[TUHS] fork

[Folkert van Heusden via TUHS]

>> I think it does because else it would not be able to communicate at 
>> all I
>> think. Maybe the ARP entries are evicted from the cache (on either 
>> side) for
>> some reason.
>> 
>> Something else is strange too: I configured the netmask to 
>> 255.255.254.0
>> (for 192.168.64.0/23) and it sets that to 255.255.255.0.
> 
> Not sure 2.11BSD supports CIDR (aka. classless) or if you have to
> enable it.
> 
> If not then 192.168.x.x will always end up with a /24 (Class C) netmask
> and things like 172.x.x.x a /16 (Class B) and 10.x.x.x with a /8 (Class 
> A)
> 
> Also such a netmask mismatch between your router/gw and the device will
> cause broadcast (and thus ARP) handling issues. Modern devices will
> (re)populate their ARP table if they receive a packet from an unknown
> device so they learn quicker and things can kinda limp along but in
> reality it is b0rken ;)

Yes, but if you take /24 they both are still in the same segment. But I 
agree it is not ideal.
So I added vxlan to my emulator allowing me to insert/retrieve the 
packets in a /24 segment I have elsewhere.

-- 
www.vanheusden.com