Prune Chunks from Deleted Snapshot - missing preferences and -print-credentials help

Support
#1

I have removed a snapshot that I don’t need anymore and now I want to clean the chunks from storage. I’m running the WebUI docker (webUI version 1.7.2, CLI 3.1.0) on Unraid. I understand I need to run duplicacy prune -exhaustive but I’m hitting 2 roadblocks. First, getting credentials:

When setting up the WebUI, I selected encryption and I don’t have a copy of the b2 id and key unencrypted. I am attempting to run duplicacy-web -print-credentials to get those pieces of information so I can attempt to prune but when I run duplicacy-web -print-credentials I receive the following message:

/config # duplicacy_web -print-credentials
Duplicacy Web Edition 1.7.2 (1B2557)
Can’t start the web server at 127.0.0.1:3875: listen tcp 127.0.0.1:3875: bind: address already in use

How do I run the duplicacy-web -print-credentials command to get the credentials when I receive the above error message?

Second: when attempting to run duplicacy_linux_x64_3.1.0 prune -all, I receive the following message:

/config/bin # ./duplicacy_linux_x64_3.1.0 prune -all
Failed to read the preference file from repository /config/bin: open /config/bin/.duplicacy/preferences: no such file or directory

Correctly, the .duplicacy folder within /config/bin does not have any files. There are various “settings” files one directory up in /config but I receive the same error when I run ./bin/duplicacy_linux_x64_3.1.0 prune -all from the /config directory. What “preferences” is the duplicacy command looking for and how I do point the command at those files?

After this - to sanity check myself - I believe I’ll need to do the following:

  1. run duplicacy-web -print-credentials and record the credentials
  2. run duplicacy_linux_x64_3.1.0 init <snapshot ID> <storage URL> to initialize the storage
  3. run ‘duplicacy_linux_x64_3.1.0 prune -exhaustive’ to remove chunks associated with the now-deleted snapshot, presuming I can also find the preferences that the duplicacy command wants.

Any and all help is appreciated as this is the first time I’m trying to interact with the CLI to do anything.

#2

Why don’t you run prune from the UI?

Create a new schedule for prune, uncheck all days so it does not run. Edit options (click on existing options), add -exhaustive -a flags. Then start it manually by clicking the triangle.

Once done — delete the schedule. Or keep it for future use.

#3

I tried running a prune operation as the UI presents it but as the progress bar never showed anything in an hour I assumed it wasn’t working so jumped straight to trying with the CLI. I’m running as you have suggested now but the progress bar still isn’t moving. Is this expected? Any way to calculate how long this might take?

#4

WebUI runs CLI. It won’t be any different if you run it manually, just extra work configuring the repository again.

You can watch the log in the UI to monitor what’s going on. You can add a global flag -d to increase log verbosity.

It’s probably takes long time to enumerate all chunks. Some remotes are worse than others. For example, Google Drive with about 2 tb of data stored took two weeks for me to run exhaustive prune.

#5

Oh good lord. I was absolutely not expecting that kind of time. I figured 48 hours would be long but reasonable. I’m running this against a 4tb B2 remote that I’ve neglected to ever have pruned. I guess I’ll be waiting a while.

Am I blind: where can I view the logs in the UI? Or do you mean check in /logs through the terminal?