanning@ISI.EDU Mon Oct 21 22:09:18 2002
From: bmanning@ISI.EDU (Bill Manning)
Date: Mon, 21 Oct 2002 14:09:18 -0700 (PDT)
Subject: [6bone] Re: IPv6-only IXP's are absolutely wrong
In-Reply-To: from Arien Vijn at "Oct 21, 2 10:45:03 pm"
Message-ID: <200210212109.g9LL9IG05192@boreas.isi.edu>
% > then there are issues wrt RA/ND on an exchange... having all the
% > participants trying to "stamp" the fabric with their version of
% > which prefix to use is noisy at best and an effective DOS at worst.
% >
%
% In practice this is not an issue, at least in my experience. We explicitly
others have had different experiences. as usual, YMMV, and
being more proactive helps with the social engineering.
% > then there are mgmt issues. most L2 fabrics do not have up to date
% > v6 mib support, so the stats/traffic collection is not as accurate
% > as it should be.
%
% This is a quite annoying problem indeed. Not only limited to v6 MIB support
% BTW. However I do not see how a separate v6 exchange will solve that.
well, on a separate v6 fabric, one does not have the "fudging"
that occurs when the v4 counters get confused w/ v6 stuff.
% Except that one just can count the port statistics which is exactly what you
% can do when everyone hooks up a separate IPv6 router.
it has been entertaining to compare the switch numbers with the
connecting router numbers. you would like them to at least be
in the same ballpark. :)
% > these things, in addition to the issues with the connecting gear
% > (v4 in HW, v6 in SW), tend to argue that a mixed-mode exchange may
% > be less stable / harder to troubleshoot than an single use
% > facility.
%
% We actually had a separate v6 VLAN for a number of years and concluded that
% it was not a problem at all to use both protocols in the same VLAN. The
% benefit of doing that is that members can use dual stack routers which is
% exactly what most v6 enabled IXPs do.
Again, YMMV. Two of my native v6 peers refuse to use their native
v4 boxes because of the performance hit. That and something about
using EFT/beta code in production boxes. :)
Not saying that these issues can't be worked around. Its that there
are real issues here and for somefolks, it may justify an independent
fabric. For others, a separate VLAN will work and for others there
is no problem with dumping all the traffic on the same media.
% Kind regards, Arien
%
--
--bill
From jeroen@unfix.org Mon Oct 21 22:16:49 2002
From: jeroen@unfix.org (Jeroen Massar)
Date: Mon, 21 Oct 2002 23:16:49 +0200
Subject: [6bone] IPv6 via NAT - problems
In-Reply-To: <005d01c27943$f13ef540$020010ac@worrells.org>
Message-ID: <003001c27947$2c2471a0$210d640a@unfix.org>
Jeremy Worrells [mailto:jeremy@worrells.org] wrote:
> Unfortunately, 172.16.0.4 is on a non-routeable subnet that I
> am using. I tried using 172.16.0.4 as my address, but freenet6 throws
> away any request using 10/8, 172.16/12, or 192.168/16 addresses, as
it's not
> routable. I thought that using my DSL router's address would let the
> packets be NATted and work just fine.
You should request the tunnel with your external (public) interfaces IP
address.
Then you should reconfigure your gif0 endpoint to 172.16.0.4.
Check the ifconfig line I gave below, it will work ;)
Greets,
Jeroen
> > Jeremy Worrells wrote:
> >
> >
> > > 11:33:50.929054 worrells.dsl.xmission.com > www.freenet6.net:
> > > jworrells.tsps1.freenet6.net >
> > > 3ffe:505:2008:1:2e0:18ff:fea8:4533: icmp6:
> > > echo request (encap)
> > > 11:33:51.358566 www.freenet6.net > 172.16.0.4:
> > > 3ffe:505:2008:1:2e0:18ff:fea8:4533 >
> > > jworrells.tsps1.freenet6.net: icmp6:
> > > echo reply (encap)
> >
> > Tunnel endpoint is 172.16.0.4 but:
> >
> >
> >
> > > gif0: flags=8051 mtu 1280
> > > tunnel inet 166.70.27.6 --> 206.123.31.114
> > > inet6 fe80::203:47ff:fe98:96f5%gif0 prefixlen 64
> > scopeid 0x7
> > > inet6 3ffe:b80:2:f1a6::2 --> 3ffe:b80:2:f1a6::1
> > prefixlen 128
> >
> > It's configured as 166.70.27.6.
> >
> > Change that and it will work:
> > ifconfig gif0 tunnel 172.16.0.4 206.123.31.114
> >
> > The kernel throws away your packets as the local endpoint
> > doesn't match.
From pekkas@netcore.fi Mon Oct 21 22:19:33 2002
From: pekkas@netcore.fi (Pekka Savola)
Date: Tue, 22 Oct 2002 00:19:33 +0300 (EEST)
Subject: [6bone] IPv6 via NAT - problems
In-Reply-To: <005001c27938$faffe900$020010ac@worrells.org>
Message-ID:
On Mon, 21 Oct 2002, Jeremy Worrells wrote:
> Good afternoon, 6bone!
>
> I am a career sysadmin with a great interest in IPv6. I am trying to get a
> small IPv6 network running at home, using Freenet6 and a FreeBSD 4.6.2
> server. I have the software installed, and the interface comes up. However,
> I cannot get a ping6 or traceroute6 to supply any answers. My Cisco 678 DSL
> router is set up to NAT all protocol 41 and TCP port 4343 traffic to my
> FreeBSD server. Here is the output from a tcpdump on another machine on the
> LAN:
>
> 11:33:50.929054 worrells.dsl.xmission.com > www.freenet6.net:
> jworrells.tsps1.freenet6.net > 3ffe:505:2008:1:2e0:18ff:fea8:4533: icmp6:
> echo request (encap)
> 11:33:51.358566 www.freenet6.net > 172.16.0.4:
> 3ffe:505:2008:1:2e0:18ff:fea8:4533 > jworrells.tsps1.freenet6.net: icmp6:
> echo reply (encap)
>
> It appears that the packets are going to the right places, but are for some
> reason not getting to the IPv6 stack. Here are my interface details:
>
> fxp0: flags=8843 mtu 1500
> inet 172.16.0.4 netmask 0xffffff00 broadcast 172.16.0.255
> inet6 fe80::203:47ff:fe98:96f5%fxp0 prefixlen 64 scopeid 0x1
> ether 00:03:47:98:96:f5
> media: Ethernet autoselect (10baseT/UTP)
> status: active
> gif0: flags=8051 mtu 1280
> tunnel inet 166.70.27.6 --> 206.123.31.114
^^^^^^^^^^^
there's probably a check in the decapsulator that the packets must come
from this/to this address, perhaps you should try changing it to
172.16.0.4 ? Else you have to hack the kernel source, I'm afraid.
Ugly way to get IPv6, but if it works.... :-)
--
Pekka Savola "Tell me of difficulties surmounted,
Netcore Oy not those you stumble over and fall"
Systems. Networks. Security. -- Robert Jordan: A Crown of Swords
From chuck+6bone@snew.com Mon Oct 21 22:21:16 2002
From: chuck+6bone@snew.com (Chuck Yerkes)
Date: Mon, 21 Oct 2002 14:21:16 -0700
Subject: Toning it down. Re: [6bone] FAQ about pTLA request NDSOFTWARE - review closes 23 October 2002
In-Reply-To: ; from tvo@EnterZone.Net on Sun, Oct 20, 2002 at 10:14:20PM -0400
References: <1035133678.4729.343.camel@wks1.fr.corp.ndsoftware.com>
Message-ID: <20021021142116.B28318@snew.com>
Intended generally, to the participants in this discussion:
I'll happily allow that M. DEFFAYET does not use English as his
first language and suggest that highlighting slight misuses of it
is not acceptable in this review phase; instead, asking for and
dealing with the requests and information contained within the words
is appropriate.
Let's re-elevate the discussion back to a civil level without taking
language and the lack of nuance in email into it.
That said, I have a couple thoughts, first being that being a website
hosting/design company that ONLY serves IPv6 would seem a tad
premature to offer with any hope of success.
The next thought being that I too have a CoLo'd machine and one
that, until last friday, was at home on a static IP address. A
move blew the latter situation apart.
I never felt a real need for multiple IP addresses at the Colo
machine, though it does provide tunnels and IPSec endpoints for
several people. I've gotten on just fine with a fairly small
allocation.
I'd much rather have upstream ISPs route my once portable Class C
(24 bit is useless these days).
A pTLA never seemed justified except in the lab, where a fictional
one (blocked at the router) was used to learn a bit about some of
the capabilities of some software and hardware tools we were playing
with.
I've not seen anything to justify a pTLA; where folks I know a
Motorola (one of the larger networks on the planet) and the like
certain DO justify it.
Thank you,
chuck
Quoting John Fraizer (tvo@EnterZone.Net):
> On 20 Oct 2002, Nicolas DEFFAYET wrote:
> > Here a FAQ, no more reply on this list about "pTLA request NDSOFTWARE -
> > review closes 23 October 2002".
> >
> > NDSoftware pTLA request is fully compliant with RFC2772.
> > http://mailman.isi.edu/pipermail/6bone/2002-October/006364.html
> >
> > Don't try to find a bug, there is no bugs.
> > If you are jealous, search another victim.
> >
...
> I don't know about "bugs" in your application but, there are several
> compliance issues, which I have already pointed out in previous emails.
...
> OK. And how do you suppose that you're going to qualify for an sTLA?
>
> >
> > Is Nicolas DEFFAYET is a kid without brain who destroy 6bone ?
> >
> > I'm Nicolas DEFFAYET, i have 4 years, i got every day to school, and i
> > play with my FisherPrice routers and i destroy all 6bone.
> >
>
> I hope you don't mind but, I'm going to use that in my signature line from
> now on. It's so funny!
From ipv6@worrells.org Mon Oct 21 22:22:10 2002
From: ipv6@worrells.org (Jeremy Worrells)
Date: Mon, 21 Oct 2002 15:22:10 -0600
Subject: [6bone] IPv6 via NAT - problems
References: <003001c27947$2c2471a0$210d640a@unfix.org>
Message-ID: <007101c27947$eb55c1a0$020010ac@worrells.org>
Thanks Jeroen! That did the trick. Looks like I'm up!
frink# ping6 www.jp.freebsd.org
PING6(56=40+8+8 bytes) 3ffe:b80:2:f1a6::2 -->
3ffe:505:2008:1:2e0:18ff:fea8:4533
16 bytes from 3ffe:505:2008:1:2e0:18ff:fea8:4533, icmp_seq=0 hlim=56
time=445.559 ms
Fantastic!
Jeremy
----- Original Message -----
From: "Jeroen Massar"
To: "'Jeremy Worrells'"
Cc: "'6bone'" <6bone@isi.edu>
Sent: Monday, October 21, 2002 3:16 PM
Subject: RE: [6bone] IPv6 via NAT - problems
Jeremy Worrells [mailto:jeremy@worrells.org] wrote:
> Unfortunately, 172.16.0.4 is on a non-routeable subnet that I
> am using. I tried using 172.16.0.4 as my address, but freenet6 throws
> away any request using 10/8, 172.16/12, or 192.168/16 addresses, as
it's not
> routable. I thought that using my DSL router's address would let the
> packets be NATted and work just fine.
You should request the tunnel with your external (public) interfaces IP
address.
Then you should reconfigure your gif0 endpoint to 172.16.0.4.
Check the ifconfig line I gave below, it will work ;)
Greets,
Jeroen
> > Jeremy Worrells wrote:
> >
> >
> > > 11:33:50.929054 worrells.dsl.xmission.com > www.freenet6.net:
> > > jworrells.tsps1.freenet6.net >
> > > 3ffe:505:2008:1:2e0:18ff:fea8:4533: icmp6:
> > > echo request (encap)
> > > 11:33:51.358566 www.freenet6.net > 172.16.0.4:
> > > 3ffe:505:2008:1:2e0:18ff:fea8:4533 >
> > > jworrells.tsps1.freenet6.net: icmp6:
> > > echo reply (encap)
> >
> > Tunnel endpoint is 172.16.0.4 but:
> >
> >
> >
> > > gif0: flags=8051 mtu 1280
> > > tunnel inet 166.70.27.6 --> 206.123.31.114
> > > inet6 fe80::203:47ff:fe98:96f5%gif0 prefixlen 64
> > scopeid 0x7
> > > inet6 3ffe:b80:2:f1a6::2 --> 3ffe:b80:2:f1a6::1
> > prefixlen 128
> >
> > It's configured as 166.70.27.6.
> >
> > Change that and it will work:
> > ifconfig gif0 tunnel 172.16.0.4 206.123.31.114
> >
> > The kernel throws away your packets as the local endpoint
> > doesn't match.
From riel@conectiva.com.br Mon Oct 21 22:31:12 2002
From: riel@conectiva.com.br (Rik van Riel)
Date: Mon, 21 Oct 2002 19:31:12 -0200 (BRST)
Subject: [6bone] very drafty draft on 6bone routing mess
In-Reply-To:
Message-ID:
On Mon, 21 Oct 2002, Pekka Savola wrote:
> http://www.netcore.fi/pekkas/ietf/6bone-messv2.txt
I like it. The only thing I'm missing is a way for sites to
easily see if other sites are transit sites or not, ie. should
you accept announced transit paths from a site or not ?
I suppose this can be taken care of by having routing peers
talk to each other, but that might lead to the same lack of
hierarchy the 6bone has today.
It boils down to a way the "serious" (whatever that is) 6bone
sites can exclude transit via the sites that aren't suitable
for transit, without relying on those sites for that. This is
essential since some of the 6bone sites are badly administrated.
Most 6bone sites may be administrated fine, but having routing
messed up by a few badly run sites would suck immensely ;)
kind regards,
Rik
--
Bravely reimplemented by the knights who say "NIH".
http://www.surriel.com/ http://distro.conectiva.com/
Current spamtrap: october@surriel.com
From arien+6bone@ams-ix.net Mon Oct 21 23:56:18 2002
From: arien+6bone@ams-ix.net (Arien Vijn)
Date: Tue, 22 Oct 2002 00:56:18 +0200
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October
2002
In-Reply-To: <20021021082734.GA14333@bfib.colo.bit.nl>
Message-ID:
On 21-10-2002 10:27AM, "Pim van Pelt" wrote:
> On Mon, Oct 21, 2002 at 10:21:33AM +0200, Stephane Bortzmeyer wrote:
> | On Sun, Oct 20, 2002 at 10:14:45PM +0200,
> | Pim van Pelt wrote
> | a message of 76 lines which said:
> |
> | > Last month, I saw a small thread on the lir-wg@ripe.net mailinglist,
> | > where Nicolas complained in this public forum about the fact that the
> | > NCC did not grant him an AS number request.
> |
> | RIPE-NCC gave a stupid reply ("you do not need an AS number for IPv6")
> | and fixed the mistake afterwards with apologies. In that case, it was
> | not Nicolas' fault.
> Stephane,
>
> To be frank: the answer was not stupid, nor was there a need to
> appologise for it. To run IPv6, one does not need an AS number. One
> needs a machine which supports the protocol.
>
True, but this is not the initial response from the NCC. Which was:
"Thank you for your request for an AS number, with IPv6 you do not need to
use any AS numbers nor route objects."
And that is a slightly inaccurate statement isn't it?
Anyway I was rather surprised that Mr Deffayet got his AS number that fast
after all. But let's trust the NCC decided on the merits of the request they
got via his LIR.
Arien
--
Arien Vijn
Amsterdam Internet Exchange
http://www.ams-ix.net
From gert@space.net Tue Oct 22 08:33:55 2002
From: gert@space.net (Gert Doering)
Date: Tue, 22 Oct 2002 09:33:55 +0200
Subject: [6bone] Re: IPv6-only IXP's are absolutely wrong
In-Reply-To: <200210212109.g9LL9IG05192@boreas.isi.edu>; from bmanning@ISI.EDU on Mon, Oct 21, 2002 at 02:09:18PM -0700
References: <200210212109.g9LL9IG05192@boreas.isi.edu>
Message-ID: <20021022093355.Z94537@Space.Net>
Hi,
On Mon, Oct 21, 2002 at 02:09:18PM -0700, Bill Manning wrote:
> % We actually had a separate v6 VLAN for a number of years and concluded that
> % it was not a problem at all to use both protocols in the same VLAN. The
> % benefit of doing that is that members can use dual stack routers which is
> % exactly what most v6 enabled IXPs do.
>
> Again, YMMV. Two of my native v6 peers refuse to use their native
> v4 boxes because of the performance hit. That and something about
> using EFT/beta code in production boxes. :)
On the DECIX, we have a number of participants with Junipers, who do v4+v6
on the same box, and a number of Cisco users, who have separate boxes for
v4 and v6 connectivity.
We are using one VLAN for both, so the Juniper people can save the second
port / second router.
> Not saying that these issues can't be worked around. Its that there
> are real issues here and for somefolks, it may justify an independent
> fabric. For others, a separate VLAN will work and for others there
> is no problem with dumping all the traffic on the same media.
Thanks for your insights, though. We haven't seen any problems yet, but
will closely monitor those issues.
Gert Doering
-- NetMaster
--
Total number of prefixes smaller than registry allocations: 48282 (47686)
SpaceNet AG Mail: netmaster@Space.Net
Joseph-Dollinger-Bogen 14 Tel : +49-89-32356-0
80807 Muenchen Fax : +49-89-32356-299
From chbm@cprm.net Tue Oct 22 10:33:34 2002
From: chbm@cprm.net (Carlos Morgado)
Date: Tue, 22 Oct 2002 10:33:34 +0100
Subject: [6bone] very drafty draft on 6bone routing mess
In-Reply-To: ; from riel@conectiva.com.br on Mon, Oct 21, 2002 at 07:31:12PM -0200
References:
Message-ID: <20021022103334.A5725@cprm.net>
On Mon, Oct 21, 2002 at 07:31:12PM -0200, Rik van Riel wrote:
> On Mon, 21 Oct 2002, Pekka Savola wrote:
>
> > http://www.netcore.fi/pekkas/ietf/6bone-messv2.txt
>
> I like it. The only thing I'm missing is a way for sites to
> easily see if other sites are transit sites or not, ie. should
> you accept announced transit paths from a site or not ?
>
That info can be gathered to some extent from whois. Apart from that
is commercial dealings.
> I suppose this can be taken care of by having routing peers
> talk to each other, but that might lead to the same lack of
> hierarchy the 6bone has today.
>
Internet currently works with peers talking to each other. Also, you
either buy/negotiate transits or you do non-transit peering
so there's not a lot of ambiguity there :)
Do you have other scenarios in mind ?
--
Carlos Morgado - Internet Engineering - Phone +351 214146594
GPG key: 0x75E451E2 FP: B98B 222B F276 18C0 266B 599D 93A1 A3FB 75E4 51E2
The views expressed above do not bind my employer.
From pekkas@netcore.fi Tue Oct 22 11:13:37 2002
From: pekkas@netcore.fi (Pekka Savola)
Date: Tue, 22 Oct 2002 13:13:37 +0300 (EEST)
Subject: [6bone] very drafty draft on 6bone routing mess
In-Reply-To:
Message-ID:
Hello,
Thanks for the comments.
On Mon, 21 Oct 2002, Rik van Riel wrote:
> On Mon, 21 Oct 2002, Pekka Savola wrote:
>
> > http://www.netcore.fi/pekkas/ietf/6bone-messv2.txt
>
> I like it. The only thing I'm missing is a way for sites to
> easily see if other sites are transit sites or not, ie. should
> you accept announced transit paths from a site or not ?
I'm not sure what you mean, exactly.
If I understood your question correctly, this should be done by you, by
observing which paths the neighbor advertises.
After that, if you decide you want only specific, non-transit (at least
for the whole globe) routes, you can apply route-map's in inbound, like:
AS1234 = your peer
AS5678 = your peer's customer
ip as-path access-list 50 ^1234+$
ip as-path access-list 50 ^1234+ 5678+$
(you can leave out '+' if you don't want to allow prepending, of course.)
But I'm not sure if I answered your question...?
[...]
> It boils down to a way the "serious" (whatever that is) 6bone
> sites can exclude transit via the sites that aren't suitable
> for transit, without relying on those sites for that.
Basically, you have to ensure that those sites never get your route,
especially without having no-export/no-advertise community tagged to it or
before having as-path prepended so much it doesn't matter if they
re-advertise it.
That decision is easy when those sites are your direct peers, but unless
they are.. there may be some coordination between the parties involved, or
you could just not advertise the route to them at all (via that path).
We could, of course, try to specify a few default communities like:
: and :, : , like:
6680:666
6680:667
1234:668
and the definition would be either (to be examined by every AS):
1) if your AS is and is "666", do not announce this
to any external peer.
2) if your AS is and is "667" you may only announce
this to external peers using no-export community.
3) if your *neighbor's* AS is and is "668", do not
announce this route to the neighbor at all. (This is particularly nasty
and could be used to create total blockades :-).
Option a) retain this community when advertising to valid peers (could get nasty..)
Option b) clear this community when advertising to valid peers
.. but I'm not sure how useful these would be, as the whole point is
trying to get rid of 6bone routing mess, not try to manage with it (as it
currently is). Also, I'm not sure how widely these would be implemented
(as least 3 would require a lot of work) -- perhaps those who really want
to do the best thing would do it only.
I don't think this can be done in a global scope, and I doubt reserving
specific communities would be accepted by IESG, but in some smaller
scopes, why not..?
> This is
> essential since some of the 6bone sites are badly administrated.
> Most 6bone sites may be administrated fine, but having routing
> messed up by a few badly run sites would suck immensely ;)
Yep, a "solution" try to avoid dealing with upstreams that have anything
to do with those badly administrative systems, but that isn't easy..
--
Pekka Savola "Tell me of difficulties surmounted,
Netcore Oy not those you stumble over and fall"
Systems. Networks. Security. -- Robert Jordan: A Crown of Swords
From bmanning@ISI.EDU Tue Oct 22 16:17:34 2002
From: bmanning@ISI.EDU (Bill Manning)
Date: Tue, 22 Oct 2002 08:17:34 -0700 (PDT)
Subject: [6bone] recursive DNS servers?
Message-ID: <200210221517.g9MFHYC27511@boreas.isi.edu>
a question came up recently that I could not answer.
how many recursive IPv6 transport aware DNS
servers are there?
I have a couple that are "out there" for folks to use
but don't know of any others. Are people running these
things "in the wild" or are they mostly behind walled-gardens?
--
"When in doubt, Twirl..." -anon
From kim@tac.nyc.ny.us Tue Oct 22 16:42:22 2002
From: kim@tac.nyc.ny.us (Kimmo Suominen)
Date: Tue, 22 Oct 2002 11:42:22 -0400
Subject: [6bone] recursive DNS servers?
In-Reply-To: <200210221517.g9MFHYC27511@boreas.isi.edu>
from Bill Manning on Tue, 22 Oct 2002 08:17:34 -0700
References: <200210221517.g9MFHYC27511@boreas.isi.edu>
Message-ID: <20021022154222.283677E4C@beowulf.gw.com>
All of my DNS servers are IPv6-aware/enabled, but only the
non-recursive authoritative servers are accessible from the
general public (grendel.gw.com, morgoth.gw.com, pyry.gw.com).
The registrar I use does not support AAAA records, however,
and I guess most don't (none?)...
If I had an open recursive DNS server, shouldn't I be afraid
to use it myself?
Regards,
+ Kim
| From: Bill Manning
| Date: Tue, 22 Oct 2002 08:17:34 -0700
|
|
| a question came up recently that I could not answer.
|
| how many recursive IPv6 transport aware DNS
| servers are there?
|
| I have a couple that are "out there" for folks to use
| but don't know of any others. Are people running these
| things "in the wild" or are they mostly behind walled-gardens?
|
|
| --
| "When in doubt, Twirl..." -anon
| _______________________________________________
| 6bone mailing list
| 6bone@mailman.isi.edu
| http://mailman.isi.edu/mailman/listinfo/6bone
|
From bmanning@ISI.EDU Tue Oct 22 16:46:26 2002
From: bmanning@ISI.EDU (Bill Manning)
Date: Tue, 22 Oct 2002 08:46:26 -0700 (PDT)
Subject: [6bone] recursive DNS servers?
In-Reply-To: <20021022154222.283677E4C@beowulf.gw.com> from Kimmo Suominen at "Oct 22, 2 11:42:22 am"
Message-ID: <200210221546.g9MFkQ719624@boreas.isi.edu>
% All of my DNS servers are IPv6-aware/enabled, but only the
% non-recursive authoritative servers are accessible from the
% general public (grendel.gw.com, morgoth.gw.com, pyry.gw.com).
thanks.
% The registrar I use does not support AAAA records, however,
% and I guess most don't (none?)...
not in production - yet.
when I still did .int, we supported AAAA records.
% If I had an open recursive DNS server, shouldn't I be afraid
% to use it myself?
Not at all. :)
%
% Regards,
% + Kim
%
%
% | From: Bill Manning
% | Date: Tue, 22 Oct 2002 08:17:34 -0700
% |
% |
% | a question came up recently that I could not answer.
% |
% | how many recursive IPv6 transport aware DNS
% | servers are there?
% |
% | I have a couple that are "out there" for folks to use
% | but don't know of any others. Are people running these
% | things "in the wild" or are they mostly behind walled-gardens?
% |
% |
% | --
% | "When in doubt, Twirl..." -anon
% | _______________________________________________
% | 6bone mailing list
% | 6bone@mailman.isi.edu
% | http://mailman.isi.edu/mailman/listinfo/6bone
% |
%
--
--bill
From gert@space.net Tue Oct 22 17:19:12 2002
From: gert@space.net (Gert Doering)
Date: Tue, 22 Oct 2002 18:19:12 +0200
Subject: [6bone] recursive DNS servers?
In-Reply-To: <200210221517.g9MFHYC27511@boreas.isi.edu>; from bmanning@ISI.EDU on Tue, Oct 22, 2002 at 08:17:34AM -0700
References: <200210221517.g9MFHYC27511@boreas.isi.edu>
Message-ID: <20021022181912.Q94537@Space.Net>
Hi,
On Tue, Oct 22, 2002 at 08:17:34AM -0700, Bill Manning wrote:
> a question came up recently that I could not answer.
>
> how many recursive IPv6 transport aware DNS
> servers are there?
>
> I have a couple that are "out there" for folks to use
> but don't know of any others. Are people running these
> things "in the wild" or are they mostly behind walled-gardens?
Our official recursive name server, ns2.space.net, is doing v4+v6
transport (DJB dnscache plus IPv6 patches).
It's available only for our customers, so it's partly "in the wild" and
partly "walled garden".
Gert Doering
-- NetMaster
--
Total number of prefixes smaller than registry allocations: 48282 (47686)
SpaceNet AG Mail: netmaster@Space.Net
Joseph-Dollinger-Bogen 14 Tel : +49-89-32356-0
80807 Muenchen Fax : +49-89-32356-299
From david@IPRG.nokia.com Tue Oct 22 18:55:59 2002
From: david@IPRG.nokia.com (David Kessens)
Date: Tue, 22 Oct 2002 10:55:59 -0700
Subject: sTLA alloc policies [Re: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002]
In-Reply-To: ; from pekkas@netcore.fi on Sun, Oct 20, 2002 at 05:36:10PM +0300
References: <20021020142528.X94537@Space.Net>
Message-ID: <20021022105559.G29674@iprg.nokia.com>
Pekka,
On Sun, Oct 20, 2002 at 05:36:10PM +0300, Pekka Savola wrote:
>
> Oh, Nokia must also have colored the truth slightly..
What are you trying to insinuate here ?!?
Please refrain from such comments if you don't know the details.
We got our address space under the old rules. Despite this, it really
shouldn't be too hard for any large multinational company to show
plans for assigning address space to 200 other organizations. No need
to color the truth at all.
David K.
---
From john@sixgirls.org Tue Oct 22 23:50:34 2002
From: john@sixgirls.org (John Klos)
Date: Tue, 22 Oct 2002 18:50:34 -0400 (EDT)
Subject: [6bone] recursive DNS servers?
In-Reply-To: <200210221517.g9MFHYC27511@boreas.isi.edu>
Message-ID:
Hello,
> a question came up recently that I could not answer.
>
> how many recursive IPv6 transport aware DNS
> servers are there?
>
> I have a couple that are "out there" for folks to use
> but don't know of any others. Are people running these
> things "in the wild" or are they mostly behind walled-gardens?
I don't suppose that there are that many yet, but NetBSD's installer has a
built-in list of IPv6 name servers because you can do a full install
entirely via IPv6. I suppose that while those servers are certainly
willing to allow connections for people installing NetBSD, you might want
to ask each of the server's administrators if they would allow other uses.
In the meanwhile, SCL's main server acts as a DNS and NTP server for many
networks in New York City and elsewhere, and may be used as a public IPv6
server:
reva.sixgirls.org 3ffe:b80:133c:1::1
John Klos
Sixgirls Computing Labs
From dlc-6bone@halibut.com Wed Oct 23 00:22:40 2002
From: dlc-6bone@halibut.com (David Carmean)
Date: Tue, 22 Oct 2002 16:22:40 -0700
Subject: [6bone] recursive DNS servers?
In-Reply-To: <200210221517.g9MFHYC27511@boreas.isi.edu>; from bmanning@ISI.EDU on Tue, Oct 22, 2002 at 08:17:34AM -0700
References: <200210221517.g9MFHYC27511@boreas.isi.edu>
Message-ID: <20021022162240.A26255@halibut.com>
On Tue, Oct 22, 2002 at 08:17:34AM -0700, Bill Manning wrote:
>
> a question came up recently that I could not answer.
>
> how many recursive IPv6 transport aware DNS
> servers are there?
A semi-related question: how many IPv4 transport
authoritative nameservers are out there that don't just
blackhole AAAA queries? Doubleclick is the biggest
thorn in my web-browsing side at the moment. When I'm
using Mozilla on my 6bone-connected machine, ad-rich
sites are incredibly slow to load because of all the
timeouts.
From cfaber@fpsn.net Wed Oct 23 01:05:40 2002
From: cfaber@fpsn.net (Colin Faber)
Date: Tue, 22 Oct 2002 18:05:40 -0600
Subject: [6bone] recursive DNS servers?
References: <200210221517.g9MFHYC27511@boreas.isi.edu> <20021022162240.A26255@halibut.com>
Message-ID: <3DB5E7D4.D4BD8A35@fpsn.net>
I would suggest using the "Block images from this server" option in
mozilla on such sites. It's really just a work around until doubleclick
figures out how much the IPv6 community needs to see their ads.
:-)
David Carmean wrote:
>
> On Tue, Oct 22, 2002 at 08:17:34AM -0700, Bill Manning wrote:
> >
> > a question came up recently that I could not answer.
> >
> > how many recursive IPv6 transport aware DNS
> > servers are there?
>
> A semi-related question: how many IPv4 transport
> authoritative nameservers are out there that don't just
> blackhole AAAA queries? Doubleclick is the biggest
> thorn in my web-browsing side at the moment. When I'm
> using Mozilla on my 6bone-connected machine, ad-rich
> sites are incredibly slow to load because of all the
> timeouts.
>
> _______________________________________________
> 6bone mailing list
> 6bone@mailman.isi.edu
> http://mailman.isi.edu/mailman/listinfo/6bone
--
Colin Faber
(303) 736-5160
fpsn.net, Inc.
* Black holes are where God divided by zero. *
From fink@es.net Wed Oct 23 02:28:58 2002
From: fink@es.net (Bob Fink)
Date: Tue, 22 Oct 2002 18:28:58 -0700
Subject: [6bone] pTLA request NDSOFTWARE - change of review date to 30 October
2002
Message-ID: <5.1.0.14.0.20021022182536.030eca68@imap2.es.net>
6bone Folk,
When I announced the pTLA review for NDSOFTWARE I stated the closing date
incorrectly as 23 October (only 7 days). It should be 30 October (14 days).
My fault. Sorry.
Thanks,
Bob
From hansolofalcon@worldnet.att.net Wed Oct 23 03:51:19 2002
From: hansolofalcon@worldnet.att.net (Gregg C Levine)
Date: Tue, 22 Oct 2002 22:51:19 -0400
Subject: [6bone] pTLA request NDSOFTWARE - change of review date to 30 October 2002
In-Reply-To: <5.1.0.14.0.20021022182536.030eca68@imap2.es.net>
Message-ID: <000f01c27a3f$10fdec00$b658580c@who>
Hello from Gregg C Levine
Indeed. I noticed that. Apology accepted. But I am still inclined to
insist that this request be denied.
-------------------
Gregg C Levine hansolofalcon@worldnet.att.net
------------------------------------------------------------
"The Force will be with you...Always." Obi-Wan Kenobi
"Use the Force, Luke." Obi-Wan Kenobi
(This company dedicates this E-Mail to General Obi-Wan Kenobi )
(This company dedicates this E-Mail to Master Yoda )
> -----Original Message-----
> From: 6bone-admin@mailman.isi.edu [mailto:6bone-admin@mailman.isi.edu]
On
> Behalf Of Bob Fink
> Sent: Tuesday, October 22, 2002 9:29 PM
> To: 6BONE List
> Subject: [6bone] pTLA request NDSOFTWARE - change of review date to 30
> October 2002
>
> 6bone Folk,
>
> When I announced the pTLA review for NDSOFTWARE I stated the closing
date
> incorrectly as 23 October (only 7 days). It should be 30 October (14
days).
> My fault. Sorry.
>
>
> Thanks,
>
> Bob
>
> _______________________________________________
> 6bone mailing list
> 6bone@mailman.isi.edu
> http://mailman.isi.edu/mailman/listinfo/6bone
From berni@birkenwald.de Wed Oct 23 08:58:22 2002
From: berni@birkenwald.de (Bernhard Schmidt)
Date: Wed, 23 Oct 2002 09:58:22 +0200
Subject: [6bone] recursive DNS servers?
In-Reply-To: <200210221517.g9MFHYC27511@boreas.isi.edu>
References: <200210221517.g9MFHYC27511@boreas.isi.edu>
Message-ID: <20021023075822.GA46092@thor.birkenwald.de>
On Tue, Oct 22, 2002 at 08:17:34AM -0700, Bill Manning wrote:
> how many recursive IPv6 transport aware DNS
> servers are there?
We have an authorative and a recursive IPv6-enabled nameserver out there
waiting for clients. Both using BIND 9.2.2rc1 with FreeBSD 4.(6|7).
Resolver: resns6.eurocyber.net 2001:768:1:3::2
Authorative: authns6.eurocyber.net 2001:768:1:3::4
Both free for use from everywhere.
--
bye bye
Bernhard
From feico@pasta.cs.uit.no Wed Oct 23 15:22:28 2002
From: feico@pasta.cs.uit.no (Feico Dillema)
Date: Wed, 23 Oct 2002 16:22:28 +0200
Subject: [6bone] recursive DNS servers?
In-Reply-To: <200210221517.g9MFHYC27511@boreas.isi.edu>
References: <200210221517.g9MFHYC27511@boreas.isi.edu>
Message-ID: <20021023142228.GV3253@pasta.cs.uit.no>
On Tue, Oct 22, 2002 at 08:17:34AM -0700, Bill Manning wrote:
> a question came up recently that I could not answer.
>
> how many recursive IPv6 transport aware DNS
> servers are there?
>
> I have a couple that are "out there" for folks to use
> but don't know of any others. Are people running these
> things "in the wild" or are they mostly behind walled-gardens?
dns.pasta.cs.uit.no is recursive and authorative and currently open
and "in the wild".
Feico.
From andreas.bergstrom@hiof.no Wed Oct 23 16:34:35 2002
From: andreas.bergstrom@hiof.no (=?iso-8859-1?Q?Andreas_Bergstr=F8m?=)
Date: Wed, 23 Oct 2002 17:34:35 +0200 (CEST)
Subject: [6bone] recursive DNS servers?
In-Reply-To: <20021023142228.GV3253@pasta.cs.uit.no>
Message-ID:
Today, Feico Dillema uttered these words:
> dns.pasta.cs.uit.no is recursive and authorative and currently open
> and "in the wild".
The same goes for shienar.ipv6.hiof.no
May you live long and spamless,
Andreas Bergstrøm
--
My favourite Linux distro: http://www.mandrake-linux.com
'Carpe diem and reach for the stars!'
Can't see my attachments, please read:
http://odin.hiof.no/~andreasb/outlook.html
From lamia@ati.tn Wed Oct 23 19:37:45 2002
From: lamia@ati.tn (Lamia Chaffai)
Date: Wed, 23 Oct 2002 19:37:45 +0100
Subject: [6bone] pTLA request NDSOFTWARE - review closes 23 October 2002
References: <000001c277be$45fe9900$5c59580c@who> <1035121435.4771.54.camel@wks1.fr.corp.ndsoftware.com>
Message-ID: <3DB6EC79.9000105@ati.tn>
--------------090105000307020306000203
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Dear list members,
First of all, on behalf of the Tunisian Internet Agency (ATI) team, I
would like to thank all 6bone mailing list members for their generous
help and support to foster experimentation and deployment of IPv6 by all
the Internet community.
We also would like to take this opportunity to react to the recent
E-mail from Mr. Nicolas Defayett since ATI was mentioned as one of its
tunnel customers.
In fact, Mr Defayett helped us in establishing our first tunnel with
6Bone but we faced some trouble with IPv6 routing since he is using a
private ASN. The reason why we established a second tunnel with XS26 as
a transition phase.
And now we've gone a step forward by sending a Sub-TLA request to RIPE
to get our IPv6 production address block.
We hope that by this we bring some clarifications regarding our
participation in the 6Bone Network.
Best Regards
Lamia Chaffai
Tunisian Internet Agency
http://www.ati.tn
http://www.ipv6net.tn
Nicolas DEFFAYET wrote:
>On Sun, 2002-10-20 at 00:24, Gregg C Levine wrote:
>
>Hello Gregg,
>
>
>
>>I have been lurking on this list, for a good number of years now.
>>Sometimes even posting a comment, or a gripe. This is more along the
>>lines of both. I have been monitoring the traffic discussing
>>NDSOFTWARE's request. Both finding the original message to Bob Fink, and
>>the list, and everything. Sorry Mr. Deffayet, I disagree. For one, you
>>do need to spell out who will be using the service. Is it for your
>>company? Yourself? What? Who, even? Unless you can spell out neatly the
>>answers to my questions, I am inclined to agree with everyone else. I am
>>also agreeing with the people I have disagreed with early on. I might
>>also, add, even Master Yoda's methods of speaking isn't that confusing.
>>
>>
>
>Who will be using the service:
>
>- NDSoftware (my company)
>
>- many projects (here a list of main projects):
>
>* IPv6-FR
>A non profit organisation for the developement of IPv6 in France
>IPv6-FR run a tunnel broker and have currently 200 users, each user have
>a /48
>=> NDSoftware provide to IPv6-FR: 1 /35 and a native IPv6 connectivity.
>
>* NextGenCollective
>IPv6 research
>http://www.nextgencollective.net/
>NGC[NextGen Collective] provides IPv6-over-IPv4 tunnels to people all
>over the world.
>ASpath-tree: http://www.nextgencollective.net/bgp4/ (AS65526 is my old
>private ASN)
>=> NDSoftware provide to NextGenCollective: 2 /44, 1 /40, 1 /36 and a
>tunnel IPv6 over IPv4 with BGP (full transit).
>
>* IPng.org.uk
>IPv6 tunnel broker
>http://www.ipng.org.uk/
>ASpath-tree: http://www.ipng.org.uk/bgp/bgp-page-complete.html (AS65526
>is my old private ASN)
>=> NDSoftware provide to IPng.org.uk: 1 /44, 1 /40 and a tunnel IPv6
>over IPv4 with BGP (full transit).
>
>* ILS
>Italian Linux Society
>ILS provide IPv6 connectivity to italian user groups and organizations
>experimenting with IPv6.
>ILS host the IPv6 IRC server calvino.freenode.net
>=> NDSoftware provide to ILS: 1 /44 and a tunnel IPv6 over IPv4 with BGP
>(full transit).
>
>* ATI
>A tunisian ISP
>http://www.ipv6net.tn/
>http://www.ipv6net.tn/ipv6-Tunisia.pdf
>=> NDSoftware provide to ATI: 1 /44 and a tunnel IPv6 over IPv4 with BGP
>(full transit). NDSoftware have help ATI for the IPv6 deployement in
>tunisia. ATI plan later to request a pTLA.
>
>* FABIONNE
>A projet for do IPv6 Debian package
>http://debian-ipv6.fabionne.net/
>=> NDSoftware provide to FABIONNE: a tunnel IPv6 over IPv4 with BGP
>(full transit) and host a mirror for this projet
>(http://debian-ipv6.mirrors.ndsoftwarenet.com/).
>
>ESMT
>An university in Senegal
>=> NDSoftware provide to ESMT: 1 /44 and a tunnel IPv6 over IPv4.
>
>
>Best regards,
>
>Nicolas DEFFAYET
>
>_______________________________________________
>6bone mailing list
>6bone@mailman.isi.edu
>http://mailman.isi.edu/mailman/listinfo/6bone
>
>
>
>
--------------090105000307020306000203
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit
Dear list members,
First of all, on behalf of the Tunisian Internet Agency (ATI) team, I would
like to thank all 6bone mailing list members for their generous help and
support to foster experimentation and deployment of IPv6 by all the Internet
community.
We also would like to take this opportunity to react to the recent E-mail
from Mr. Nicolas Defayett since ATI was mentioned as one of its tunnel
customers.
In fact, Mr Defayett helped us in establishing our first tunnel with 6Bone
but we faced some trouble with IPv6 routing since he is using a private
ASN. The reason why we established a second tunnel with XS26 as a transition
phase.
And now we've gone a step forward by sending a Sub-TLA request to RIPE to
get our IPv6 production address block.
We hope that by this we bring some clarifications regarding our participation
in the 6Bone Network.
Best Regards
Lamia Chaffai
Tunisian Internet Agency
http://www.ati.tn
http://www.ipv6net.tn
Nicolas DEFFAYET wrote:
On Sun, 2002-10-20 at 00:24, Gregg C Levine wrote:
Hello Gregg,
I have been lurking on this list, for a good number of years now.
Sometimes even posting a comment, or a gripe. This is more along the
lines of both. I have been monitoring the traffic discussing
NDSOFTWARE's request. Both finding the original message to Bob Fink, and
the list, and everything. Sorry Mr. Deffayet, I disagree. For one, you
do need to spell out who will be using the service. Is it for your
company? Yourself? What? Who, even? Unless you can spell out neatly the
answers to my questions, I am inclined to agree with everyone else. I am
also agreeing with the people I have disagreed with early on. I might
also, add, even Master Yoda's methods of speaking isn't that confusing.
Who will be using the service:
- NDSoftware (my company)
- many projects (here a list of main projects):
* IPv6-FR
A non profit organisation for the developement of IPv6 in France
IPv6-FR run a tunnel broker and have currently 200 users, each user have
a /48
=> NDSoftware provide to IPv6-FR: 1 /35 and a native IPv6 connectivity.
* NextGenCollective
IPv6 research
http://www.nextgencollective.net/
NGC[NextGen Collective] provides IPv6-over-IPv4 tunnels to people all
over the world.
ASpath-tree: http://www.nextgencollective.net/bgp4/ (AS65526 is my old
private ASN)
=> NDSoftware provide to NextGenCollective: 2 /44, 1 /40, 1 /36 and a
tunnel IPv6 over IPv4 with BGP (full transit).
* IPng.org.uk
IPv6 tunnel broker
http://www.ipng.org.uk/
ASpath-tree: http://www.ipng.org.uk/bgp/bgp-page-complete.html (AS65526
is my old private ASN)
=> NDSoftware provide to IPng.org.uk: 1 /44, 1 /40 and a tunnel IPv6
over IPv4 with BGP (full transit).
* ILS
Italian Linux Society
ILS provide IPv6 connectivity to italian user groups and organizations
experimenting with IPv6.
ILS host the IPv6 IRC server calvino.freenode.net
=> NDSoftware provide to ILS: 1 /44 and a tunnel IPv6 over IPv4 with BGP
(full transit).
* ATI
A tunisian ISP
http://www.ipv6net.tn/
http://www.ipv6net.tn/ipv6-Tunisia.pdf
=> NDSoftware provide to ATI: 1 /44 and a tunnel IPv6 over IPv4 with BGP
(full transit). NDSoftware have help ATI for the IPv6 deployement in
tunisia. ATI plan later to request a pTLA.
* FABIONNE
A projet for do IPv6 Debian package
http://debian-ipv6.fabionne.net/
=> NDSoftware provide to FABIONNE: a tunnel IPv6 over IPv4 with BGP
(full transit) and host a mirror for this projet
(http://debian-ipv6.mirrors.ndsoftwarenet.com/).
ESMT
An university in Senegal
=> NDSoftware provide to ESMT: 1 /44 and a tunnel IPv6 over IPv4.
Best regards,
Nicolas DEFFAYET
_______________________________________________
6bone mailing list
6bone@mailman.isi.edu
http://mailman.isi.edu/mailman/listinfo/6bone
--------------090105000307020306000203--
From Tom@blazing.de Wed Oct 23 20:25:43 2002
From: Tom@blazing.de (Tom Spier)
Date: Wed, 23 Oct 2002 21:25:43 +0200
Subject: [6bone] recursive DNS servers?
References:
Message-ID: <002101c27ac9$fbb985d0$6405a8c0@2be>
Wednesday, October 23, 2002 5:34 PM Andreas Bergstrøm wrote:
> Today, Feico Dillema uttered these words:
>
> > dns.pasta.cs.uit.no is recursive and authorative and currently open
> > and "in the wild".
>
> The same goes for shienar.ipv6.hiof.no
>
> May you live long and spamless,
>
> Andreas Bergstrøm
>
> --
ns1.blazing.de open&available, too.
Greets
Tom
From riel@conectiva.com.br Wed Oct 23 21:54:04 2002
From: riel@conectiva.com.br (Rik van Riel)
Date: Wed, 23 Oct 2002 18:54:04 -0200 (BRST)
Subject: [6bone] recursive DNS servers?
In-Reply-To: <200210221517.g9MFHYC27511@boreas.isi.edu>
Message-ID:
On Tue, 22 Oct 2002, Bill Manning wrote:
>
> a question came up recently that I could not answer.
>
> how many recursive IPv6 transport aware DNS
> servers are there?
NL.linux.org and imladris.surriel.com are up and running.
Rik
--
Bravely reimplemented by the knights who say "NIH".
http://www.surriel.com/ http://distro.conectiva.com/
Current spamtrap: october@surriel.com
From pekkas@netcore.fi Wed Oct 23 22:12:29 2002
From: pekkas@netcore.fi (Pekka Savola)
Date: Thu, 24 Oct 2002 00:12:29 +0300 (EEST)
Subject: [6bone] Re: sTLA alloc policies
In-Reply-To: <20021022105559.G29674@iprg.nokia.com>
Message-ID:
On Tue, 22 Oct 2002, David Kessens wrote:
> On Sun, Oct 20, 2002 at 05:36:10PM +0300, Pekka Savola wrote:
> >
> > Oh, Nokia must also have colored the truth slightly..
>
> What are you trying to insinuate here ?!?
> Please refrain from such comments if you don't know the details.
Indeed, the applications are not public so I do not, unfortunately, know
details :-(
> We got our address space under the old rules.
Ah, I didn't notice this.
> Despite this, it really
> shouldn't be too hard for any large multinational company to show
> plans for assigning address space to 200 other organizations. No need
> to color the truth at all.
You must be using some other definition of other organizations than I do.
Further, I don't believe there are even 200 countries out there. :-)
Let's see. Extending your interpretation any company with 200 employees
could be entitled to a block: they _do_ want to provide xDSL service and
proper addresses to their employees (who are private users) using the
recommended /48 assignment!
Right..
--
Pekka Savola "Tell me of difficulties surmounted,
Netcore Oy not those you stumble over and fall"
Systems. Networks. Security. -- Robert Jordan: A Crown of Swords
From tony@lava.net Thu Oct 24 10:03:00 2002
From: tony@lava.net (Antonio Querubin)
Date: Wed, 23 Oct 2002 23:03:00 -1000 (HST)
Subject: [6bone] recursive DNS servers?
In-Reply-To: <200210221517.g9MFHYC27511@boreas.isi.edu>
Message-ID:
On Tue, 22 Oct 2002, Bill Manning wrote:
> a question came up recently that I could not answer.
>
> how many recursive IPv6 transport aware DNS
> servers are there?
>
> I have a couple that are "out there" for folks to use
> but don't know of any others. Are people running these
> things "in the wild" or are they mostly behind walled-gardens?
We had 2 until yesterday: ns1.ipv6.lava.net, ns2.ipv6.lava.net. The ns1
box is undergoing an OS upgrade this week.
From dlc-6bone@halibut.com Thu Oct 24 20:33:46 2002
From: dlc-6bone@halibut.com (David Carmean)
Date: Thu, 24 Oct 2002 12:33:46 -0700
Subject: [6bone] recursive DNS servers?
In-Reply-To: ; from john@sixgirls.org on Tue, Oct 22, 2002 at 08:55:25PM -0400
References: <3DB5E7D4.D4BD8A35@fpsn.net>
Message-ID: <20021024123346.B16206@halibut.com>
On Tue, Oct 22, 2002 at 08:55:25PM -0400, John Klos wrote:
> Hi,
>
> > I would suggest using the "Block images from this server" option in
> > mozilla on such sites. It's really just a work around until doubleclick
> > figures out how much the IPv6 community needs to see their ads.
>
> I simply added my own zone for doubleclick and for a few other nasty ad
> servers. At first I used my main web server's IP address, but sometimes
> people would get the whole page updated with the Apache error message, so
> I changed the IP address they all point to to 127.0.0.1. Instant time-out
> and no full page errors.
I can do this for my personal (home DSL/work desktop) nameservers,
but it's probably not appropriate for me to do so on my production
nameservers, especially since one of my employer's products is a web
cache and we're integrating and testing IPv6 code.
I hacked up a perl script to test public (authoritative) nameserver
behavior w.r.t. A vs. AAAA queries over IPv4. Extracting hostnames
from our production HTTP cache logs, for an 80-minute period, I found
the following:
Of 7,131 unique hostnames, for which I queried 6,808 unique listed
nameservers,
Queries for 36 of those hostnames produced valid A responses but
had timeouts for AAAA. Only about 1/4 of those seem to be dedicated
spam^W ad/banner servers, but they have more "weight" due to their
usage pattern.
23 queries resulted in a NOERROR reponse for A, but NXDOMAIN for AAAA.
17 queries resulted in a valid response for A, but SERVFAIL for AAAA.
1 query resulted in a NOERROR reponse for A, but NOTIMP for
AAAA. (That server also answers NOTIMP for CH TXT version.bind query.)
Maybe it's time to hack a "match-query-type" option into BIND 9's
"view" concept? :o)
From david@IPRG.nokia.com Thu Oct 24 23:57:37 2002
From: david@IPRG.nokia.com (David Kessens)
Date: Thu, 24 Oct 2002 15:57:37 -0700
Subject: [6bone] Re: sTLA alloc policies (fwd)
In-Reply-To: ; from pekkas@netcore.fi on Thu, Oct 24, 2002 at 09:24:13AM +0300
References:
Message-ID: <20021024155737.F827@iprg.nokia.com>
Pekka,
On Thu, Oct 24, 2002 at 09:24:13AM +0300, ext Pekka Savola wrote:
>
> I wonder what interpretation of "other organizations" is used when RIPE
> NCC evaluates the applications..?
Ask the RIPE NCC, fill in an application yourself, ask other people
how they qualified (that's why some smart people hire a consult with
experience with this kind applications to help them out).
> ---------- Forwarded message ----------
> Date: Thu, 24 Oct 2002 00:12:29 +0300 (EEST)
> From: Pekka Savola
> To: David Kessens
> Cc: 6bone@ISI.EDU
> Subject: [6bone] Re: sTLA alloc policies
>
> On Tue, 22 Oct 2002, David Kessens wrote:
> > On Sun, Oct 20, 2002 at 05:36:10PM +0300, Pekka Savola wrote:
> > >
> > > Oh, Nokia must also have colored the truth slightly..
> >
> > What are you trying to insinuate here ?!?
> > Please refrain from such comments if you don't know the details.
>
> Indeed, the applications are not public so I do not, unfortunately, know
> details :-(
so don't spread rumors about us coloring the truth while you have no
proof of any wrongdoing.
> > We got our address space under the old rules.
>
> Ah, I didn't notice this.
again, please check the facts before you send a mail.
> > Despite this, it really
> > shouldn't be too hard for any large multinational company to show
> > plans for assigning address space to 200 other organizations. No need
> > to color the truth at all.
>
> You must be using some other definition of other organizations than I do.
> Further, I don't believe there are even 200 countries out there. :-)
>
> Let's see. Extending your interpretation any company with 200 employees
> could be entitled to a block: they _do_ want to provide xDSL service and
> proper addresses to their employees (who are private users) using the
> recommended /48 assignment!
No. No. No!
I didn't say that 'other organizations' are equal to one employee. You
are making that up in your own fantasy world.
You seem to be living in a very simple world. Do you have any idea how
multi-national organizations work ?!? They have partnerships, joint
ventures, cooperations with other organizations, are member of
associations, support non-profit organizations, are part of standards
bodies etc.. Many of these organizations are not owned or operated in
any kind of way by the multi-national organization and still the
multi-national organization can provide ip addresses and connectivity
to such 'other organizations'.
David K.
PS also, instead of suggesting there might not be 200 countries, you
could have done some more research:
the UN has 191 memberstates.
the number of countries in the world is close to this number,
depending on your definition of what constitutes a country.
So you are correct that there are less than 200 countries, but
it is close enough to 200 that it really doesn't help you.
---
From pekkas@netcore.fi Fri Oct 25 08:38:32 2002
From: pekkas@netcore.fi (Pekka Savola)
Date: Fri, 25 Oct 2002 10:38:32 +0300 (EEST)
Subject: [6bone] Re: sTLA alloc policies (fwd)
In-Reply-To: <20021024155737.F827@iprg.nokia.com>
Message-ID:
On Thu, 24 Oct 2002, David Kessens wrote:
> On Thu, Oct 24, 2002 at 09:24:13AM +0300, ext Pekka Savola wrote:
> >
> > I wonder what interpretation of "other organizations" is used when RIPE
> > NCC evaluates the applications..?
>
> Ask the RIPE NCC, fill in an application yourself, ask other people
> how they qualified (that's why some smart people hire a consult with
> experience with this kind applications to help them out).
I don't think RIPE NCC would respond to queries like that with useful
answers, or people would tell exactly how they qualified (or often colored
the truth). In any case, we already have sTLA so there is no need to fill
anything.
> > > What are you trying to insinuate here ?!?
> > > Please refrain from such comments if you don't know the details.
> >
> > Indeed, the applications are not public so I do not, unfortunately, know
> > details :-(
>
> so don't spread rumors about us coloring the truth while you have no
> proof of any wrongdoing.
Depending on the interpretation of "other organizations", the behaviour is
self-evidently "questionable" or self-evidently "okay".
> > > Despite this, it really
> > > shouldn't be too hard for any large multinational company to show
> > > plans for assigning address space to 200 other organizations. No need
> > > to color the truth at all.
> >
> > You must be using some other definition of other organizations than I do.
> > Further, I don't believe there are even 200 countries out there. :-)
> >
> > Let's see. Extending your interpretation any company with 200 employees
> > could be entitled to a block: they _do_ want to provide xDSL service and
> > proper addresses to their employees (who are private users) using the
> > recommended /48 assignment!
>
> No. No. No!
>
> I didn't say that 'other organizations' are equal to one employee. You
> are making that up in your own fantasy world.
I didn't say you said so: read again. I said that is relatively logical
extension of the arguments you said. "employees" are private persons,
different legal entities.
> You seem to be living in a very simple world. Do you have any idea how
> multi-national organizations work ?!? They have partnerships, joint
> ventures, cooperations with other organizations, are member of
> associations, support non-profit organizations, are part of standards
> bodies etc.. Many of these organizations are not owned or operated in
> any kind of way by the multi-national organization and still the
> multi-national organization can provide ip addresses and connectivity
> to such 'other organizations'.
No, I really don't have that much idea.
I'm not saying that using multiple /48 for big multinational organizations
is easy. Far from it. It must be more difficult than some small ISP's
getting sTLA. But that's how the current rules are.
> PS also, instead of suggesting there might not be 200 countries, you
> could have done some more research:
> the UN has 191 memberstates.
> the number of countries in the world is close to this number,
> depending on your definition of what constitutes a country.
> So you are correct that there are less than 200 countries, but
> it is close enough to 200 that it really doesn't help you.
I don't waste my time on irrelevant research, which is why I used 'might
not be'. Whether there is 190 or 210 doesn't really matter.
--
Pekka Savola "Tell me of difficulties surmounted,
Netcore Oy not those you stumble over and fall"
Systems. Networks. Security. -- Robert Jordan: A Crown of Swords
From tvo@EnterZone.Net Fri Oct 25 11:15:51 2002
From: tvo@EnterZone.Net (John Fraizer)
Date: Fri, 25 Oct 2002 06:15:51 -0400 (EDT)
Subject: [6bone] private ASNs and the Default-Free-Zone
Message-ID:
This is not a problem:
ipv6-site: COMPENDIUM-AR
origin: AS45328
descr: Compendium, Buenos Aires, AR
country: AR
prefix: 3FFE:8260::/28
*THIS* is a problem:
Border2-BGP> sh ipv6 bgp 3ffe:8260:2010:1:2a0:c9ff:fe01:9600
BGP routing table entry for 3ffe:8260::/28
Paths: (11 available, best #8, table Default-IP-Routing-Table)
1930 2200 5511 1752 1849 1890 45328
3ffe:31ff:0:ffff::50 from 3ffe:31ff:0:ffff::50 (193.136.2.250)
(fe80::c188:202)
Origin IGP, localpref 100, valid, external
Last update: Fri Oct 25 04:21:38 2002
24765 1849 1890 45328
3ffe:4005:0:1::26 from 3ffe:4005:0:1::26 (62.24.229.1)
Origin IGP, localpref 100, valid, external
Community: 24765:100 24765:750 24765:1000 24765:6001
Last update: Fri Oct 25 04:20:33 2002
8002 5594 8277 45328
3ffe:80c0:200:5::36 from 3ffe:80c0:200:5::36 (206.252.222.79)
Origin IGP, localpref 100, valid, external
Last update: Fri Oct 25 04:20:29 2002
4554 6939 8379 8277 45328
2001:4f0:1001:1::2 from 2001:4f0:1001:1::2 (192.0.1.1)
(fe80::c620:401)
Origin IGP, metric 1, localpref 100, valid, external
Last update: Fri Oct 25 03:54:56 2002
6435 6939 8379 8277 45328
3ffe:8160:0:1::c from 3ffe:8160:0:1::c (64.65.64.152)
(fe80::4041:4098)
Origin IGP, localpref 100, valid, external
Last update: Fri Oct 25 03:47:04 2002
109 6939 8379 8277 45328
3ffe:c00:8023:4::1 from 3ffe:c00:8023:4::1 (128.107.240.254)
(fe80::806b:f0fe)
Origin IGP, localpref 100, valid, external
Come on, if Nicolas can get an ASN, so can COMPENDIUM.
Beyond that, if you peer with someone who uses a private ASN, use the
following command (or equiv for your router) on the peering session:
neighbor 3ffe:xxxx::xxxx remove-private-AS
If your router code doesn't support that command or one like it, might I
suggest that you UPGRADE?
---
John Fraizer | High-Security Datacenter Services |
President | Dedicated circuits 64k - 155M OC3 |
EnterZone, Inc | Virtual, Dedicated, Colocation |
http://www.enterzone.net/ | Network Consulting Services |
From cfaber@fpsn.net Fri Oct 25 11:48:45 2002
From: cfaber@fpsn.net (Colin Faber)
Date: Fri, 25 Oct 2002 04:48:45 -0600
Subject: [6bone] Reverse DNS with BIND and IPv6
Message-ID: <3DB9218D.C85EF042@fpsn.net>
Hi folks,
Just thought i'd let anyone that's interested know that I've put
together a simple web based cgi system to aid in the building of
IPv6 IP6.INT records for BIND:
http://tools.fpsn.net/ipv6-inaddr/
--
Colin Faber
(303) 736-5160
fpsn.net, Inc.
* Black holes are where God divided by zero. *
From gert@space.net Fri Oct 25 13:45:58 2002
From: gert@space.net (Gert Doering)
Date: Fri, 25 Oct 2002 14:45:58 +0200
Subject: [6bone] private ASNs and the Default-Free-Zone
In-Reply-To: ; from tvo@EnterZone.Net on Fri, Oct 25, 2002 at 06:15:51AM -0400
References:
Message-ID: <20021025144558.S94537@Space.Net>
Hi,
On Fri, Oct 25, 2002 at 06:15:51AM -0400, John Fraizer wrote:
> This is not a problem:
>
> ipv6-site: COMPENDIUM-AR
> origin: AS45328
> descr: Compendium, Buenos Aires, AR
> country: AR
> prefix: 3FFE:8260::/28
I disagree. It's not a major problem, like the one below, but I think
this object should not be there either.
> *THIS* is a problem:
>
>
> Border2-BGP> sh ipv6 bgp 3ffe:8260:2010:1:2a0:c9ff:fe01:9600
> BGP routing table entry for 3ffe:8260::/28
> Paths: (11 available, best #8, table Default-IP-Routing-Table)
>
> 1930 2200 5511 1752 1849 1890 45328
Strongly seconded.
[..]
> Come on, if Nicolas can get an ASN, so can COMPENDIUM.
And if you don't have an AS#, use a private AS, don't just grab any number
that seems to be available.
> Beyond that, if you peer with someone who uses a private ASN, use the
> following command (or equiv for your router) on the peering session:
>
> neighbor 3ffe:xxxx::xxxx remove-private-AS
And make sure that you never ever do transit through a private AS# - it
will really break everything related to BGP paths, like "find a short
path", or "troubleshoot weird problems".
Even better, never give transit to a private AS# either.
Give that enterprise connectivity, yes, but do it static, and don't
mess with BGP origin AS manipulations.
Gert Doering
-- NetMaster
--
Total number of prefixes smaller than registry allocations: 48540 (48282)
SpaceNet AG Mail: netmaster@Space.Net
Joseph-Dollinger-Bogen 14 Tel : +49-89-32356-0
80807 Muenchen Fax : +49-89-32356-299
From basit@basit.cc Fri Oct 25 19:11:01 2002
From: basit@basit.cc (Abdul Basit)
Date: Fri, 25 Oct 2002 13:11:01 -0500 (CDT)
Subject: [6bone] Re: pTLA request NDSOFTWARE - review closes 23 October 2002
Message-ID:
Hello folks@6bone ,
I like to favour Mr Deffayet on his pTLA request, I am speaking on behalf
of NextGenCollective (NGC), NGC is a reasearch organization that provides
ipv6 tunnels to over 150 users , We have NDSOFTWARE as our ipv6
upstream. NGC is housed at WSU (Wichita state university), USA, As a
matter of fact, in very begining , we were unable to find any support
of ipv6 , it was Mr Deffayet who helped us a lot, so its wrong to say that
he is an offensive person, I see him working almost all the time
supporting his customers. and not to mention NDSOFTWARE does supports
many research based ipv6 projects, NGC, and IPng are one of them.
and NGC in turns plan to deploy IPv6 on whole WSU network infrastructure.
We also plan to tunnel with KU and KSU in cojunction with HiPECC
(http://www.hipecc.twsu.edu) Internet2 project at WSU. Even we are
analyzing the feasibility of deploying IPv6 on laptops issued to students.
NGC is also providing IPSec tunnels to those who wants with dynamic updates.
We are also planning to look for feasbility for establishing a mobile
cluster using ipv6.
http://www.linuxsymposium.org/2002/view_txt.php?text=abstract&talk=36
As a matter of fact, being on irc is fun, you meet nice persons
and get help, it always the case with me. On the other hand, persons
working in cisco, motorolla, Nokia are of no real help, I tried to contact
some persons in Nokia (listed at http://www.nokia.com/ipv6/ ) several
times , no reply i got, but if i email Mr Deffayet for some question or
ask for some support, i usually get reply on the same day. I give credit
to Mr Deffayet of what is currently being done under NGC as he is the one
who is supporting that research work. I beleive NOKIA , cisco and other
BIG giants don't support IPv6 work at universities that much that
NDSOFTWARE is trying to do.
My opinion is , if NDSOFTWARE's request is fully RFC2772 compliant,
there should be no objection of approving its pTLA 's request, or
ppl who are against it should provide strong reasons in terms of RFC2772
to oppose the request. Its not ONLY the right of BIG organizations
(like Nokia, Cisco etc) to become pTLA, other relatively small
organization can serve as a pTLA 's also if they obey the rules
stated in RFC. We should encourage all.
- basit
Graduate Student
MS Computer Science
Wichita state university
http://basit.cc
http://ip6.basit.cc
From nicolas.deffayet@ndsoftware.net Fri Oct 25 13:58:35 2002
From: nicolas.deffayet@ndsoftware.net (Nicolas DEFFAYET)
Date: 25 Oct 2002 14:58:35 +0200
Subject: [6bone] Reverse DNS with BIND and IPv6
In-Reply-To: <3DB9218D.C85EF042@fpsn.net>
References: <3DB9218D.C85EF042@fpsn.net>
Message-ID: <1035550715.590.63.camel@wks1.fr.corp.ndsoftware.com>
On Fri, 2002-10-25 at 12:48, Colin Faber wrote:
Hi Colin,
> Just thought i'd let anyone that's interested know that I've put
> together a simple web based cgi system to aid in the building of
> IPv6 IP6.INT records for BIND:
>
> http://tools.fpsn.net/ipv6-inaddr/
>
Do you plan to add too a .arpa support ?
Best Regards,
Nicolas DEFFAYET
From pekkas@netcore.fi Fri Oct 25 14:03:26 2002
From: pekkas@netcore.fi (Pekka Savola)
Date: Fri, 25 Oct 2002 16:03:26 +0300 (EEST)
Subject: [6bone] I-D ACTION:draft-savola-v6ops-6bone-mess-00.txt (fwd)
Message-ID:
I updated this slightly over v2 version.
Comments etc. of course appreciated.
--
Pekka Savola "Tell me of difficulties surmounted,
Netcore Oy not those you stumble over and fall"
Systems. Networks. Security. -- Robert Jordan: A Crown of Swords
---------- Forwarded message ----------
Date: Fri, 25 Oct 2002 07:28:14 -0400
From: Internet-Drafts@ietf.org
To: IETF-Announce: ;
Cc: v6ops@ops.ietf.org
Subject: I-D ACTION:draft-savola-v6ops-6bone-mess-00.txt
A New Internet-Draft is available from the on-line Internet-Drafts directories.
Title : Moving from 6bone to IPv6 Internet
Author(s) : P. Savola
Filename : draft-savola-v6ops-6bone-mess-00.txt
Pages : 13
Date : 2002-10-24
Currently, IPv6 Internet is, globally considered, inseparable from
the 6bone network. The 6bone has been built as a tighly meshed
tunneled topology. As the number of participants has grown, it has
become an untangible mess, hindering the real deployment of IPv6 due
to low quality of connections. This memo discusses the nature and
the state of 6bone/IPv6 Internet, points out problems and outlines a
few ways to start fixing them; also, some rough operational
guidelines for different-sized organizations are presented. The most
important issues are: not offering transit to everyone and real
transit operators being needed to take a more active role.
A URL for this Internet-Draft is:
http://www.ietf.org/internet-drafts/draft-savola-v6ops-6bone-mess-00.txt
To remove yourself from the IETF Announcement list, send a message to
ietf-announce-request with the word unsubscribe in the body of the message.
Internet-Drafts are also available by anonymous FTP. Login with the username
"anonymous" and a password of your e-mail address. After logging in,
type "cd internet-drafts" and then
"get draft-savola-v6ops-6bone-mess-00.txt".
A list of Internet-Drafts directories can be found in
http://www.ietf.org/shadow.html
or ftp://ftp.ietf.org/ietf/1shadow-sites.txt
Internet-Drafts can also be obtained by e-mail.
Send a message to:
mailserv@ietf.org.
In the body type:
"FILE /internet-drafts/draft-savola-v6ops-6bone-mess-00.txt".
NOTE: The mail server at ietf.org can return the document in
MIME-encoded form by using the "mpack" utility. To use this
feature, insert the command "ENCODING mime" before the "FILE"
command. To decode the response(s), you will need "munpack" or
a MIME-compliant mail reader. Different MIME-compliant mail readers
exhibit different behavior, especially when dealing with
"multipart" MIME messages (i.e. documents which have been split
up into multiple messages), so check your local documentation on
how to manipulate these messages.
Below is the data which will enable a MIME compliant mail reader
implementation to automatically retrieve the ASCII version of the
Internet-Draft.
From nicolas.deffayet@ndsoftware.net Fri Oct 25 14:13:00 2002
From: nicolas.deffayet@ndsoftware.net (Nicolas DEFFAYET)
Date: 25 Oct 2002 15:13:00 +0200
Subject: [6bone] private ASNs and the Default-Free-Zone
In-Reply-To:
References:
Message-ID: <1035551580.585.80.camel@wks1.fr.corp.ndsoftware.com>
On Fri, 2002-10-25 at 12:15, John Fraizer wrote:
> This is not a problem:
>
> ipv6-site: COMPENDIUM-AR
> origin: AS45328
> descr: Compendium, Buenos Aires, AR
> country: AR
> prefix: 3FFE:8260::/28
Do you think that it's normal to allocate a pTLA with an unallocated ASN
?
> Beyond that, if you peer with someone who uses a private ASN, use the
> following command (or equiv for your router) on the peering session:
>
> neighbor 3ffe:xxxx::xxxx remove-private-AS
>
> If your router code doesn't support that command or one like it, might I
> suggest that you UPGRADE?
remove-private-AS will remove the private ASN in ASpath, not the route
with private ASN...
Exemple:
3ffe:ffff::/32
1 2 3 65000
If AS3 use remove-private-AS, other network will get this:
3ffe:ffff::/32
1 2 3
AS3 is not the source of 3ffe:ffff::/32, the source is 65000
=> private ASN _MUST_ send their routes with the community no-export
(like i do before)
Using this for don't announce route with private ASN is better:
ip as-path access-list private-asn-out deny
_(6451[2-9]|645[2-9][0-9]|64[6-9][0-9][0-9]|65[0-4][0-9][0-9]|655[0-2][0-9]|6553[0-5])_
ip as-path access-list private-asn-out permit .*
Best Regards,
Nicolas DEFFAYET
From bortzmeyer@gitoyen.net Fri Oct 25 14:29:28 2002
From: bortzmeyer@gitoyen.net (Stephane Bortzmeyer)
Date: Fri, 25 Oct 2002 15:29:28 +0200
Subject: [6bone] Reverse DNS with BIND and IPv6
In-Reply-To: <3DB9218D.C85EF042@fpsn.net>
References: <3DB9218D.C85EF042@fpsn.net>
Message-ID: <20021025132928.GA15363@nic.fr>
On Fri, Oct 25, 2002 at 04:48:45AM -0600,
Colin Faber wrote
a message of 18 lines which said:
> Just thought i'd let anyone that's interested know that I've put
> together a simple web based cgi system to aid in the building of
> IPv6 IP6.INT records for BIND:
>
> http://tools.fpsn.net/ipv6-inaddr/
I do not find it as useful as the command-line tool ipv6calc which can
do many other fun things.
ftp://ftp.bieringer.de/pub/linux/IPv6/ipv6calc
~ % ipv6calc --addr_to_ip6arpa 3ffe:b80:138c::/48
c.8.3.1.0.8.b.0.e.f.f.3.ip6.arpa.
From raphit@noemie.org Fri Oct 25 14:30:48 2002
From: raphit@noemie.org (Raphael Bouaziz)
Date: Fri, 25 Oct 2002 15:30:48 +0200
Subject: [6bone] private ASNs and the Default-Free-Zone
In-Reply-To: ; from tvo@EnterZone.Net on Fri, Oct 25, 2002 at 06:15:51AM -0400
References:
Message-ID: <20021025153048.A3495@noemie.org>
On Fri, Oct 25, 2002, John Fraizer wrote:
> 109 6939 8379 8277 45328
> 3ffe:c00:8023:4::1 from 3ffe:c00:8023:4::1 (128.107.240.254)
> (fe80::806b:f0fe)
> Origin IGP, localpref 100, valid, external
>
>
> Come on, if Nicolas can get an ASN, so can COMPENDIUM.
>
>
> Beyond that, if you peer with someone who uses a private ASN, use the
> following command (or equiv for your router) on the peering session:
But note that AS45328 does not belong to the private AS numbers
range (AS64512 - AS65535).
So, such announcements are totally bogus, _even internal_.
--
Raphael Bouaziz.
raphit@noemie.org - http://noemie.nerim.net/
Sysadmin Power Forever(TM).
From riel@conectiva.com.br Fri Oct 25 15:05:05 2002
From: riel@conectiva.com.br (Rik van Riel)
Date: Fri, 25 Oct 2002 12:05:05 -0200 (BRST)
Subject: [6bone] private ASNs and the Default-Free-Zone
In-Reply-To: <1035551580.585.80.camel@wks1.fr.corp.ndsoftware.com>
Message-ID: