Ruby is the gateway machine. DNS is set up ok, and the addressing appears to be ok.
but they can't see each other
?
Everything appears to work fine on each machine - but they cant see each other.
>From nanguo:
$ ping6 ruby
ping6: UDP connect: No route to host
$
>From ruby:
$ ping6 nang
- Original Message -
From: "Lars Eggert" <[EMAIL PROTECTED]>
To: "Dennis Pedersen" <[EMAIL PROTECTED]>
Cc: <[EMAIL PROTECTED]>
Sent: Monday, April 08, 2002 11:23 PM
Subject: Re: IPsec tunnel mode
> Dennis Pedersen wrote:
> > Because on the [EMAIL PROTECTED] Lars Eggert said something ab
At 12:16 9-4-2002 +0200, Dennis Pedersen wrote:
>But uhm is there a 'simple' way of doing this? (as in just adding the IP of
>the other ends gif interface as destinatio in my routes?
>The setup today i an exact copy of (other IP's of course)
>www.freebsddiary.org/ipsec-tunnel.php
>This works just
Hello.
Since I keep running into problems with NAT traversal when implementing
IPSEC solutions, I've been looking at solutions offered by quite a few
of the commerical setups.
Seems that a pretty widely used solution is to send ESP encapsulated in
udp packets. Specifics for this kind of configura
From: "Rogier R. Mulhuijzen" <[EMAIL PROTECTED]>
> At 12:16 9-4-2002 +0200, Dennis Pedersen wrote:
> >But uhm is there a 'simple' way of doing this? (as in just adding the IP
of
> >the other ends gif interface as destinatio in my routes?
> >The setup today i an exact copy of (other IP's of cours
I have problem with my dumb terminal.
I'm new employee is my office, and I'm a newbie at
FreeBSD Unix.
At my office, there is one set of computer :
1 unit Digital Server with processors Alpha (at the
monitor and
Casing writed DIGITAL), this server using OS Digital
Unix Ver 3.2C
5 unit dumb termi
From: "Merlin" <[EMAIL PROTECTED]>
> If I understand it correctly, all I need to do on a network host is set
> ipv6_enable="YES"
>
> and the rest is done automagically.
Correct.
> net.inet6.forwarding=0
> net.inet6.accept_rtadv=1
> rtsol
>
> are all set automatically from the rc.network6 sta
I have problem with my dumb terminal.
I'm new employee is my office, and I'm a newbie at
FreeBSD Unix.
At my office, there is one set of computer :
1 unit Digital Server with processors Alpha (at the
monitor and
Casing writed DIGITAL), this server using OS Digital
Unix Ver 3.2C
5 unit dumb termi
Robert wrote:
> Ruby is the gateway machine. DNS is set up ok, and the addressing appears to be ok.
>but they can't see each other
> ?
> Everything appears to work fine on each machine - but they cant see each other.
>
> From nanguo:
> $ ping6 ruby
> ping6: UDP connect: No route to host
Dennis Pedersen wrote:
> But uhm is there a 'simple' way of doing this? (as in just adding the IP of
> the other ends gif interface as destinatio in my routes?
> The setup today i an exact copy of (other IP's of course)
> www.freebsddiary.org/ipsec-tunnel.php
> This works just fine besides til pro
Dennis Pedersen wrote:
> But uhm is there a 'simple' way of doing this?
Did you look at the KAME newsletters? (URL in a previous email)
Lars
--
Lars Eggert <[EMAIL PROTECTED]> Information Sciences Institute
http://www.isi.edu/larse/ University of Southern California
My packet traces show that FreeBSD 4.5 TCP connections conatin TCP timestamp
option in the TCP header. So, the TCP/IP header size is 52 bytes (instead of
40) most of the time.
Is there a reason for including the timestamp option on most of the TCP
packets?
___
- Original Message -
From: "Lars Eggert" <[EMAIL PROTECTED]>
To: "Dennis Pedersen" <[EMAIL PROTECTED]>
Cc: <[EMAIL PROTECTED]>
Sent: Tuesday, April 09, 2002 6:18 PM
Subject: Re: IPsec tunnel mode
> Dennis Pedersen wrote:
> > But uhm is there a 'simple' way of doing this? (as in just add
At 13:36 9-4-2002 +0200, Dennis Pedersen wrote:
>Uhm okai, but where do i see the port number for the 2 natd processes? , kan
>i specify it somewhere or?
From natd(8):
-port | -p port
Read from and write to divert(4) port port, distinguishing
packets as
Dennis Pedersen wrote:
>>Setting up the other approach (IPIP tunnel + IPsec transport mode) works
>>by first setting up the tunnels (see the gifconfig/ifconfig man pages)
>>and stringing the topology together with route (route man page). No
>>other commands are needed. Once this works (i.e. you se
hello,
I installed an ASUS A7N266 with an Athlon XP 1700+ as a
diskless machine booting pxeboot over a 3Com 3c905C-TX
connecting to a isc-dhcp3-3.0.1.r6 server installed
from ports.
With a 128M DIMM installed, everything works fine:
bootpc_init: wired to interface 'xl0'
Sending DHCP Discover
Few quick questions.
1. Has anyone been able to establish a successful VPN connection
between FreeBSD and a Cisco VPN server?
2. If not, is it possible?
3. Has anyone attempted to port the Linux Cisco VPN client to
FreeBSD?
I'm not on the list, so pl
Nelson, Trent . wrote:
> Few quick questions.
>
> 1. Has anyone been able to establish a successful VPN connection
> between FreeBSD and a Cisco VPN server?
What protocol? IPsec? PPTP?
PPTP with mpd has some problems, at least with the Cisco box we have.
The Cisco box likes to a
Hi,
Actually I rewrote loadd (a load balancing daemon)
for running with netgraph. In the original version, loadd use Divert sockets
with ipfw for taking incoming/outgoing traffic.
What node I must use for doing the same thing
?
I try to use ng_ether but if I do that, People that
run pppoe
there is a node called etf (ethertype filter)
that can allow different nodes to share an ethernet interface
The ppp daemon needs to be modified a bit to be able to use it..
Brian, what does it take to allow one to specify a different node and
hook?
I.e.
if I hook an ethertype filter onto the int
Ok julian,
So I must write loadd with ng_ether for the moment and wait for ng_etf
include in FreeBSD 4.x.
I can modify my code after if ng_etf is available...
Last question, when I write a packet on the lower hook, must I include a
CRC32 field at the end of the ethernet packet ?
I create a littl
(If this shouldn't be on -net please accept my apologies. It seemed all the
networking gurus are there and this sort of overlaps onto that subject.)
I have a 4.5 release box that is acting as a bridging firewall with ipfw
for an internet connected network and I'm having some issues with ftp
(as
Sebastien Petit writes:
> Actually I rewrote loadd (a load balancing daemon) for running with netgraph. In the
>original version, loadd use Divert sockets with ipfw for taking incoming/outgoing
>traffic.
> What node I must use for doing the same thing ?
> I try to use ng_ether but if I do that,
On Tue, 9 Apr 2002, Sebastien Petit wrote:
> Ok julian,
>
> So I must write loadd with ng_ether for the moment and wait for ng_etf
> include in FreeBSD 4.x.
> I can modify my code after if ng_etf is available...
> Last question, when I write a packet on the lower hook, must I include a
> CRC32
> Sebastien Petit writes:
> > Actually I rewrote loadd (a load balancing daemon) for running with
netgraph. In the original version, loadd use Divert sockets with ipfw for
taking incoming/outgoing traffic.
> > What node I must use for doing the same thing ?
> > I try to use ng_ether but if I do th
On Tue, 9 Apr 2002, Archie Cobbs wrote:
> Sebastien Petit writes:
> > Actually I rewrote loadd (a load balancing daemon) for running with netgraph. In
>the original version, loadd use Divert sockets with ipfw for taking incoming/outgoing
>traffic.
> > What node I must use for doing the same t
I've been wondering if I shouldn't incorporate the etf functionality
into the ether node..
On Tue, 9 Apr 2002, Sebastien Petit wrote:
> > Sebastien Petit writes:
> > > Actually I rewrote loadd (a load balancing daemon) for running with
> netgraph. In the original version, loadd use Divert socke
On Tuesday 09 April 2002 22:29, Julian Elischer wrote:
> On Tue, 9 Apr 2002, Sebastien Petit wrote:
> > Ok julian,
> >
> > So I must write loadd with ng_ether for the moment and wait for ng_etf
> > include in FreeBSD 4.x.
> > I can modify my code after if ng_etf is available...
> > Last question,
Sebastien Petit wrote:
> With tcpdump I see bad cksum 0!, so I think I must recompute ip checksum
> before resending the packet.
The xl interface offloads checksumming to the NIC now, which is why you
see zero checksums when you tcpdump on the source or sink hosts. (There
was a thread on this
On Wednesday 10 April 2002 00:42, Lars Eggert wrote:
> Sebastien Petit wrote:
> > With tcpdump I see bad cksum 0!, so I think I must recompute ip checksum
> > before resending the packet.
>
> The xl interface offloads checksumming to the NIC now, which is why you
> see zero checksums when you tcpd
On Wed, 10 Apr 2002, Sebastien Petit wrote:
> On Tuesday 09 April 2002 22:29, Julian Elischer wrote:
> > On Tue, 9 Apr 2002, Sebastien Petit wrote:
> > > Ok julian,
> > >
> > > So I must write loadd with ng_ether for the moment and wait for ng_etf
> > > include in FreeBSD 4.x.
> > > I can modif
Sebastien Petit wrote:
> And I apologize for my duplicate thread about this...
None needed - I wasn't trying to imply that the thread back then was
about the netgraph issue you described and you had missed it (it was
about the apparently incorrect checksums with tcpdump), and I'm sorry if
I ca
Several years ago I wrte the following macro
to update a checksum when changing a word
in an IP packet.
here's the code...
#define FIXSUM16(c, op, np) \
do { \
(c) -= (u_int16_t) ~*((u_int16_t *) (
> to update a checksum when changing a word
>
> #define FIXSUM16(c, op, np) \
> do { \
> (c) -= (u_int16_t) ~*((u_int16_t *) (op)); \
> if ((c) < 0) {
just looking at this (it's been a while)
On Tue, 9 Apr 2002, Bakul Shah wrote:
> > to update a checksum when changing a word
> >
> > #define FIXSUM16(c, op, np) \
> > do { \
> > (c) -= (u_in
I have the Border router set up and working it seems, and the client/host on the same
network and ping it - but it knows
itself as the interface only ? (fe80+MAC)
How do I convince the client that it is actually 2002:cb01:6006::2
I can't discover how to put a 2002 address onto the rl0 interface.
> > > (c) -= (u_int16_t) *((u_int16_t *) (np)); \
> ^^ hmm should that be "+=" ?
I believe your original code is correct. See the paragraph
right before section 6 on Page 4, rfc1624.
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-net" i
Hello,
has anyone successfully connected an ORiNOCO Gold or similar wi0 running
under FreeBSD 4.5-STABLE to a Lucent/Agere Access Server AS-2000? Do I
have to port the Lucent asclient for Linux to FreeBSD or will the usual
FreeBSD included tools suffice?
-Bjorn Fischer
To Unsubscribe: send mai
On Wednesday 10 April 2002 01:01, you wrote:
> Sebastien Petit wrote:
> > And I apologize for my duplicate thread about this...
>
> None needed - I wasn't trying to imply that the thread back then was
> about the netgraph issue you described and you had missed it (it was
> about the apparently in
39 matches
Mail list logo
