(You must already be logged in to Gitlab or you’ll hit a 404)
The automated backup system contained in the Symbiosis system protects against accidental deletion or corruption of files. These backups are designed to run once per day and archive the contents of a number of important system directories.
Having backups stored locally is not sufficient to provide real protection from accidents though, as they might be removed or deleted. Therefore your local backups should be archived to a remote machine.
For Bytemark customers the backup script is configured to attempt to do this, using the remote backup space provided by Bytemark. This remote backup space is provided for Legacy Virtual Machines and Dedicated Servers only. For Cloud Server customers, additional storage can be purchased separately and can be mounted at
/var/backups, so that all backups are stored on a separate disk to the local system.
Your system will maintain full backups of the following locations:
Additionally each MySQL and PostgreSQL database you have upon your system will also be exported and backed up.
The backup system uses the backup2l program, which is configured to backup the
files in the above locations into the directory
more information about backup2l, please refer to its
As mentioned above, the backup script will attempt to ensure that your local backups are uploaded to a remote server, to protect against data loss if your system fails catastrophically.
For Bytemark customers using Legacy VMs or Dedicated Servers this location should be determined automatically.
It is also possible specify the correct location in
/etc/symbiosis/dns.d/backup.name. This should be a fully-specified rsync path. For example:
Every day, when runs it generates output saying what has been backed up, and if there were any errors during the backup process. This email will get sent to the root account of the local hostname, e.g., “firstname.lastname@example.org”.
It’s important to realise that the automated backups, especially their transfer to the remote backup space, is done on a best effort basis. You should carefully check the
backup2lreport for errors and from time to time practise recovering files at random from the remote server, to ensure that there are recoverable backups.