How to backup multiple directories to the same backup destination?

I have few directories containing documents that sometimes get moved from one to another. I would like to backup the both directory trees to the same backup destination (B2 Bucket) to save in data transfer time in case documents get moved form one directory-tree to another.

Should I:

  1. init the backup two times (once for each directory) for the same repository ?

OR

  1. Use symlinks and create a separate (artificial) “root” directory and place the symlinks there pointing to the root of both the directories?

If the approach #1 also works, can I backup partially overlapping repositories from different computers to the same backup Storage? I mean the case were some computers have only subset of the total content in a repository.

Both 1) and 2) should work, although 2) is more of a workaround for the GUI version since it can only backup one directory. For the CLI version 1) is recommended.

And yes, you can backup different directories on the same or different computers to the same B2 bucket. Duplicacy is very flexible in this regard due to the lock-free deduplication approach.

Thanks, I went to read the design docs more carefully and went to test #1. Works all good.

Hello,

When I will be able to backup more than one directory with the GUI version?

It has been going slow due to recent Crashplan effect. The new estimate is end of October if nothing special happens again before then.

What is the “Crashplan effect”? I mean, I know what happened to CP, but are you overwhelmed with support due to this? Just curious…

Yes, there had been a sudden surge on emails, issues, posts, etc. It has calmed down a bit so I started to have more time on development.

Ok, thanks for the clarification.
Hope that this will lead to a more mature product eventually :slight_smile:

Wait, how does 1) actually work (CLI)? Two different repository directories but with the same repository id? This configuration doesn’t seem to be mentioned in any of the docs… After init, does backup have to be run from each directory? My first thought is the snapshot state will get nuked by the other.

I’m currently evaluating the CLI at the moment and thought I knew how it worked. Sorry to add to the CrashPlan effect. :slight_smile:

What JarnoP meant was to backup two directories to the same storage. Our terminology is somewhat confusing because the meaning of repository is different from what is usual for other backup tools.

Aha ok. Yea I’m aware of the github talk re terminology. So for CLI, you would suggest in order to handle multiple backup directories, instead of symlinks, to have unique ‘repository ids’ for each directory root and then to iterate over them in a batch file say? This does seem better considering the current issue surrounding VSS on other volumes…

Same request here, need to be able schedule backup of multiple repositories on same PC through GUI. I’m aware that I can do this through CLI, but scheduling and emailing of logs from GUI is very helpful. I have no desire to learn batch scripts and CLI just to do something the GUI can mostly do already.

Is just to confirm. If I have to different directories im backing up and they back up to the same b2 destination, this will be ok?

Will the backup checker freak out?

1 Like

All ok. You can back up to the same destination as many different folders as you want, even from different computers. All you should care is to name each repository different (different snapshot name), so that duplicacy can properly keep the history per-repo. that’s all

I am backing up mac#1 and mac#2 to my synology NAS. I have setup folder_A on synology and mac#1 is backing up to folder_A.

When i go to setup mac#2, I get this “The storage has been initialized with a password. Please enter a unique name for the storage and the password below.”

So, do i enter the SAME storage name and password as used for mac#1…OR… can i use a different storage name? I am fairly certain the password has to be the same, correct?

Thanks.

Same password, yep… that unlocks the storage for any machine backup/restore etc…

The storage name can be anything as it’s just a friendly name for the Web interface. May as well use the same name, though, since it’s the same storage and it’s probably best to identify it by computer name / location. For consistency.