Hello ramon,

due to spoof prevention... have a look in you boot scripts for rp_filter

and change...

for f in /proc/sys/net/ipv4/conf/*/rp_filter; do
    echo 1 > $f
done

to

for f in /proc/sys/net/ipv4/conf/*/rp_filter; do
    echo 0 > $f
done

-- 
Best regards,
 evilbunny                            mailto:[EMAIL PROTECTED]

Tuesday, April 23, 2002, 11:47:36 AM, you wrote:

rb> Hi peoples,

rb> I've got a tricky one which has got me a little stumped..

rb> I'm installing satellite for my net connection. Ive got it all working
rb> The ISP is a reseller of the iHUG service.

rb> The Dial up account is with a local ISP. (not an iHug specific account)
rb> Ive got the card all configured (ie the driver loads etc)
rb> and the Dialup appears to allow me to change my source IP outgoing.

rb> How does it the satellite stuff work:
rb>         data goes out via the ppp0:0 (dialup) interface
rb>         with a source address of the Satellite card

rb>         The satellite card is configured with a dummy IP address
rb>         and it receives packets (from my ISP) sent to the external IP Address.
rb>         (there is an app running in bg that 'locks' on to the satellite and get the
rb> data)

rb> How is it setup:?
rb>         (say the satellite External IP address is 1.2.3.4 ie real.
rb>         the card drivers are loaded
rb>         The module is loaded into the kernel and then the app is loaded.
rb>         The the 'NIC' side of the card is configured.
rb>         # ifconfig sm200d 10.0.0.1

rb>         Dialup the ISP via modem (local IP given 5.5.5.1, remote ip given
rb> 5.5.5.254)
rb>         # ifconfig ppp0:0 1.2.3.4 netmask 255.255.255.255 pointopoint 5.5.5.254

rb>         delet the old ppp0 route
rb>         # route del default

rb>         add new route (via ppp0:0 so it sets the source of the IP to the sat IP)
rb>         # route add default gw 5.5.5.254 dev ppp0:0

rb> What happens ?
rb>         I can ping an external host
rb>         # ping www.ibm.com

rb>         The return ping response actually DO come back on the sm200d interface

rb>       sm200d    Link encap:Ethernet  HWaddr 00:xx:xx:xx:xx:xx
rb>           inet addr:10.0.0.1  Bcast:10.255.255.255  Mask:255.0.0.0
rb>           UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
rb>    ---->  RX packets:97 errors:0 dropped:0 overruns:0 frame:0
rb>           TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
rb>           collisions:0 txqueuelen:100
rb>           Interrupt:9

rb>         The box I am doing this on is a firewall (smoothwall) but I have dropped
rb> all the
rb>         ipchains ruls and set default policies for input,output as ACCEPT and
rb>         forward as MASQ (because there is an internal network attached to eth0 @
rb> 192.168.1.0

rb>         And I also set some ipchains logging with ipchains -A input -d 1.2.3.4 -l
rb>         and I can see all the reponses come back ..

rb> my question?
rb>         How do I get the card/linux to deliver them to the application (eg ping) or
rb> whatever
rb>         given that they come back on a different interface to what sent out on?

rb>         How can I debug this further to see where/why they are going no where?

rb>         How is this 'jump' between NIC's configured ?



rb> Any suggestions are MUCHly appreciated! :^)

rb> - ramon.






rb> This email was sent to you by Ramon Buckland, Software Engineer at f5.
rb> You can reach Ramon directly on 0421 379 694.

-- 
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug

Reply via email to