I'm trying to get NAT-PT to work on a host with FreeBSD 4.2 (Pentium II, 200
MHz, KAME-snapshot 2001-01-08). I've compiled the kernel with the option
NAT_PT in the Configuration File (from GENERIC.KAME, only the NAT_PT option
changed), and I've compiled the nat-pt userland files successfully. A f
>I'm trying to get NAT-PT to work on a host with FreeBSD 4.2 (Pentium II, 200
>MHz, KAME-snapshot 2001-01-08). I've compiled the kernel with the option
>NAT_PT in the Configuration File (from GENERIC.KAME, only the NAT_PT option
>changed), and I've compiled the nat-pt userland files successfully.
Memory statistics by type Type Kern
Type InUse MemUse HighUse Limit Requests Limit Limit Size(s)
routetbl728180102400K 102401K102400K 73822480 0 16,32,64,128,256
100MB in use by the routing table? There are only 6 routes... :P
--
Richard A Stee
Richard,
what version of the OS are you running ?
Also, could you do a 'netstat -p tcp' and send me the output.
jayanth
Richard A. Steenbergen ([EMAIL PROTECTED]) wrote:
> Memory statistics by type Type Kern
> Type InUse MemUse HighUse Limit Requests Limit Li
On Mon, 15 Jan 2001, jayanth wrote:
> Richard,
>
> what version of the OS are you running ?
> Also, could you do a 'netstat -p tcp' and send me the output.
4.2-BETA presently, I'm upgrading to -stable to see if it'll go away.
tcp:
1561757204 packets sent
1036973008 data
Hi together,
for quite a while I have been looking around for a way to limit the bandwith
for each IP that accesses my server. I want to slow down any connektion
to 128 KBit/s.
The only thing I found was Dummynet in combination with ipfw. I am using
ipf as firewall an for IP-accounting. It does a
There is an annoying bug in FreeBSD networking/routing which has been
around at least since 3.4.
Symptoms are -- if you change the address to an interface,
packets to destinations to which you have talked to in the past
will still go out with the previous address unless
you delete and reinstall
check the arp cache. You might want to do a arp -ad and try again.
jayanth
Luigi Rizzo ([EMAIL PROTECTED]) wrote:
> There is an annoying bug in FreeBSD networking/routing which has been
> around at least since 3.4.
>
> Symptoms are -- if you change the address to an interface,
> packets to des
>Example:
>ifconfig ed0 10.0.0.1
>ping 10.0.0.20 # works fine
>ifconfig ed0 10.0.0.2
>ping 10.0.0.20 # no reply, tcpdump shows traffic coming from 10.0.0.1
>route delete 10.0.0.2
>ping 10.0.0.20 # now things work as expected
>So it seems that the old address is stored
> >Any idea on where the old address is stored ?
>
> try using
> # route -n get 10.0.0.0
> and you'll see rt_ifa holding pointer to 10.0.0.1. rt_ifa is used for
> source address selection.
thanks, that was it (with a -v flag to see all...)
> IMHO IPv4 code is not
>> IMHO IPv4 code is not very friendly with multiple addresses on single
>> interface. i believe the following items are assumed for the use
>> of rt_ifa.
>but it seems that when you change interface address the call to
>in_ifscrub() should take care of removing the old address...
Hola!
I'm experimenting with ipv6 and i need some help. I've a gif
tunnel between a FBSD 4.0 and a FBSD 4.2.
In the 4.2 side i get lots of messages like:
Jan 16 02:34:05 ipv6-gw /kernel: nd6_lookup: failed to add route for a
neighbor(3ffe:2900:b00d:ff00::0002), errno=17
Jan 16 02:34:05 ipv6-gw
On Tue, Jan 16, 2001 at 02:43:19AM -0300, [EMAIL PROTECTED] wrote:
> Hola!
>
> I'm experimenting with ipv6 and i need some help. I've a gif
> tunnel between a FBSD 4.0 and a FBSD 4.2.
>
> In the 4.2 side i get lots of messages like:
>
> Jan 16 02:34:05 ipv6-gw /kernel: nd6_lookup: failed to add
¡Hola!
> > I'm experimenting with ipv6 and i need some help. I've a gif
> > tunnel between a FBSD 4.0 and a FBSD 4.2.
> >
> > In the 4.2 side i get lots of messages like:
> >
> > Jan 16 02:34:05 ipv6-gw /kernel: nd6_lookup: failed to add route for a
>neighbor(3ffe:2900:b00d:ff00::0002), errno=
LVS problem
15 matches
Mail list logo
