I have a Red Hat Enterprise Linux 4.7 server with two ethernet NICs on the same subnet. (Note that Linux names ethernet cards eth0, eth1,... eth
n.)
Code:
# ifconfig | egrep 'eth|inet'
eth0 Link encap:Ethernet HWaddr 00:22:19:04:EA:7A
inet addr:172.16.83.27 Bcast:172.16.83.31 Mask:255.255.255.224
eth3 Link encap:Ethernet HWaddr 00:15:17:8C:7C:D1
inet addr:172.16.83.29 Bcast:172.16.83.31 Mask:255.255.255.224
eth0 (172.16.83.27) has sshd listening on tcp port 22. eth3 (172.16.83.29) has httpd listening on tcp port 443.
Code:
# netstat -ltn
Active Internet connections (only servers)
Proto Recv-Q Send-Q Local Address Foreign Address State
tcp 0 0 172.16.83.27:22 0.0.0.0:* LISTEN
tcp 0 0 172.16.83.29:443 0.0.0.0:* LISTEN
I'm noticing some odd behavior: my host-level firewall logging shows that sometimes traffic to 172.16.83.29:443 is coming in on eth3, but other times traffic to 172.16.83.29:443 is coming in on eth0.
Here is my routing table:
Code:
# netstat -rn
Kernel IP routing table
Destination Gateway Genmask Flags MSS Window irtt Iface
172.16.83.0 0.0.0.0 255.255.255.224 U 0 0 0 eth0
172.16.83.0 0.0.0.0 255.255.255.224 U 0 0 0 eth3
169.254.0.0 0.0.0.0 255.255.0.0 U 0 0 0 eth3
0.0.0.0 172.16.83.1 0.0.0.0 UG 0 0 0 eth0
I suspect the behavior I'm seeing could be solved by a routing table entry, but I am not confident that I know exactly what that entry should be. I'm willing to experiment a bit (while physically on site, in case I break the networking), but could someone speak to whether this makes sense and/or whether I appear to be barking up the right tree?