Facebook
TwitterWe are running CentOS 6.4 with minimal desktop environment and plesk. When we build the server we had only 2 IPs x.x.x.42 (primary) and x.x.x.45 (shared). Now after adding a third IP (shared) via Plesk control panel problems appear.
Under default CentOS network manager, plesk created these connections:
eth0 (this should be the primary connection, it has the primary IP, gateway and other settings)
System eth0:1 (only the x.x.x.45 shared IP)
System eth0:2 (only the newly added x.x.x.46 shared IP)
After every server restart, CentOS chooses the System eth0:2 as an active connection. Then the errors start to pop up:
Unable to configure a web server on the host <xxx>
nginx: [emerg] bind() to x.x.x.45:80 failed (99: Cannot assign requested address)
nginx: configuration file /etc/nginx/nginx.conf test failed
Unable to configure a web server on the host <xxx>
nginx: [emerg] bind() to x.x.x.46:80 failed (99: Cannot assign requested address)
nginx: configuration file /etc/nginx/nginx.conf test failed
After that none of the other services work (bind, mail, ...)
To fix this, we must manually set eth0 as the active network connection, then go to Plesk control panel under IP management, choose the option to reread the IPs and then to repair IPs x.x.x.45 and x.x.x.46. After that is done, everything starts working again.
I would greatly appreciate any advice you can give me. I'm really not a Linux expert and have great trouble understanding networking in linux. I found 3 sets of network configuration files (with identical settings) under 3 directories and have no idea which one are actually in use.
Under default CentOS network manager, plesk created these connections:
eth0 (this should be the primary connection, it has the primary IP, gateway and other settings)
System eth0:1 (only the x.x.x.45 shared IP)
System eth0:2 (only the newly added x.x.x.46 shared IP)
After every server restart, CentOS chooses the System eth0:2 as an active connection. Then the errors start to pop up:
Unable to configure a web server on the host <xxx>
nginx: [emerg] bind() to x.x.x.45:80 failed (99: Cannot assign requested address)
nginx: configuration file /etc/nginx/nginx.conf test failed
Unable to configure a web server on the host <xxx>
nginx: [emerg] bind() to x.x.x.46:80 failed (99: Cannot assign requested address)
nginx: configuration file /etc/nginx/nginx.conf test failed
After that none of the other services work (bind, mail, ...)
To fix this, we must manually set eth0 as the active network connection, then go to Plesk control panel under IP management, choose the option to reread the IPs and then to repair IPs x.x.x.45 and x.x.x.46. After that is done, everything starts working again.
I would greatly appreciate any advice you can give me. I'm really not a Linux expert and have great trouble understanding networking in linux. I found 3 sets of network configuration files (with identical settings) under 3 directories and have no idea which one are actually in use.
