How do I remove a backup from my storage on Unraid GUI version?

Support
#1

Hi,

I’d like to remove a particular backup from my storage using the Unraid docker GUI version, but I"m not sure if it’s possible. Hoping someone could either confirm it’s not, or point me in the right direction if is is.

I’d been backing up my AppdataBackup folder, but have realised that, as it’s a new single compressed archive each day, no de-duplication by Duplicacy is possible so it’s taking up a load of space.

I now have a script running to take an uncompressed backup each night (from here if anyone’s interested: Reddit - Dive into anything). This is now part of my daily Duplicacy backup job (called AppdataBackupUncompressed), and the compressed backup job has been stopped.

Is there a way to remove the AppdataBackup backup from my storage?

Thanks

#2

The easiest would be to delete the subfolder corresponding to the snapshot under “snapshots” folder on your storage, and then run prune -a -exhaustive to clean up all orphaned chunks. You can do that from WebUI by creating a prune schedule and adding -a and -exhaustive flags.

#3

Amazing, thanks for the extremely quick and useful reply.

Just a confirmation if I could before I go ahead.

  • If my unwanted backup is called “AppdataBackup”
  • and its snapshots are here:
    /mnt/user/Backups/Duplicacy/snapshots/AppdataBackup
  • and there are 38 files in that folder called “1”, “2”, etc., each around 1kB in size

Would I delete this whole folder:
/mnt/user/Backups/Duplicacy/snapshots/AppdataBackup
then set up and run that prune job you specified?

Thanks again

#4

Yep, exactly. Those files are “snapshots” of each backup revision, once you delete them the storage will forget they ever existed, but the actual data chunks snapshots files were pointing to will still be in the storage. The prune exhaustive command will look at all chunks that are not referenced by any backup revisions and delete them, thus freeing the space.

It may take a few invocations as it tries to safely process concurrency scenarios: e.g. some backup job that is in progress may have just uploaded new chunk but not yet uploaded a snapshot, deleting that chunk would be a bad idea.

But if you want to purge immediately — ensure no other duplicacy job is running or can start against that storage and add -exclusive flag to the prune command. This turns off safety and will delete orphaned data immediately. Be very sure this prune is the only process running. Or bad things will happen :slight_smile:

#5

Excellent, thanks for the clarification. Appreciate the help!