Perhaps this experience might help someone in the future.
IRLP IP address problem:
The installation of CentOS 4.9 was done at home with my Belkin router. The IP address was set to 192.168.2.10.
After the setup, the node was moved back to it's home in the Civic Centre. I edited the /etc/sysconfig/network-scripts/ifcfg-eth0 file to change the IP address back to the original 192.168.1.150. Before editing, I dutifully copied ifcfg-eth0 to ifcfg-eth0-at-home to make a backup. I didn't reboot (blush). The node worked fine, but I realized later that I hadn't run pulsecheck and that the call sign was wrong on the off.wav file. I returned to make the changes and rebooted. The node couldn't get to the internet! At this point, Yvonne, VE3RYA (UNIX person) got involved.
Ifconfig eth0 showed all the right numbers. From a computer on the LAN, we could ping the node at 192.168.1.150, and we could ping another computer on the LAN. We could not ping outside the LAN.
Troubleshoot-irlp had the DNS error and all ports were not forwarded. Much hairpulling…
We finally noticed that a ping from the node showed as going from 192.168.2.10!!!!!!!!
Watching the boot carefully, we saw that the init used ifcfg-eth0, then also used ifcfg-eth0-at-home! The IP was being set to .1.150, then being set to .2.10, but ifconfig still showed .1.150. We deleted ifcfg-eth0-at-home, and the problem was solved.
Clearly, Linux is an operating system designed to make people feel good after beating their heads against the wall for hours before finding their problems ;-)
Bob VE3YX (and VE3GQW) Node 2520