IPv6 bolloxed again and other problems
John Naismith
AAISP machines just like the last time.
I _do_ have to say that AAISP is getting flakier by the week on a
number of things - DNS resolution/accuracy and routing being prime
amongst them. These problems don't appear to be getting resolved
either and have been ongoing for some considerable time now.
I can work around DNS problems but routing problems are outwith my
control on ADSL. They are now so commonplace on MFNX that I have lists
of proxies to hand so that I can tunnel traffic when (note the when,
_not_ if) their routing falls apart. It is a daily occurrence now
which I don't even bother reporting unless it's like that for 24
hours+. For all I know it could well be your BGP implementation
falling apart - I don't know. All I do know is that _every day_ there
are chunks of the internet that disappear via AAISP - destination
unreachable. It is _invariably_ the case that the address blocks
concerned are _not_ unroutable and in fact don't have _any_ problem at
all.
You need to start sorting out the basics Adrian. You have the
following problems which affect me regularly :
1) Primary and secondary DNS machines are all too frequently reporting
different lookup results - sometimes for days! There's a post in here
_every_ week about it;
2) MFNX is (not to put too fine a point on it) bloody awful at the
moment. Either that or you have bolloxed up BGP. I don't know which is
the case as I don't deal with MFNX. Other transit suppliers appear to
have no issues when I check via them so it's either you or MFNX.
Either way it is getting more than tedious;
3) You have a strange IPv6 tunnel setup which simply doesn't work with
WindowsXP, 2k or 2k3 and results in the above issue. If it is working
as intended then _please_ say so and then we can make appropriate
decisions. Do bear in mind that tunnels are _meant_ to work point(s)
to point(s) though and not accept return traffic from undefined IP
addresses?
I hope you take this in the manner it was intended but to be honest
I'm not that bothered at the moment as to whether you do or not. You
have basic issues which require attention _now_ so do you think you
could _please_ concentrate on them?
Thanks.
NB - As I'm aware that this group is "monitored" by several entities
I'd like to say that these issues are not something that average users
of a "normal" ISP would even notice. However we're not "average
users", this isn't a "normal ISP" and as such they are not acceptable.
--
John Naismith
Jim Crowther
<john$E200...@naismith.org.uk> writes:
>3) You have a strange IPv6 tunnel setup which simply doesn't work with
>WindowsXP, 2k or 2k3 and results in the above issue. If it is working
>as intended then _please_ say so and then we can make appropriate
>decisions. Do bear in mind that tunnels are _meant_ to work point(s)
>to point(s) though and not accept return traffic from undefined IP
>addresses?
My ipv6 setup fails regularly every 24hrs, (this is of course a clue
it's a problem my end). However I re-boot, all is well.
Running local DNS (not advertised). listen-on-v6 { any; }; is set in
named.conf.
(XPpro, Treewalk, ZAPro, non-NAT router, all firewalls *I think*
allowing ipv6 i/o, at least named.exe has full rights.)
I've been looking at some ipv6 mailing-lists, are we going to be stuck
with nothing happening, or are ISPs (AAISP of course) doing things a
particular way? It all looks disappointingly like unnc.
--
Jim Crowther "It's MY computer" (tm SMG)
Avoid Swen by disposing of any previous Usenet addresses,
put 'spam' or 'delete' somewhere in the Reply-to: header.
Help yourself avoid the spam: <http://keir.net/k9.html>
Julie Brandon
> I _do_ have to say that AAISP is getting flakier by the week on a
> number of things - DNS resolution/accuracy
Agreed! Because of this we're now using our own cache anyway.
> and routing being prime
> amongst them.
I disagree, as far as IPv4 is concerned (not quite doing IPv6 yet so can't
comment on that.) In my experience IPv4 routing has been pretty much rock
solid. Any difficulties I've had have been intermittant and outside of
A&A, or remote-server-load related (and hence obviously also outside of
A&A.)
--
Julie Brandon http://www.computergeeks.co.uk/
_______________________________________________________________________________
>>=- WEBPAGE UPDATE: 360' panoramic virtual tour [incl. 8 panoramas] -=<<
John Naismith
<nos...@honeypot.computergeeks.me.uk.invalid> wrote:
>On Mon, 22 Mar 2004 01:20:58 +0000, John Naismith (john$E200...@naismith.org.uk) said:
>> I _do_ have to say that AAISP is getting flakier by the week on a
>> number of things - DNS resolution/accuracy
>
>Agreed! Because of this we're now using our own cache anyway.
>
>> and routing being prime
>> amongst them.
>
>I disagree, as far as IPv4 is concerned (not quite doing IPv6 yet so can't
>comment on that.) In my experience IPv4 routing has been pretty much rock
>solid. Any difficulties I've had have been intermittant and outside of
>A&A.)
As in MFNX like I said Julie? Unless you can see what is going on at
AAISPs BGP interface then you can't tell whether it's AAISP or MFNX. I
_strongly_ suspect it is the latter but until alternate transit is _in
place_ then it's impossible for a user to tell.
--
John Naismith
John Naismith
<Don't_bo...@blackhole.do-not-spam.me.uk> wrote:
>In message <dmds50lfmr9sefngf...@4ax.com>, John Naismith
><john$E200...@naismith.org.uk> writes:
>
>>3) You have a strange IPv6 tunnel setup which simply doesn't work with
>>WindowsXP, 2k or 2k3 and results in the above issue. If it is working
>>as intended then _please_ say so and then we can make appropriate
>>decisions. Do bear in mind that tunnels are _meant_ to work point(s)
>>to point(s) though and not accept return traffic from undefined IP
>>addresses?
>
>My ipv6 setup fails regularly every 24hrs, (this is of course a clue
>it's a problem my end). However I re-boot, all is well.
>
>Running local DNS (not advertised). listen-on-v6 { any; }; is set in
>named.conf.
This isn't a DNS issue Jim. It's because (for internal AAISP
addresses) Adrian has set it up so that a machine _other_ than the
defined tunnel endpoint is replying. External addresses work fine.
It wouldn't really be a problem other than the fact that all the AAISP
servers are on IPv6 so you have to wait for the ipv6 timeout (varies
depending on stack) and then fallback to ipv4. Irritating does _not_
begin to cover it ;-)
--
John Naismith
Rev Adrian Kennard
> On 22 Mar 2004 04:34:22 GMT, Julie Brandon
> <nos...@honeypot.computergeeks.me.uk.invalid> wrote:
>
>
>>On Mon, 22 Mar 2004 01:20:58 +0000, John Naismith (john$E200...@naismith.org.uk) said:
>>
>>>I _do_ have to say that AAISP is getting flakier by the week on a
>>>number of things - DNS resolution/accuracy
I thought the DNS issues were resolved. Are people saying there is still
a problem?
>>Agreed! Because of this we're now using our own cache anyway.
>>
>>
>>>and routing being prime
>>>amongst them.
>>
>>I disagree, as far as IPv4 is concerned (not quite doing IPv6 yet so can't
>>comment on that.) In my experience IPv4 routing has been pretty much rock
>>solid. Any difficulties I've had have been intermittant and outside of
>>A&A.)
>
> As in MFNX like I said Julie? Unless you can see what is going on at
> AAISPs BGP interface then you can't tell whether it's AAISP or MFNX. I
> _strongly_ suspect it is the latter but until alternate transit is _in
> place_ then it's impossible for a user to tell.
The cables were meant to be in last week - it really is any day now.
--
_ Rev. Adrian Kennard, Andrews & Arnold Ltd / AAISP
(_) _| _ . _ _ Broadband, fixed IPs, no min term http://adsl.ms/
( )(_|( |(_|| ) Asterisk VoIP based PABXs, SNOM200 http://aa.gg/
~~~~~~~~~~~~~~~~ Bond two ADSL lines? http://www.FireBrick.info/
Rev Adrian Kennard
>...
> 1) Primary and secondary DNS machines are all too frequently reporting
> different lookup results - sometimes for days! There's a post in here
> _every_ week about it;
I was under the impression the DNS problems we had were sorted.
Can you give me an example I can actually work on?
> 2) MFNX is (not to put too fine a point on it) bloody awful at the
> moment. Either that or you have bolloxed up BGP. I don't know which is
> the case as I don't deal with MFNX. Other transit suppliers appear to
> have no issues when I check via them so it's either you or MFNX.
> Either way it is getting more than tedious;
I agree. Level 3 will be on line any day now.
> 3) You have a strange IPv6 tunnel setup which simply doesn't work with
> WindowsXP, 2k or 2k3 and results in the above issue. If it is working
> as intended then _please_ say so and then we can make appropriate
> decisions. Do bear in mind that tunnels are _meant_ to work point(s)
> to point(s) though and not accept return traffic from undefined IP
> addresses?
Twice endless has lost the ability to see the local interface for IP6. I
do not yet know why this is. I am looking in to this.
The fact that traffic routed to 217.169.20.1 works is not the solution
to the problem. I need to find why endless is losing local routing.
> I hope you take this in the manner it was intended but to be honest
> I'm not that bothered at the moment as to whether you do or not. You
> have basic issues which require attention _now_ so do you think you
> could _please_ concentrate on them?
Certainly.
> NB - As I'm aware that this group is "monitored" by several entities
> I'd like to say that these issues are not something that average users
> of a "normal" ISP would even notice. However we're not "average
> users", this isn't a "normal ISP" and as such they are not acceptable.
Thanks.
John Naismith
>John Naismith wrote:
>>...
>> 1) Primary and secondary DNS machines are all too frequently reporting
>> different lookup results - sometimes for days! There's a post in here
>> _every_ week about it;
>
>I was under the impression the DNS problems we had were sorted.
>Can you give me an example I can actually work on?
Well Message-ID: <405da01b$0$63625$5a6a...@news.aaisp.net.uk> was
certainly true at the time the post was made. Several other similar
posts have been made in the last month or so. I haven't seen _any_
response from yourself or other AAISP staff although I may have missed
them. This is (IMO) a different issue from the one which you sorted
out a couple of weeks ago.
>> 2) MFNX is (not to put too fine a point on it) bloody awful at the
>> moment. Either that or you have bolloxed up BGP. I don't know which is
>> the case as I don't deal with MFNX. Other transit suppliers appear to
>> have no issues when I check via them so it's either you or MFNX.
>> Either way it is getting more than tedious;
>
>I agree. Level 3 will be on line any day now.
Having used Level 3 in the past I dunno whether to say "good" or not
;-) A second transit supplier (almost regardless of who it is) is
badly needed so it'll have to be a qualified "good".
>> 3) You have a strange IPv6 tunnel setup which simply doesn't work with
>> WindowsXP, 2k or 2k3 and results in the above issue. If it is working
>> as intended then _please_ say so and then we can make appropriate
>> decisions. Do bear in mind that tunnels are _meant_ to work point(s)
>> to point(s) though and not accept return traffic from undefined IP
>> addresses?
>
>Twice endless has lost the ability to see the local interface for IP6. I
>do not yet know why this is. I am looking in to this.
>
>The fact that traffic routed to 217.169.20.1 works is not the solution
>to the problem. I need to find why endless is losing local routing.
Ah my apologies then. I interpreted your response to the previous
occurrence of this problem incorrectly. I suspect endless has in fact
been losing the local interface rather more than twice. I'm 99%
certain this has been happening for 3 months or so now. Fixing it
would be much appreciated although if its intermittent I quite
understand the difficulties.
--
John Naismith
Rev Adrian Kennard
Hash: SHA1
John Naismith wrote:
>...
> Well Message-ID: <405da01b$0$63625$5a6a...@news.aaisp.net.uk> was
> certainly true at the time the post was made. Several other similar
Sorry, what was the subject?
- --
Rev Adrian Kennard
Andrews & Arnold Ltd / AAISP www.aaisp.net.uk
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFAXrVxHBb4e52L0Y0RAkIwAJ0f9B1Uq0wpX0pPcuGwxf45KSNlNwCfYnsN
xZf64kKaUkzsmWx4lQDmP+M=
=cGUR
-----END PGP SIGNATURE-----
John Naismith
<john$E200...@naismith.org.uk> wrote:
>On Mon, 22 Mar 2004 08:29:09 +0000, Rev Adrian Kennard <`@o.gg> wrote:
>
>>John Naismith wrote:
>>>...
>>> 1) Primary and secondary DNS machines are all too frequently reporting
>>> different lookup results - sometimes for days! There's a post in here
>>> _every_ week about it;
>>
>>I was under the impression the DNS problems we had were sorted.
>>Can you give me an example I can actually work on?
>
>Well Message-ID: <405da01b$0$63625$5a6a...@news.aaisp.net.uk> was
>certainly true at the time the post was made.
Oh and I saw something similar with the domain which Andrew registered
for me last week - the MX10 was OK on one of the DNS servers and not
the other. I had intended to contact you about it as it'd been like
that for more than 24 hours but a minor crisis got in the way and when
I checked again the DNS servers (correctly) agreed on the MX10 result.
--
John Naismith
John Naismith
>-----BEGIN PGP SIGNED MESSAGE-----
>Hash: SHA1
>
>John Naismith wrote:
>>...
>> Well Message-ID: <405da01b$0$63625$5a6a...@news.aaisp.net.uk> was
>> certainly true at the time the post was made. Several other similar
>
>Sorry, what was the subject?
Heh - get a newsreader that can deal with message ID's ;-)
Cut and paste follows:
Path: news.aaisp.net.uk!news-peer-lilac.gradwell.net!not-for-mail
From: "Alastair" <em...@address.invalid>
Newsgroups: uk.net.providers.aaisp
Subject: DNS
Date: Sun, 21 Mar 2004 14:01:12 -0000
X-Priority: 3
X-MSMail-Priority: Normal
X-Newsreader: Microsoft Outlook Express 6.00.2800.1158
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1165
Lines: 12
Message-ID: <405da01b$0$63625$5a6a...@news.aaisp.net.uk>
NNTP-Posting-Date: 21 Mar 2004 14:00:59 GMT
NNTP-Posting-Host: 81.2.107.176
X-Trace: 1079877659 news.aaisp.net.uk 63625 81.2.107.176
X-Complaints-To: aaispne...@gradwell.net
Xref: news.aaisp.net.uk uk.net.providers.aaisp:3251
We started the transfer of the domain HWCA.COM to A&A
nameservers last Wednesday.
Currently 217.169.20.21 is correctly giving HWCA.COM and
WWW.HWCA.COM as 81.2.67.202 (from A records on A&A
nameservers) whereas 217.169.20.20 is still giving out
212.67.202.35 (from A records on HostEurope nameservers).
This seems to be taking longer than usual. Has something
gone wrong? Or, more likely, have we done something wrong?
--
John Naismith
Rev Adrian Kennard
Hash: SHA1
John Naismith wrote:
> On Mon, 22 Mar 2004 09:44:17 +0000, Rev Adrian Kennard <`@o.gg> wrote:
>
>
>>-----BEGIN PGP SIGNED MESSAGE-----
>>Hash: SHA1
>>
>>John Naismith wrote:
>>
>>>...
>>>Well Message-ID: <405da01b$0$63625$5a6a...@news.aaisp.net.uk> was
>>>certainly true at the time the post was made. Several other similar
>>
>>Sorry, what was the subject?
>
>
> Heh - get a newsreader that can deal with message ID's ;-)
Indeed.
> Currently 217.169.20.21 is correctly giving HWCA.COM and
> WWW.HWCA.COM as 81.2.67.202 (from A records on A&A
> nameservers) whereas 217.169.20.20 is still giving out
> 212.67.202.35 (from A records on HostEurope nameservers).
>
> This seems to be taking longer than usual. Has something
> gone wrong? Or, more likely, have we done something wrong?
This is one I sorted manually this morning. It is down to caching times,
and depending on the old name server settings, etc, a transfer can take
a few days to catch up.
- --
Rev Adrian Kennard
Andrews & Arnold Ltd / AAISP www.aaisp.net.uk
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFAXrfRHBb4e52L0Y0RApXuAJ9OMsIqsJpJ3SnaB3emPKaQVQmTWACglAdG
RZBxPiPoTEKa6kEaE0BzG0E=
=vgrZ
-----END PGP SIGNATURE-----
j...@coris.org.uk
>>WindowsXP, 2k or 2k3 and results in the above issue. If it is working
>>as intended then _please_ say so and then we can make appropriate
>>decisions. Do bear in mind that tunnels are _meant_ to work point(s)
>>to point(s) though and not accept return traffic from undefined IP
>>addresses?
Baring in mind that "6to4" isn't quite a tunnel but more a way of
encapsulating v4 packets, I can't see the problem in sending to one v4 IP
and receiving from different ones. Anyway, isn't it possible to send to
either endless or edgeless?
> My ipv6 setup fails regularly every 24hrs, (this is of course a clue
> it's a problem my end). However I re-boot, all is well.
I've had no problems at all with my IPv6 setup. I've been using it
regularly for several months and transferring a fair amount of data over
it (e.g a 100Mb+ CVS update). Hmm, I wonder what adding the v6 traffic
would do to my usage graph ...
J
--
My other computer also runs NetBSD / Sailing at Newbiggin
http://www.netbsd.org/ / http://www.newbigginsailingclub.org/
Julie Brandon
> As in MFNX like I said Julie?
Only very temporary/transient issues.
> Unless you can see what is going on at
> AAISPs BGP interface then you can't tell whether it's AAISP or MFNX. I
> _strongly_ suspect it is the latter but until alternate transit is _in
> place_ then it's impossible for a user to tell.
IMHO it'll be mfnx, I remember what they were like when I left Demon. *:(
Rev Adrian Kennard
Hash: SHA1
j...@coris.org.uk wrote:
>>>3) You have a strange IPv6 tunnel setup which simply doesn't work with
>>>WindowsXP, 2k or 2k3 and results in the above issue. If it is working
>>>as intended then _please_ say so and then we can make appropriate
>>>decisions. Do bear in mind that tunnels are _meant_ to work point(s)
>>>to point(s) though and not accept return traffic from undefined IP
>>>addresses?
>
>
> Baring in mind that "6to4" isn't quite a tunnel but more a way of
> encapsulating v4 packets, I can't see the problem in sending to one v4 IP
> and receiving from different ones. Anyway, isn't it possible to send to
> either endless or edgeless?
Yes, except that for no apparent reason endless is losing local routing,
but edgeless (aimless+needless) is not. endless was rebooted this
morning to resolve this, but we have to find the real cause yet.
>>My ipv6 setup fails regularly every 24hrs, (this is of course a clue
>>it's a problem my end). However I re-boot, all is well.
>
> I've had no problems at all with my IPv6 setup. I've been using it
> regularly for several months and transferring a fair amount of data over
> it (e.g a 100Mb+ CVS update). Hmm, I wonder what adding the v6 traffic
> would do to my usage graph ...
<-:
- --
Rev Adrian Kennard
Andrews & Arnold Ltd / AAISP www.aaisp.net.uk
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFAXu4YHBb4e52L0Y0RAux4AJ0cpSETZpPawkeFuRyIUAq0CLqC6wCfS6sE
pOfMLcGFFfA44CshVSIn9VQ=
=UVef
-----END PGP SIGNATURE-----
John Naismith
>>>3) You have a strange IPv6 tunnel setup which simply doesn't work with
>>>WindowsXP, 2k or 2k3 and results in the above issue. If it is working
>>>as intended then _please_ say so and then we can make appropriate
>>>decisions. Do bear in mind that tunnels are _meant_ to work point(s)
>>>to point(s) though and not accept return traffic from undefined IP
>>>addresses?
>
>Baring in mind that "6to4" isn't quite a tunnel but more a way of
>encapsulating v4 packets, I can't see the problem in sending to one v4 IP
>and receiving from different ones.
I think I've said this before but 6to4 and v6v4 are not the same thing
under Windows. v6v4 _is_ a tunnel. In fact it's a tunnel with its' own
unique interface (and hence routing) on the Windows box. 6to4 is a
pseudo-interface which is preset to use MS to route ipv6 traffic.
I am _more_ than willing to be corrected on this - in fact I wish
somebody _would_ correct me. However please don't bother unless you
have managed to get Windows XP or 2k3 to work in the way you suggest.
I can't do it and I don't see anyone else managing it so far.
Don't take this personally it's just (as others can attest) getting
Windows XP/2k3 to route ipv6 over ipv4 (using v6v4) is a bitch due to
the lack of documentation. Once you figure it out then it's only 5 or
6 lines of commands to sort it, but getting there is a whole different
ballgame....
6to4 works "out of the box" but obviously routes via MS - which is no
bad thing for people with no assignment who want to tinker with ipv6.
If anyone knows how to do it then PLEASE tell us :-)
--
John Naismith
Julie Brandon
> Don't take this personally it's just (as others can attest) getting
Was it you who was giving up smoking perchance?
Ta-ra,
Julie (who's got a bit lazy with the diet... although thats partly
cos the bloody oven has broke)
John Naismith
<nos...@honeypot.computergeeks.me.uk.invalid> wrote:
>On Mon, 22 Mar 2004 13:53:28 +0000, John Naismith (john$E200...@naismith.org.uk) said:
>> Don't take this personally it's just (as others can attest) getting
>
>Was it you who was giving up smoking perchance?
It was and I have :-) 25 days and counting now.......
--
John Naismith
Julian Coleman
> under Windows. v6v4 _is_ a tunnel. In fact it's a tunnel with its' own
> unique interface (and hence routing) on the Windows box. 6to4 is a
> pseudo-interface which is preset to use MS to route ipv6 traffic.
> 6to4 works "out of the box" but obviously routes via MS - which is no
> bad thing for people with no assignment who want to tinker with ipv6.
Looking at:
http://www.ipv6style.jp/en/tryout/20030205/index.shtml
it looks to me as if you want a 6to4 configuration something like:
netsh interface ipv6 6to4 set relay 217.169.20.1
ipv6 adu 3/2002:5102:60a3::5102:60a3
ipv6 rtu 2002::/16 3 pub life 1800
ipv6 rtu ::/0 3/2002:d9a9:1401::1 pub life 1800
assuming that your endpoint is 81.2.96.163. Of course, I could be way
off base here (as I don't use windows at all).
Jim Crowther
<j...@coris.org.uk> writes:
> http://www.ipv6style.jp/en/tryout/20030205/index.shtml
Oo another good site I hadn't spotted (new at this), thanks J.
Ron Lowe
Naa, I dont think so.
[ Snippy extract from Understanding IPv6, Joseph Davies ]
RFC 2893 defined 2 types of tunnels:
-Configured
-Automatic
A configured tunnel requires manual configuration of the tunnel endpoints,
the IPv4 addresses of tunnel endpoints are not encoded in the IPv6 source or
destination addresses, nor inthe next-hop address of the matching route.
Typically, router-router and host-router tunneling configurations configured
manually. The tunnel interface configuration, consisting of the IPv4
addresses of the tunnel endpoints, must be specified manually along with the
static routes that use the tunnel interface.
To manually create configured tunnels in XP and 2k3, use the
"netsh interface ipv6 add v6v4tunnel"
command.
This is something totally diferent to 6to4, which is described in RFC 3056.
This is an address assignment and router-to-router automatic tunneling
technology, which uses global addresses like 2002:WWXX:YYZZ::/48 where
WWXXYYZZ is the hex representation of the IPv4 address w.x.y.z
Fine if you want a 2002: address.
To route _your_ assigned IPv6 allocation, you must use the manually
configured tunnel as described above.
Nothing to do with 6to4.
--
Ron
Jim Crowther
<ron-msng@{d.e.l.e.t.e.}lowe-family.me.uk> writes:
>To manually create configured tunnels in XP and 2k3, use the
>"netsh interface ipv6 add v6v4tunnel"
>command.
>
>This is something totally diferent to 6to4, which is described in RFC 3056.
>This is an address assignment and router-to-router automatic tunneling
>technology, which uses global addresses like 2002:WWXX:YYZZ::/48 where
>WWXXYYZZ is the hex representation of the IPv4 address w.x.y.z
>
>Fine if you want a 2002: address.
>
>To route _your_ assigned IPv6 allocation, you must use the manually
>configured tunnel as described above.
>Nothing to do with 6to4.
Aha. Learn more every day. So I'm using v6v4, not 6to4 then, I assume,
as other sites see my 2004: address. I hadn't appreciated the
difference before, ta.
John Naismith
<ron-msng@{d.e.l.e.t.e.}lowe-family.me.uk> wrote:
<snip>
>Nothing to do with 6to4.
Ta Ron - you explain it much better than I do ;-)
--
John Naismith
Julian Coleman
> configured tunnel as described above.
> Nothing to do with 6to4.
Um, I beg to differ. I looked at the IPv6 for Linux configuration on
the AAISP web page, then searched around with Google to translate the
IPV6_AUTOTUNNEL and sit0 parts of it. I found that this set up a *6to4*
configuration for a linux machine.
So, I built a kernel with stf (BSD equivalent), configured the files
/etc/ifconfig.stf0 and /etc/mygate6 and, now, I have working IPv6
connectivity for my netblock using 6to4. I send my encapsulated packets
to edgeless, and receive from endless. Works perfectly for me (as I
mentioned previously).
J
PS. I did also try setting up an RFC2893 tunnel but didn't get it to work.
Ron Lowe
>
> Um, I beg to differ.
I'm afraid I was unable to bridge the terminology gap between the
Linux example on the AAISP page and the terminology in Windows.
MS has made great efforts in its IPv6 documentation to use RFC terminology.
There are several RFCs describing several transition mechanisms.
I could not get an adequate response to the question:
"Exactly what tunneling mechanism(s) do AAISP support,
using _RCF references_ rather than a partially commented Linux conf file"
I tried to reverse-engineer the example conf files into RFC terminology, but
got nowhere.
So I tried diferent mechanisms untill I got it to work.
It's possible that you can build a 6to4 tunnel, but I couldnt.
I'll take your word for it that you can do this.
Untill the documentation on the IPv6 page is more formal,
using RFC descriptions of the supported tunneling mechanisms,
then we're shooting in the dark.
--
Ron
Jim Crowther
<ron-msng@{d.e.l.e.t.e.}lowe-family.me.uk> writes:
>It's possible that you can build a 6to4 tunnel, but I couldnt.
>I'll take your word for it that you can do this.
>
>Untill the documentation on the IPv6 page is more formal,
>using RFC descriptions of the supported tunneling mechanisms,
>then we're shooting in the dark.
That's true... Here's the nearest I could find to defining the commands
available:
<http://www.microsoft.com/windowsserver2003/technologies/ipv6/ipv62netshtable.mspx>
Ron Lowe
> >It's possible that you can build a 6to4 tunnel, but I couldnt.
> >I'll take your word for it that you can do this.
> >
> >Untill the documentation on the IPv6 page is more formal,
> >using RFC descriptions of the supported tunneling mechanisms,
> >then we're shooting in the dark.
>
> That's true... Here's the nearest I could find to defining the commands
> available:
>
>
<http://www.microsoft.com/windowsserver2003/technologies/ipv6/ipv62netshtabl
e.mspx>
>
yes, thanks.
I'm pretty conversant with the windows implimentation of IPv6,
it's tunneling mechanisms, and the netsh ipv6 contexts.
I just don't fully understand the linux conf files on the AAISP
page to be able to translate :-)
Essentially, the conf files don't explicitly state the type of tunnel being
created and how its configured to a level of detail I can understand and
reverse-engineer to RFC-talk, and then forward-engineer to windows-talk.
I gave up on this approach, and resorted to FWITIW.
(F**k with it till it works )
Essentially, the example on the AAISP page represents
"we FWITIW, we're not real sure how it works, but it does,
and here's the settings we used. "
--
Ron
Jim Crowther
<ron-msng@{d.e.l.e.t.e.}lowe-family.me.uk> writes:
[]
>yes, thanks.
>I'm pretty conversant with the windows implimentation of IPv6,
>it's tunneling mechanisms, and the netsh ipv6 contexts.
So I gathered. :)
>I just don't fully understand the linux conf files on the AAISP
>page to be able to translate :-)
>
>Essentially, the conf files don't explicitly state the type of tunnel being
>created and how its configured to a level of detail I can understand and
>reverse-engineer to RFC-talk, and then forward-engineer to windows-talk.
>
>I gave up on this approach, and resorted to FWITIW.
>(F**k with it till it works )
The time, oh for the time...
>Essentially, the example on the AAISP page represents
>"we FWITIW, we're not real sure how it works, but it does,
>and here's the settings we used. "
>
If you could rustle up a quick run-down of how you got it to work as
we'd all like it to work, I'm sure many would be very grateful. Yeah, I
know, there's that time thing, t'was just a thought.
Ron Lowe
> we'd all like it to work, I'm sure many would be very grateful. Yeah, I
> know, there's that time thing, t'was just a thought.
>
Well, since you ask, here's what I did to use my Win2k3 server
to route my IPv6 allocation on my LAN...
For posterity, here's a commented dump of what I did.
Sorry it's a bit long, but it may help someone.
First,Win2k3 server, tunnel endpoint and router:
Microsoft Windows [Version 5.2.3790]
(C) Copyright 1985-2003 Microsoft Corp.
# First, let's get to the netsh interface ipv6 context...
C:\Documents and Settings\Administrator.HOMENET>netsh
netsh>interface
netsh interface>ipv6
#Now, build the tunnel. I'm giving the tunnel interface the name AAISP.
#Replace 81.187.191.78 with your own tunnel endpoint address.
#Be sure this agrees with what you enter on the AAISP control page.
netsh interface ipv6>add v6v4tunnel "AAISP" 81.187.191.78 217.169.20.6
Ok.
#This will create Interface with an index >4 ( 5 in my case ).
#It will have auto-configured a link-local (fe80:: ) address.
#Let's have a look.
netsh interface ipv6>show address
Querying active state...
Interface 5: AAISP
Addr Type DAD State Valid Life Pref. Life Address
--------- ---------- ------------ ------------ ----------------------------
Link Preferred infinite infinite fe80::5:51bb:bf4e
Interface 4: Local Area Connection
Addr Type DAD State Valid Life Pref. Life Address
--------- ---------- ------------ ------------ ----------------------------
Link Preferred infinite infinite fe80::2a0:c9ff:fe1c:7391
Interface 3: 6to4 Tunneling Pseudo-Interface
Addr Type DAD State Valid Life Pref. Life Address
--------- ---------- ------------ ------------ ----------------------------
Other Preferred infinite infinite 2002:51bb:bf4e::51bb:bf4e
Interface 2: Automatic Tunneling Pseudo-Interface
Addr Type DAD State Valid Life Pref. Life Address
--------- ---------- ------------ ------------ ----------------------------
Link Preferred infinite infinite fe80::5efe:81.187.191.78
Interface 1: Loopback Pseudo-Interface
Addr Type DAD State Valid Life Pref. Life Address
--------- ---------- ------------ ------------ ----------------------------
Loopback Preferred infinite infinite ::1
Link Preferred infinite infinite fe80::1
#Add an address to the interface.
#This is where I'm unhappy.
#What V6 address rightly belongs to the tunnel?
#Here, Im using my assigned prefix (2001:8b0:b7:1)
# and assigning host address 1.
# But is that right?
#Surely a router needs to route between 2 subnets.
#And here, we're having an address on the WAN interface
#which is within the same subnet as the LAN interface is going to have.
#Whatever. It works.
#Remember to use your own assigned prefix!
netsh interface ipv6>add address "AAISP" 2001:8b0:b7:1::1
Ok.
#Let's see if that worked:
netsh interface ipv6>show address
Querying active state...
Interface 5: AAISP
Addr Type DAD State Valid Life Pref. Life Address
--------- ---------- ------------ ------------ ----------------------------
-
Manual Preferred infinite infinite 2001:8b0:b7:1::1
Link Preferred infinite infinite fe80::5:51bb:bf4e
<snip rest of output>
#OK, that worked.
#Now, set the default route to point to the tunnel.
#Again, I'm a bit queasy about this.
#I expect to set routes to point to next-hop routers,
#not to local interfaces!
#To do that, I'd need the v6 address of the router at the
#other end of the tunnel.
#But it works, so what the heck...
netsh interface ipv6>add route ::/0 "AAISP"
Ok.
#Let's see if it worked:
netsh interface ipv6>show route
Querying active state...
Publish Type Met Prefix Idx Gateway/Interface
Name
------- -------- ---- ------------------------ --- --------------------
no Manual 1 ::/0 5 AAISP
#OK, so far so good.
#Now to enable routing on this machine.
#We want to set up forwarding between the LAN and the tunnel.
#We also want to enable Router Advertisments on the LAN,
#so that clients can auto-configure addresses based on the advertised
prefix.
#Interface 4 is the LAN connection...
netsh interface ipv6>set interface 4 forwarding=enabled advertise=enabled
Ok.
#Now, the WAN interface ( the tunnel ) is interface 5...
netsh interface ipv6>set interface 5 forwarding=enabled
Ok.
#Now, we add the prefix route to the LAN and ensure it is
#published to the clients via the Router Advertisements.
#Format is 'add route prefix/length inteface# publish=yes
#Interface 4 is the LAN, remember.
#Use your own allocated address prefix!
netsh interface ipv6>add route 2001:08b0:b7:1::/64 4 publish=yes
Ok.
#Let's see what that did to the v6 route table..
netsh interface ipv6>show route
Querying active state...
Publish Type Met Prefix Idx Gateway/Interface
Name
------- -------- ---- ------------------------ --- --------------------
yes Manual 0 2001:8b0:b7:1::/64 4 Local AreaConnection
no Manual 1 ::/0 5 AAISP
#Oh, we forgot to publish the default route. (::/0).
#We need to do that, so the router advertises itself as a default route.
#I should have done that back at the start where I did the 'route add'.
# I forgot the 'publish=yes'.
#No matter, let's add it now:
netsh interface ipv6>set route ::/0 "AAISP" publish=yes
Ok.
netsh interface ipv6>show route
Querying active state...
Publish Type Met Prefix Idx Gateway/Interface
Name
------- -------- ---- ------------------------ --- --------------------
-
yes Manual 0 2001:8b0:b7:1::/64 4 Local Area
Connection
yes Manual 1 ::/0 5 AAISP
#Now, we're done on the router.
#It's all set up.
Now, I go to an XP client and crank it up.
The IPv6 stack should do Router Solicitation, and get
Router Advertisements in return.
That should auto-configure an address with the correct prefix,
and also a default route.
Microsoft Windows XP [Version 5.1.2600]
(C) Copyright 1985-2001 Microsoft Corp.
C:\Documents and Settings\Ron.HOMENET>ipconfig /all
Windows IP Configuration
Host Name . . . . . . . . . . . . : rons-pc
Primary Dns Suffix . . . . . . . : homenet.local
Node Type . . . . . . . . . . . . : Hybrid
IP Routing Enabled. . . . . . . . : No
WINS Proxy Enabled. . . . . . . . : No
DNS Suffix Search List. . . . . . : homenet.local
Ethernet adapter Local Area Connection:
Connection-specific DNS Suffix . : homenet.local.
Description . . . . . . . . . . . : Intel(R) PRO/100 S Desktop
Adapter
Physical Address. . . . . . . . . : 00-02-B3-2A-70-C0
Dhcp Enabled. . . . . . . . . . . : Yes
Autoconfiguration Enabled . . . . : Yes
IP Address. . . . . . . . . . . . : 81.187.191.66
Subnet Mask . . . . . . . . . . . : 255.255.255.240
IP Address. . . . . . . . . . . . : 2001:8b0:b7:1:94ba:f53:ed11:95ba
#The next line is the XP's 'aggregateable global unicast' V6 address.
#Prefix ( 2001:8b0:b7:1:: )came from the router advertisement.
#Host portion 202:b3ff:fe2a:70c0 is auto-generated from the
#MAC address of the LAN card, with fffe shoved in the middle,
#and the 'locally administered' bit toggled. This kludge is called EUI-64.
# Compare to physical address listed above.
IP Address. . . . . . . . . . . . : 2001:8b0:b7:1:202:b3ff:fe2a:70c0
IP Address. . . . . . . . . . . . : fe80::202:b3ff:fe2a:70c0%4
#The Default GW address is the Link-Local address of the router box.
#( Also EUI-64 generated. )
Default Gateway . . . . . . . . . : 81.187.191.65
fe80::2a0:c9ff:fe1c:7391%4
DHCP Server . . . . . . . . . . . : 81.187.191.78
DNS Servers . . . . . . . . . . . : 81.187.191.78
fec0:0:0:ffff::1%1
fec0:0:0:ffff::2%1
fec0:0:0:ffff::3%1
Primary WINS Server . . . . . . . : 81.187.191.78
Lease Obtained. . . . . . . . . . : 07 October 2003 18:12:40
Lease Expires . . . . . . . . . . : 15 October 2003 18:12:40
<snip tunneling interfaces>
#So let's try pinging the router:
Pinging homenetdc01.homenet.local [2001:8b0:b7:1:2a0:c9ff:fe1c:7391]
with 32 bytes of data:
Reply from 2001:8b0:b7:1:2a0:c9ff:fe1c:7391: time<1ms
Reply from 2001:8b0:b7:1:2a0:c9ff:fe1c:7391: time<1ms
<snip>
#This works because XP and 2k3 both register AAAA records in local DNS.
#So let's try a traceroute from the client XP to www.kame.net:
C:\Documents and Settings\Ron.HOMENET>tracert www.kame.net
Tracing route to orange.kame.net [2001:200:0:8002:203:47ff:fea5:3085]
over a maximum of 30 hops:
1 <1 ms * <1 ms 2001:8b0:b7:1:2a0:c9ff:fe1c:7391
2 26 ms 27 ms 27 ms v6-tunnel-aaisp.ipv6.btexact.com
[2001:7f8:2:801d::3]
3 27 ms 27 ms 27 ms needless.aaisp.net.uk [2001:8b0::d9a9:1408]
4 28 ms 27 ms 27 ms 2001:7f8:5::6334:1
5 27 ms 28 ms 28 ms 2001:618:c00:4:10:5cab:6a47:5555
6 29 ms 29 ms 28 ms ft-euro6ix-uk6x.ipv6.btexact.com
[2001:7f8:2:1::8]
7 38 ms 38 ms 37 ms p5-0-0.pas6ar1.paris.ipv6.opentransit.net
[2001:688:0:2:1::]
8 115 ms 116 ms 114 ms p6-0-0.nyk6ar1.new-york.ipv6.opentransit.net
[2001:688:0:2:1::7]
9 311 ms 311 ms 310 ms 2001:688:0:2:6::1
10 311 ms 312 ms 310 ms hitachi1.otemachi.wide.ad.jp
[2001:200:0:1800::9c4:2]
11 312 ms 311 ms 311 ms pc3.yagami.wide.ad.jp
[2001:200:0:1c04::1000:2000]
12 314 ms 313 ms 313 ms gr2000.k2c.wide.ad.jp
[2001:200:0:8002::2000:1]
13 312 ms 311 ms 313 ms orange.kame.net
[2001:200:0:8002:203:47ff:fea5:3085]
Trace complete.
#finally, let's crank up IE, and visit www.kame.net.
# If you have been there before using IPv4, then delete your
#temporary Internet Files to force a re-fetch.
#And there you go: a Dancing Kame.
#Scroll to the bottom of the page to see what IP address it detected you
using.
That's it.
--
Ron
Jim Crowther
<ron-msng@{d.e.l.e.t.e.}lowe-family.me.uk> writes:
>> If you could rustle up a quick run-down of how you got it to work as
>> we'd all like it to work, I'm sure many would be very grateful. Yeah, I
>> know, there's that time thing, t'was just a thought.
>>
>
>
>
>Well, since you ask, here's what I did to use my Win2k3 server
>to route my IPv6 allocation on my LAN...
>
>For posterity, here's a commented dump of what I did.
>Sorry it's a bit long, but it may help someone.
[]
You are a very kind man. I reckon many of the half-clueful (like me)
will be more than grateful for gaining a bit more of the elusive
clue-bits.
Cheers. :))
--
Jim Crowther, always learning, or at least trying to.
Ron Lowe
> will be more than grateful for gaining a bit more of the elusive
> clue-bits.
>
> Cheers. :))
With IPv6, we're all beginners.
I'm hoping that if we're beginners now,
we early adopters will be old hands when it goes mainstream!
I can't claim total expertise myself.
I called upon expert help.
I was able to get some assistance from Joseph Davies,
the IPv6 guru in MS himself. He writes the MS IPv6 books.
( But I'm not on such terms with him that I can bug him with
queries all the time! He just happened to answer me on a newsgroup! )
--
Ron
Jim Crowther
<ron-msng@{d.e.l.e.t.e.}lowe-family.me.uk> writes:
Heh, as you did me.
With a bit of luck we might get some generic WXP/W2K3 info available for
Adrian to put on the site. Me, I'm just about to start 3 long days
(finish at 22:30) so this will have to ferment...
Julie Brandon
> With a bit of luck we might get some generic WXP/W2K3 info available for
> Adrian to put on the site. Me, I'm just about to start 3 long days
> (finish at 22:30) so this will have to ferment...
That would be very very handy if that could appear on the A&A site!
Julian Coleman
> Linux example on the AAISP page and the terminology in Windows.
> It's possible that you can build a 6to4 tunnel, but I couldnt.
> I'll take your word for it that you can do this.
> Untill the documentation on the IPv6 page is more formal,
> using RFC descriptions of the supported tunneling mechanisms,
> then we're shooting in the dark.
I agree that the IPv6 page isn't very good. It took me about an hour using
Google to find out that the linux configuration related to 6to4 (RFC 3056).
Previously, I had been trying to make a gif (RFC 2893) tunnel work. I gave
up after a few hours and assumed that it wasn't supported at the AAISP end.
So, you'll just have to take my word that 6to4 tunnels work and I'll take
your word that RFC3056 (v4v6 in MS speak?) tunnels work ;-)
It shouldn't be too difficult to come up with some additional text for the
IPv6 page. I think it could do with:
1 a short paragraph about the supported tunnel types (with references)
2 an example for MS Windows (more references needed?)
3 a BSD example (with references)
Anyone from A&A care to comment? I can look at doing some of this maybe
this weekend but a more official answer for 1 above would be useful.
J
Matthew Seaman
> > I'm afraid I was unable to bridge the terminology gap between the
> > Linux example on the AAISP page and the terminology in Windows.
>
> > It's possible that you can build a 6to4 tunnel, but I couldnt.
> > I'll take your word for it that you can do this.
>
> > Untill the documentation on the IPv6 page is more formal,
> > using RFC descriptions of the supported tunneling mechanisms,
> > then we're shooting in the dark.
>
> I agree that the IPv6 page isn't very good. It took me about an hour using
> Google to find out that the linux configuration related to 6to4 (RFC 3056).
> Previously, I had been trying to make a gif (RFC 2893) tunnel work. I gave
> up after a few hours and assumed that it wasn't supported at the AAISP end.
> So, you'll just have to take my word that 6to4 tunnels work and I'll take
> your word that RFC3056 (v4v6 in MS speak?) tunnels work ;-)
That would seem to be a difference between Net- and FreeBSD then.
I've got a gif(4) RFC 2893 setup, which I slapped together in about 2
minutes by following a post by Mark Knight:
although I had to add:
network_interfaces="lo0 de0 gif0"
It's worked pretty much perfectly ever since. It might be interesting
to experiment with a stf(4) tunnel at some point, but not today.
Cheers,
Matthew
--
Dr Matthew J Seaman MA, D.Phil. 26 The Paddocks
Savill Way
PGP: http://www.infracaninophile.co.uk/pgpkey Marlow
Tel: +44 1628 476614 Bucks., SL7 1TH UK
Jim Crowther
<ron-msng@{d.e.l.e.t.e.}lowe-family.me.uk> writes
>netsh interface ipv6>add address "AAISP" 2001:8b0:b7:1::1
>Ok.
This is where I get horlicksed...
netsh interface ipv6>add address "AAISP" 2001:8b0:86:1::1
The file name is too long.
Hmm..
However the other MS (use us, ignore the rest of IPv6) automatic methods
work. (6to4, ISATAP)
Here's a site that backs up your method:
<http://www.join.uni-muenster.de/Dokumente/Howtos/Howto_IPv6-in-IPv4-Tunnel.php?lang=en>
but it fails here. :(
I dread having to post to an MS group - does anyone know of one that has
a higher-than-normal sig-noise ratio? Most of the MS mailing lists only
know of Teredo. (6to4 AIUI, please correct me if I'm wrong.)
Andrew McDonald
> In message <4060b5e4$0$63625$5a6a...@news.aaisp.net.uk>, Ron Lowe
><ron-msng@{d.e.l.e.t.e.}lowe-family.me.uk> writes
>
>>netsh interface ipv6>add address "AAISP" 2001:8b0:b7:1::1
>>Ok.
>
> This is where I get horlicksed...
>
> netsh interface ipv6>add address "AAISP" 2001:8b0:86:1::1
> The file name is too long.
Are you both running the same version of Windows and the same set of
Service Packs? I understand (not really being much of a Windows user)
that these things have changed dramatically between Service Packs.
> Most of the MS mailing lists only
> know of Teredo. (6to4 AIUI, please correct me if I'm wrong.)
Nope, Teredo (aka Shipworm) is yet another IPv4/IPv6 transition
mechanism. (Again using tunnels, but is designed to be friendlier
with NATs. IIRC it uses a central server and then creates
peer-to-peer connections where possible, but I could be wrong.)
--
Andrew McDonald
Ron Lowe
>
> netsh interface ipv6>add address "AAISP" 2001:8b0:86:1::1
> The file name is too long.
I've never seen that error message, so I dont know why you got it.
Possibly your IPv6 installation is corrupt in some way?
Is this an XP or 2k3 box?
If it's XP, ensure SP1 is present, also ensure the
Advanced Networking Pack is installed too.
This includes an updated IPv6 stack.
http://support.microsoft.com/?kbid=817778
If these were all present and correct, remove and reinstall IPv6.
If that doesn't fix it, I can query elsewhere and see if I get any
suggestions.
> I dread having to post to an MS group - does anyone know of one that has
> a higher-than-normal sig-noise ratio?
Why, any that I inhabit ;-)
>Most of the MS mailing lists only
> know of Teredo. (6to4 AIUI, please correct me if I'm wrong.)
Teredo is not 6to4.
Here's the skinny on Teredo:
http://www.microsoft.com/technet/prodtechnol/winxppro/maintain/teredo.mspx
To use teredo to access the V6 internet, you need to access
both a teredo server ( which is a meeting point where clients and
servers can set up the initial communication), _and_ a teredo relay to
actually forward packets between the IPv4 client and the IPv6 Internet.
A couple of quotes:
...
The general role of the Teredo server is to assist in the address
configuration of Teredo client and to facilitate the initial communication
between Teredo clients and other Teredo clients or between Teredo clients
and IPv6-only hosts.
...
To facilitate communication between computers using the Advanced Networking
Pack for Windows XP, Microsoft has deployed Teredo servers on the IPv4
Internet.
...
Microsoft does not plan to deploy any Teredo relays on the IPv4 Internet.
Individual Internet service providers (ISPs) could deploy their own Teredo
relays. The Advanced Networking Pack for Windows XP implementation of the
Teredo client will work with a Teredo relay when sending traffic to an
IPv6-only host on the IPv6 Internet.
...
It is important to note that Teredo is designed as a last resort transition
technology for IPv6 connectivity. If native IPv6, 6to4, or ISATAP
connectivity is present, the host does not act as a Teredo client. As more
IPv4 NATs are upgraded to support 6to4 and IPv6 connectivity become
ubiquitous, Teredo will be used less and less until finally it is not used
at all.
...
Personally, I think it all looks a bit flakey.
I didn't experiment much with it.
--
Ron
Jim Crowther
news:40634450$0$63622$5a6a...@news.aaisp.net.uk:
>> This is where I get horlicksed...
>>
>> netsh interface ipv6>add address "AAISP" 2001:8b0:86:1::1
>> The file name is too long.
>
> I've never seen that error message, so I dont know why you got it.
> Possibly your IPv6 installation is corrupt in some way?
Quite possibly.
> Is this an XP or 2k3 box?
XP, I'm away from it at the moment.
> If it's XP, ensure SP1 is present, also ensure the
> Advanced Networking Pack is installed too.
> This includes an updated IPv6 stack.
I'm pretty sure it has SP1 installed, I'll check when I can. It definitely
has the Advanced Networking Pack.
> http://support.microsoft.com/?kbid=817778
>
> If these were all present and correct, remove and reinstall IPv6.
> If that doesn't fix it, I can query elsewhere and see if I get any
> suggestions.
Again, I'll do that when I can, should be later tonight or in the wee
hours.
>> I dread having to post to an MS group - does anyone know of one that
>> has a higher-than-normal sig-noise ratio?
>
> Why, any that I inhabit ;-)
Heh, that gives me a fighting chance then.
[]
Thanks for the extra helpful info. Takes time for an old dog like me to
get my head round all the new concepts and the abbreviations that describe
them. Sometimes I don't even know there's a question, let alone any
answers. :)
--
Jim Crowther "It's MY computer" (tm SMG)
Avoid more malware by ignoring previous Usenet addresses,
Ron Lowe
>>http://www.microsoft.com/technet/prodtechnol/winxppro/maintain/teredo.mspx
> Thanks for the extra helpful info. Takes time for an old dog like me to
> get my head round all the new concepts and the abbreviations that describe
> them. Sometimes I don't even know there's a question, let alone any
> answers. :)
Conclusion I came to after reading through that:
I'd not spend too much time bending your mind round Teredo.
It's a complicated and flakey system of last resort, to allow
tunneled IPv6 connectivity to users behind NAT.
More trouble than it's worth, IMHO!
--
Ron
Jim Crowther
That's what I thought. I'll stick with getting v6v4 working. As it turns
out my XPpro *wasn't* SP1 (doh!), so spent last night getting that
installed. Looks promising, I got much further with the netsh set-up, but
didn't get to test it.
Mark Knight
Matthew Seaman <m.se...@infracaninophile.co.uk> writes
>That would seem to be a difference between Net- and FreeBSD then.
>I've got a gif(4) RFC 2893 setup, which I slapped together in about 2
>minutes by following a post by Mark Knight:
Glad it was useful ;) I can certainly confirm that rfc2893 style
configured tunnels work fine with FreeBSD's gif interface. I've not had
a day of trouble since I made my original posting.
>
>http://groups.google.co.uk/groups?hl=en&lr=&ie=UTF-8&safe=off&selm=u0ieV
>oi6Uyd%24EwLQ%40lap.knigma.org
>
>although I had to add:
>
> network_interfaces="lo0 de0 gif0"
Were you already overriding network_interfaces="auto" from
/etc/defaults/rc.conf?
Cheers,
--
Mark A. R. Knight finger: sp...@knigma.org
Tel: +44 7973 410732 http://www.knigma.org/
s/spam/markk/g
Matthew Seaman
> In message <863c7yv...@happy-idiot-talk.infracaninophile.co.uk>,
> Matthew Seaman <m.se...@infracaninophile.co.uk> writes
> > http://groups.google.co.uk/groups?hl=en&lr=&ie=UTF-8&safe=off&selm=u0ieV
> >oi6Uyd%24EwLQ%40lap.knigma.org
> >
> >although I had to add:
> >
> > network_interfaces="lo0 de0 gif0"
>
> Were you already overriding network_interfaces="auto" from
> /etc/defaults/rc.conf?
Errr... I don't think so, but it's been a while since I set the IPv6
stuff up, and I can't actually remember. I didn't have any reason not
to use the default setting -- network_interfaces="auto" -- before that
so probably not.