On Wed, Jan 02, 2002 at 01:39:20PM +0000, Ganesh Sittampalam wrote:

> > Aha, now it makes more sense.  Thanks!
> > 
> > What would be interesting to find out is why rocky is seeing sandy on
> > eth0.  Can you run a tcpdump from rocky in raw mode 'tcpdump -i eth0 -R
> > -e -x -v -n' when you try to access the outside world from sandy?
> 
> As far as I can see there's two distinct kinds of packets (and lots of 
> repeats of each):
> 
> 08:39:11.348738 0:50:56:46:63:ac 1:80:c2:0:0:0 0026 60: 802.1d config 
> 8000.00:50:56:46:63:ac.8001 root 8000.00:50:56:46:63:ac pathcost 0 age 0 max 20 
> hello 2 fdelay 15 
>                          4242 0300 0000 0000 8000 0050 5646 63ac
>                          0000 0000 8000 0050 5646 63ac 8001 0000
>                          1400 0200 0f00 a5a5 a5a5 a5a5 a5a5

These are spanning tree packets from rocky, quite normal.


> 08:39:16.118899 0:50:56:c0:0:1 ff:ff:ff:ff:ff:ff 0806 60: arp who-has 
> 10.3.8.2 tell 10.3.8.67
>                          0001 0800 0604 0001 0050 56c0 0001 0a03
>                          0843 0000 0000 0000 0a03 0802 0000 0000
>                          0000 0000 0000 0000 0000 0000 0000

OK, sandy thinks 10.3.8.2 is behind its 'normal' interface (not the
interface to rocky) for some reason.  Can you get a dump of your sandy
routing table (c:\winnt\route print)?


cheers,
Lennert
_______________________________________________
Bridge mailing list
[EMAIL PROTECTED]
http://www.math.leidenuniv.nl/mailman/listinfo/bridge

Reply via email to