|
Hello,
On Thu, 11 Oct 2001, djamil wrote:
> oh ! thank you for replaying !
> well you must be right, altho i followed some of the examples at the site
> [about the netmask and broadcast addresses].... because when i run ssh in
> debug mode it finally got me an error message : no route to host !
Following the examples is not always enough :)
> and mozilla finally got me an connection refused ....
> for hiding it i will try it tomorow [it's 00:00 here hehe !]
>
> but even after reading the docs i thaught i wouldn have because "UP RUNNING
> NOARP" in the ifconfig ...
Make your tcpdumps. NOARP does not play here. Where are these
wrong docs?
> also there is a firewall altho i double checked the trafic needed does go
> thru, unless u tell me there must be a special "opening" which would be
> surprizing 'cause ipip should be just another packet like another .... just a
> little bit longer ...
>
> i hope you have noticed the two machines are in two different dataceters ...
I see these different networks:
1. 212.43.221.77/32
2. 212.43.221.128/27
3. 212.73.232.192/26
You have the assumption that (1) can talk ARP with (3)
> also: .200 pings .146 and vice-vers-ca
>
> .200 pings .77 internally
> .146 pings .77 internally
>
> i hope this is normal !
ping to local IP is not enough for this setup to work. You
have to persuade gateway 212.73.232.193 to reply to ARP
probes in the form "who-has 212.73.232.193 tell 212.43.221.77".
If you hide 221.77 the probe will be "who-has 212.73.232.193 tell 200"
> snap-snap
>
> > >
> > > eth0:0 Link encap:Ethernet HWaddr 00:D0:B7:00:B1:DC
> > > inet addr:212.43.221.77 Bcast:212.43.221.77
> > > Mask:255.255.255.255
> > > UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
> > > Interrupt:10
> >
> > Are you sure that you can talk with 221.77? I don't see valid
> > network for it, it is not from the 221.146/27 network.
>
> what do u mean ? is the Bcast/Mask wrong ?
> i wish it is so i can move on !
221.146/27 covers 221.128 - 221.159, 77 is not there but I don't
understand your setup completely. An ASCII chart would be helpful if
you still can't find the problem with tracing. In any case, next step
is we to see a tcpdump output for your real test from client host.
Regards
--
Julian Anastasov <ja@xxxxxx>
|
