Hi,
I'm not sure whether this is a bug or not, so I'm throwing
this out here.
On a NetBSD 4 system, I have a public subnet routed through
a gif
interface. Because I wanted to use all of the IPv4
addresses, I set
net.inet.ip.hostzerobroadcast to 0 and set up a machine at
that address.
On the router, my setup looks like:
ex0:
flags=8863<UP,BROADCAST,NOTRAILERS,RUNNING,SIMPLEX,MULTIC
AST> mtu 1500
capabilities=3f00<IP4CSUM_Rx,IP4CSUM_Tx,TCP4CSUM_Rx,TCP4C
SUM_Tx,UDP4CSUM_Rx,UDP4CSUM_Tx>
enabled=3f00<IP4CSUM_Rx,IP4CSUM_Tx,TCP4CSUM_Rx,TCP4CSUM_T
x,UDP4CSUM_Rx,UDP4CSUM_Tx>
address: 00:50:04:d0:32:a8
media: Ethernet autoselect (100baseTX full-duplex)
status: active
inet 72.34.229.6 netmask 0xfffffff0 broadcast
72.34.229.15
inet6 fe80::250:4ff:fed0:32a8%ex0 prefixlen 64
scopeid 0x1
ex1:
flags=8863<UP,BROADCAST,NOTRAILERS,RUNNING,SIMPLEX,MULTIC
AST> mtu 1500
capabilities=3f00<IP4CSUM_Rx,IP4CSUM_Tx,TCP4CSUM_Rx,TCP4C
SUM_Tx,UDP4CSUM_Rx,UDP4CSUM_Tx>
enabled=3f00<IP4CSUM_Rx,IP4CSUM_Tx,TCP4CSUM_Rx,TCP4CSUM_T
x,UDP4CSUM_Rx,UDP4CSUM_Tx>
address: 00:50:da:ca:10:31
media: Ethernet autoselect (100baseTX full-duplex)
status: active
inet 192.168.10.1 netmask 0xffffff00 broadcast
192.168.10.255
inet alias 38.98.18.161 netmask 0xfffffff0
broadcast 38.98.18.175
inet6 fe80::250:daff:feca:1031%ex1 prefixlen 64
scopeid 0x2
gif0: flags=8051<UP,POINTOPOINT,RUNNING,MULTICAST> mtu
1280
tunnel inet 72.34.229.6 --> 38.98.18.100
inet 192.168.255.1 -> 192.168.255.0 netmask
0xfffffffe
inet6 fe80::250:4ff:fed0:32a8%gif0 -> prefixlen
64 scopeid 0x5
This machine also provides IP NAT for, amongst other
networks,
192.168.10.x. Routing works fine, and all of the machines,
including
38.98.18.160, are available on the public internet.
The strange part is this: when I try to ssh to 38.98.18.160
from somewhere
on the 192.168.10.x network, I get connected to the
router's ssh daemon.
Does anyone know why routing would work fine to
38.98.18.160, but NAT
wouldn't?
Thanks,
John Klos
--
Today is the day that my destiny calls me!
Meowth
|