To use Amanda on the network, you need to run two servers on the Amanda server as well as an Amanda client on each network host. Access must be enabled for both the clients and the server.
The Amanda server runs through xinetd, using xinetd service files located in /etc/xinetd.d. The two service files are amidxtape and amandaidx. Restart the xinetd daemon to have it take immediate effect.
For clients to be able to recover backups from the server, the clients' hostnames must be placed in the .amandahosts file in the server's Amanda user home directory, /var/lib/amanda. On the server, /var/lib/amanda/.amandahosts will list all the hosts that are backed up by Amanda.
Each host needs to allow access by the Amanda server. To do this, you place the hostname of the Amanda server in each client's .amandahosts dot file. This file is located in the client's Amanda user home directory, /var/lib/amanda.
Each host needs to run the Amanda client daemon, amanda, which also runs under xinetd. The /etc/xinetd.d/amanda configuration file is used to control enabling Amanda.
Tip If your server and hosts have firewalls, you will need to allow access through the ports that Amanda uses, usually 10080,10082, and 10083.
Was this article helpful?