I took some time to read the whole thread you gave me. Here is the
situation (all CIP, VIP and RIP are at the same subnet).
RS$tcpdump -n src host CIP and dst host VIP
19:12:52.685309 tunl0 < CIP.32819 > VIP.telnet blah blah...
So client packets got to the tunl0.
RS$tcpdump -n src host VIP and dst host CIP
nothing happened. Bad.
RS$traceroute -n -s VIP CIP (failed)
and
Client$tcpdump -n host VIP
eth0 < VIP.33838 > CIP.33437: udp ...
eht0 > CIP > VIP:cimp: CIP udp port 33437 unreachable
RS$ip route get from CIP to VIP iif dummy0
local VIP from CIP dev lo src VIP cache <local> iif dummy0
It seems that the real server get the ipip packet in tunl0, but somehow
failed to send out packet back to client. I don't think it's the route
problem because everything is on the same subnet. Below is the output
of ip route:
RS$ip route
x.y.z.128/25 dev eth0 proto kernel scope link src RIP
127.0.0.0/8 dev lo scope link
default via GW dev eth0
I'm feeling I'm almost getting it but not yet.
Please help me out,
-Tao
|