Can't remove old snapshots

I am still in testing mode here. I noticed on the 21st of July I had a big spike in stored data on B2. Looking back through logs I noticed that I was backing up transcoded data off of my Jellyfin install. I have now excluded them with the following backup options:

* +*/
* -jellyfin/config/transcodes/*

I decided to run a prune using -keep 0:1 -exhaustive -a

But in the logs I can see that nothing have been removed. I still have 12GB of data on B2 that should be gone. I assumed that keeping 0 snapshots older than 1 days (-keep 0:1) and specifying -exhaustive I would remove the previous snapshot on 21/07 and its referenced chunks.

Running prune command from /cache/localhost/all
Options: [-log prune -storage b2_sapling-backup01 -keep 0:1 -exhaustive -a]
2022-07-24 15:02:12.769 INFO STORAGE_SET Storage set to b2://sapling-backup01
2022-07-24 15:02:13.499 INFO BACKBLAZE_URL download URL is: https://f002.backblazeb2.com
2022-07-24 15:02:15.195 INFO RETENTION_POLICY Keep no snapshots older than 1 days
2022-07-24 15:02:16.383 INFO FOSSIL_COLLECT Fossil collection 1 found
2022-07-24 15:02:16.383 INFO FOSSIL_POSTPONE Fossils from collection 1 can't be deleted because deletion criteria aren't met
2022-07-24 15:02:16.385 INFO FOSSIL_COLLECT Fossil collection 2 found
2022-07-24 15:02:16.385 INFO FOSSIL_POSTPONE Fossils from collection 2 can't be deleted because deletion criteria aren't met
2022-07-24 15:02:16.388 INFO FOSSIL_COLLECT Fossil collection 3 found
2022-07-24 15:02:16.388 INFO FOSSIL_POSTPONE Fossils from collection 3 can't be deleted because deletion criteria aren't met
2022-07-24 15:02:16.390 INFO FOSSIL_COLLECT Fossil collection 4 found
2022-07-24 15:02:16.390 INFO FOSSIL_POSTPONE Fossils from collection 4 can't be deleted because deletion criteria aren't met
2022-07-24 15:02:18.986 INFO FOSSIL_COLLECT Fossil collection 5 saved

It looks like I have removed anything previous to the 3 test backups I ran today so why do I still have 12GB stored in B2. Shouldn’t all those transcoding file chucks be removed?

Log from 21/07:

jellyfin/config/transcodes/01b77eeb7b0ed38df5bf17601f9634f9126.ts (5426620)
2022-07-21 03:36:39.715 INFO UPLOAD_FILE Uploaded jellyfin/config/transcodes/01b77eeb7b0ed38df5bf17601f9634f9127.ts (5037084)
2022-07-21 03:36:39.715 INFO UPLOAD_FILE Uploaded jellyfin/config/transcodes/01b77eeb7b0ed38df5bf17601f9634f9128.ts (10123988)
2022-07-21 03:36:39.715 INFO UPLOAD_FILE Uploaded jellyfin/config/transcodes/01b77eeb7b0ed38df5bf17601f9634f9129.ts (4332084)
2022-07-21 03:36:39.715 INFO UPLOAD_FILE Uploaded jellyfin/config/transcodes/01b77eeb7b0ed38df5bf17601f9634f913.ts (20192140)

There are hundreds of these files. They are no longer on the file system (that directory is empty) and there arena snapshots prior to today, 24/07. How can the chunks still be on B2?

There are a lot of failsafes built in, eventually files will be deleted if you keep running it, I think it’s about a week.

If you want to purge the datastore immediately — add -exclusive flag but make absolutely sure no other client is touching the datastore during this.