Okay, so I have searched around for the past hour or so and I am seeking some clarity on retention.
I have the current retention policy set at “Delete backups that are older than” 30 days.
My concern is that of individual files. Let’s assume we are speaking of an individual file that has been present for the past 30 days and backups are performed daily. I am hoping that the retention setting that I have chosen means that 1. if a file is deleted from the source it will be retained @ the destination for 30 days. 2. if the file is modified at the source 10 different times over 10 days the original version AND the multiple modified versions are retained @ the destination for 30 days since the first modification.
Not sure I am understanding exactly what you mean.
It’s actually very simple but sometimes we tend to overthink it.
With 30 day retention, you can restore files as they were up to 30 days ago. If a file is deleted, you have 30 days to realize it and restore the file.
If you are doing daily backups with a 30 day retention then yes, you could choose which version to restore within the past 30 days. You’d have 30 versions of the file to select from, one for each day. If you were doing hourly backups with 30 day retention, you could restore any of those hourly versions in the past 30 days. You’d technically have 720 past versions to choose from (30 x 24).
With my current retention setting only files that have been modified since last backup are backed up? Or, is this a full backup of all files every day?
I’m sorry for all the questions… I’m used to systems that do a weekly or monthly full and then do incremental or differential backups in between.
Only changed blocks of data are backed up. If a file is not modified, Duplicati doesn’t need to back it up again. When thinking about retention you can disregard that detail. Duplicati knows the file still exists so it isn’t going to purge its data from back end storage.
You can prove this to yourself. Check yesterday’s backup and you will see all files that existed on your system yesterday, even if they weren’t modified after the previous backup.
Block-based storage engine explains how Duplicati 1 did a full and then a chain of incrementals, whereas in Duplicati 2 it’s basically a block-based design where a file is saved as a series of blocks. Block changes get uploaded. Unchanged blocks simply get referenced again in the file list that defines the new backup. Details.