I’m a new Web GUI user and successfully had my first backup with Wasabi using the Quick Start tutorial. I’m still lost on when to use the Prune and Check functionality but see that I can add that to a job. I did a search but didn’t see it laid out in any tutorial or similar newbie friendly topics. Does anyone have any recommendations on Prune and Check best practices for a personal dateset that may change monthly and I’m doing weekly backups?
If you’re doing weekly backups, I would suggest running a check job and a prune job after every backup in the same schedule. The option for the prune job may be -keep 0:1800 -a which means to delete backups older than 5 years. You can enter any value at the configuration page of the prune job – after creating the prune job, click the option to change it to -keep 0:1800 -a.
If you’re doing daily backups, can check and prune also run after every backup, or should they run on a different (less frequent) schedule?
If they run in the same schedule, should they run sequentially?
It’s entirely up to you.
The only caveat is that pretty graphs in web ui update based on the information generated by check. So if you want to see data there — have check run periodically.
Duplicacy is fully concurrent, you can sequence the operations, run them in parallel, or even from different hosts.
That said, running check makes sense after prune, to make sure datastore is consistent after the operation that deletes data. Running check after every backup is not that important— backup only adds new data. It does not modify existing files. (But if you want graphs - you have to)
For average user data the good approach is to do backup hourly, check daily, and prune weekly, with the default retention rules. And then tweak from there according to your needs.
Personally, i backup daily, check monthly, and never prune — but you may decide some other approach works better.
1 Like