On Wednesday 03 April 2002 03:09, Crist J. Clark wrote:
> On Tue, Apr 02, 2002 at 06:52:26PM -0500, Scott Ullrich wrote:
> > Correct. The master and backup settings and/will override the RFC. Can
> > anyone suggest a few ways that this could all be improved at the kernel
> > level?
>
> I think i
Hi, all.
I'm testing to use FreeBSD box as PPPoE server and found that ppp(8)
uses 255.255.255.254 (special address to get IP address from NAS pool
defined in RADIUS protocol) as p2p address when "set radius" is used
in ppp.conf. I found the same trouble at
http://docs.freebsd.org/cgi/getmsg.cgi
Tatsumi Hosokawa writes:
> Following patch can fix this problem. Please review it.
>
> - if (bundle->radius.valid && bundle->radius.ip.s_addr != INADDR_NONE) {
> + if (bundle->radius.valid && bundle->radius.ip.s_addr != INADDR_NONE &&
> + bundle->radius.ip.s_addr != RADIUS_INADDR_POOL) {
>
I'm slogging through the KAME IPsec code looking at adding support for
crypto hardware (and NICs that do onboard IPSEC processing). The OpenBSD
IPsec implementation already has this and doing something similar to what
OpenBSD has done requires restructuring large parts of the KAME code in a
simil
have you asked the KAME people if they have plans to
do such suppport themselves?
On Wed, 3 Apr 2002, Sam Leffler wrote:
> I'm slogging through the KAME IPsec code looking at adding support for
> crypto hardware (and NICs that do onboard IPSEC processing). The OpenBSD
> IPsec implementation a
On Wed, 3 Apr 2002, Sebastien Petit wrote:
>
> freevrrpd actually use RFC MAC addresses (00:00:5E:00:01:VRID) as ethernet
> source address when it send to the multicast address (as described in the
> RFC). Actually, FreeBSD doesn't support multiple ethernet address on one
> physical interface
Kris Kirby wrote:
> Let say I have a machine I want to attach to internet subnet
> 216.6.6.129/25. But the machine is at my house, NAT'd from the world. So
> to network the machine, I'd have to "bridge" across something like a VLAN
> over an IPSEC tunnel. Is this right? Can it be done that way? Is
Yes and no. I was told they wanted to add hardware support but I've been
unable to reach the "right people" to start a dialogue, which is why I sent
my note.
Sam
- Original Message -
From: "Julian Elischer" <[EMAIL PROTECTED]>
To: "Sam Leffler" <[EMAIL PROTECTED]>
Cc: <[EMAIL PROTEC
Hi
I have reached the 655 firewalling rules limit (with discrete values)
in ipfw and I was wondering why ipfw will not let the user select
the incremental step value in rules numbering ? also it should be
possible to renumber these rules on the fly
(though, i agree this is not this useful)
--
=
On Tue, 2 Apr 2002 12:00:30 -0500
Scott Ullrich <[EMAIL PROTECTED]> wrote:
> The HUT Project includes FreeVRRPD. Since Sebastien hasn't rung in here, I
> will try to clear the air.
>
> Sebastien and I are currently rewriting FreeVRRPD to take care of the
> remaining RFC issues and to cleanup
Sam Leffler wrote:
> Yes and no. I was told they wanted to add hardware support but I've been
> unable to reach the "right people" to start a dialogue, which is why I sent
> my note.
Try [EMAIL PROTECTED]; you'll have a response in a few hours (when
daylight hits Japan :-)
Lars
--
Lars Eggert
Hi
I have reached the 655 firewalling rules limit (with discrete values)
in ipfw and I was wondering why ipfw will not let the user select
the incremental step value in rules numbering ? also it should be
possible to renumber these rules on the fly
(though, i agree this is not this useful)
--
=
On Wed, Apr 03, 2002 at 08:59:23PM +0200, Christophe Prévotaux wrote:
> Hi
>
> I have reached the 655 firewalling rules limit (with discrete values)
> in ipfw and I was wondering why ipfw will not let the user select
> the incremental step value in rules numbering ? also it should be
> possible t
On Tue, 2 Apr 2002, Christophe [ISO-8859-1] Prévotaux wrote:
> Hi
>
> I have reached the 655 firewalling rules limit (with discrete values)
> in ipfw and I was wondering why ipfw will not let the user select
> the incremental step value in rules numbering ? also it should be
> possible to renu
And what would this code do on a Sparc? Don't assume little-endian
hardware. As this is hardly time-critical code, memcmp or equivalent
is fine.
On Wed, Apr 03, 2002 at 07:31:16PM +0900, Tatsumi Hosokawa wrote:
> Hi, all.
>
> I'm testing to use FreeBSD box as PPPoE server and found that ppp(8)
At Wed, 3 Apr 2002 15:02:38 -0500,
Barney Wolff wrote:
>
> And what would this code do on a Sparc? Don't assume little-endian
> hardware. As this is hardly time-critical code, memcmp or equivalent
> is fine.
Oops, sorry. I'll use htonl(0xfffe) instead. I don't think it
isn't time critica
Typo.
At Thu, 04 Apr 2002 08:51:23 +0900,
Tatsumi Hosokawa wrote:
>
> Oops, sorry. I'll use htonl(0xfffe) instead. I don't think it
> isn't time critical code because it executed only once per auth
~ is :-)
> session of ppp.
--
Tatsumi Hosokawa
<[EMAIL PROTECTED]>
http://FromTo.Cc/
At Wed, 03 Apr 2002 19:31:16 +0900,
Tatsumi Hosokawa wrote:
>
> I'm testing to use FreeBSD box as PPPoE server and found that ppp(8)
> uses 255.255.255.254 (special address to get IP address from NAS pool
> defined in RADIUS protocol) as p2p address when "set radius" is used
> in ppp.conf. I fou
On Wed, Apr 03, 2002 at 12:06:20PM +0200, Sebastien Petit wrote:
[snip]
> Design of freevrrpd cause a problem actually because when a MASTER server
> leave LAN (cable problem), SLAVE take his place and send gratuitous ARP for
> update ARP cache of all hosts on the same LAN.
That's not really a
- Original Message -
From: "Crist J. Clark" <[EMAIL PROTECTED]>
To: "Sebastien Petit" <[EMAIL PROTECTED]>
Cc: "Scott Ullrich" <[EMAIL PROTECTED]>; "'Barney Wolff'"
<[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>
Sent: Thursday, April 04, 2002 7:45 AM
Subject: Re: HUT Project
> On Wed, Apr 03,
On Thu, Apr 04, 2002 at 09:12:40AM +0200, Sebastien Petit wrote:
[snip]
> with the RFC2338, FreeBSD must respond to ARP query on 10.0.1.1 and
> 172.16.2.1 with 00:00:5E:01:01 MAC address and not with the real MAC
> addresses of physical interfaces. Then when a switching between SLAVE and
> MASTER
21 matches
Mail list logo
