Which do you think is the best and most simple way to prevent automatic backups filling all disk space in Linux, before it actually happens?

I would like to automatically recycle the oldest ones when the disk reaches, let's say 85% full

  • What exactly is your problem? What are you doing, what are you expected to get, what you've got instead? – Nikita Kipriyanov Feb 15 at 19:24
  • Look at sizes of previous backups, compare them with df output. – berndbausch Feb 15 at 20:16
  • 4
    Remember to have a good monitoring system that will alert you before your disk fills up. – Michael Hampton Feb 16 at 13:38

I generally use rsnapshot with appropriate retention policies. It has the (big) added advantage to provide an incremental backup approach (ie: almost no additional space is used by two identical backups) while giving you always up-to-date "full tree" in the latest snapshot (ie: what other products call "syntetic full backup").

However, this approach must be combined with robust free space monitor/alarm, should used space increase faster then expected (for example via unique data, which are not shareable between incremental runs).


You should use a separated storage for backup, may be a network storage or a disk partition. That will keep your system safe from your backup job.

I agree with @shodanshok monitoring your system with a tool is another key. have a look to netdata !

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.