Backup

We use both backupninja, which provides a tool-agnostic framework for configuring backup jobs, and borgbackup, which performs a network backup of the data.

Via ansible, we can designate a server’s backup_profile as either a target (the backup server) or a source (any server that should specify directories to backup).

When defining sources, you can specify a list of remotes (in case you want to backup to more then one remote backup servers), and also a list of directories to include.

Setting up Backup Servers

We use the xfs filesystem and create big partitions in /home.

To get started:

mkfs -t xfs /dev/sdb

Then you can add it to /etc/fstab like any other filesystem.

Transitioning from puppet

At the moment, our main backup servers are managed by puppet. Yet, our new ansible servers still need to backup to them.

To handle the transition:

• Backup sources in ansible: When specifying a server that should be the source of a backup,, use the new, ansible way (.e.g. specify that backup_profile is set to source, and designate the target backup server and the directories to be backed up like you normally would.

• For each source that you add, you need to add a setting in the old puppet configuration too, since you can’t set it up in ansible.

  • Pick the right .pp file. Since we don’t have a .pp file for all the new ansible virtual servers, instead choose the .pp file for the physical host. We will maintain both ansible .yml and dupcliate puppet .pp files files for all physical hosts during the transition.

  • Add an entry in the hosts .pp file along these lines:

if ( $::fqdn == "XXXXX.mayfirst.org" ) {
  mayfirst::m_backupninja::server::configure_node { "XXX003": }
}

• Add the server’s public key to the $mfpl_ssh_public_keys dictionary in the globals.pp file keyed to the server name as specified in the configure_node line.