Package: sbackup Version: 0.8-1 Severity: wishlist It's well known amongst sysadmins that you don't have backups unless you've _tested_ them.
To ensure that the backup arrangements are working correctly, the system should periodically guide the user through a test restore, in which the original live data is moved aside, and data from the just-performed backup is restored in its place. Then, if anything is wrong with the backup system the user will notice and be able to put the live data back before investigating, rather than finding during disaster recovery that the backups are broken. Exactly how this should be done and what double-checks should be made (for example, regarding parts of the system _not_ restored) is a nontrivial question. (FYI: my tests have all been on Ubuntu Breezy. Adam Conrad said I should send my request to the Debian BTS.) Ian. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
