Duplicacy Web - Restore not showing all files that were backed up

Support
,
#1

I am running duplicacy web in a docker container on unraid, this is happeneing with all storages i have set up, SFTP and a local one as well running on the same server. I check the backup logs and it all looks fine, files are backing up as they should. However when i go to restore a file i cannot find that file that in the logs showed was backed up into that revision

Another and maybe unrelated issue is that prune commands arent working at all, they list the retention rules properly in the logs then state no revisions to remove when, for example i have one that does an hourly backup and is supposed to remove everything after 48 hours there are over 507 revisions now, so not a single one has been removed.

#2

So you couldn’t find the file you want to restore but did you see other files in the same subdirectory for example?

What are the prune options that you’re using for the prune job?

#3

Most of the folder structure exists but only some random odd files nothing significant.

The prune command gives output in the logs with the right retention duration just then states 0 revisions to remove even though there’s over 500 taken hourly and its set to remove any over 2 days old

#4

Is there anything special about these files? Perhaps some interesting characters in the filename? If the backup log indicate they were backed up – they are backed up. Perhaps webUI can’t parse the path for display? Can you use CLI to list files in that specific revision? (you can setup temporary repository in the temporary folder on another machine to check that).

Please post the prune command from the first few lines from the log. I’m wondering if for some reason id of the snapshot is missing or the -a flag.

#5

Nothing special, mostly documents, some with underscores. The CLI can list the files in the revision but cant seem to restore them it just makes empty directories

Running prune command from /cache/localhost/all
Options: [-log prune -storage production_snapshots -keep 0:2 -d]
2024-12-03 17:19:07.423 INFO STORAGE_SET Storage set to /backuproot/nbbackup/Production_Snapshots
2024-12-03 17:19:07.564 INFO RETENTION_POLICY Keep no snapshots older than 2 days
2024-12-03 17:19:11.683 INFO SNAPSHOT_NONE No snapshot to delete

Even thgough there are over 500 snapshots so definetly some to delete

#6

This does not specify which snapshot ID to process, so it does none. Add -a flag to process all snapshot IDs.

Any errors in the log? Or just silent failure? Maybe try -ignore-owner flag?

#7

-a worked thank you

I managed to get it to work running the restore from another computer to a different directory. I am still at a loss as to why the webui is not showing my files. As a paying customer of commercial web edition it degrades the service

#8

Did you select the correct backup id on the restore page (the second dropdown from the top)? You may have multiple backup ids there.

#9

Webgui simply parses what CLI outputs. Perhaps some bugs are there.

Many have voiced the same concern before, including myself here:

and in the whole this thread there:

Whether those are bugs or usability issues (UX bugs) – GUI is what users see and interact with.

It’s closed source, and hence something only @gchen could address.

#10

Yes i did and correct revision, which i confirmed contains the files with the list command in CLI

#11

The CLI just makes empty directories when you restore? You should post more logs from your backup, list and restore commands, and I suggest adding the -v flag - e.g.: duplicacy -v list -all to get more info.