Seemingly, all you'd need is a static route table entry on "Bill" to say that
the gateway for the 172.16.1.x network is 188.8.131.52. However, this business
with the "tunnel failing" when route table entries are changed is pretty much
unheard of. You don't show what the non CIPE (Ethernet, I guess) network
addresses are, but it's my guess that when you are changing the routing table
entries, you're somehow invalidating the traffic needed to implement the
What happens when, without mucking with the routing table, the tunnel is
enabled and you ping 172.16.0.2 -> 172.16.0.1 ?
On Wednesday 09 April 2003 12:55 pm, Jake Bullet wrote:
> I'm trying to setup a link between two VPNs.
> Bill Tux Fred
> 172.16.0.2 172.16.0.1 172.16.1.1 172.16.1.2
> Windows 2k <----> Linux Server <----> Windows 2k
> CIPE CIPE
> However I seem to be having some trouble with the routing. When I try to
> ping 172.16.1.1 from Bill the request is timing out. Running tcpdump on
> Bill's VPN shows that the ping requests aren't coming down the tunnel, so
> it's not a routing issue on the linux side, Bill isn't sending the packets
> down the tunnel at all.
> I tried rearranging the routing on Bill using the "route" command, but
> this only seems to break the tunnel completely. I have set the netmask on
> each of the windows clients to be 255.255.0.0 and 255.255.255.0 on Tux in
> the hope that the packets would get routed properly.
> The CIPE adapter doesn't seem to be acting like a normal interface as the
> moment you do anything to the routing table the tunnel fails and the
> tunnel has to be disabled in the cipe control panel and re-enabled to get
> it going again.
> Can anyone please help? It's really important that I get this working.
> P.S. I've tried it out on WindowsXP and it seems to work just fine, as too
> for Windows 2003 Server Beta RC2.