Clarification on backup IDs vs repositories vs Directories

[sshanky]

The docs mention that I can use list -all to list snapshots with any id. I tried this from a different directory where no init had been done but of course it tells me that the storage has not been initialized.

I’m wondering if there is a way to list all backup ids that are in a storage, so I don’t have to keep track of which directories have been initialized. Maybe I’m looking at this wrong and I shouldn’t have multiple initialized directories and try to work out of the root directory with a set of exclusions. But if this is the case I’m still not able to understand why for example I would ever need to use a command like list -all, since there will always only be one id for each repository initialized?

EDIT: I see that if I am in one of the directories that back up to the same storage, list -all does list backups from all ids that back up to that storage.

Thanks

[tangofan]

@sshanky

I’m wondering if there is a way to list all backup ids that are in a storage, so I don’t have to keep track of which directories have been initialized. Maybe I’m looking at this wrong and I shouldn’t have multiple initialized directories and try to work out of the root directory with a set of exclusions. But if this is the case I’m still not able to understand why for example I would ever need to use a command like list -all, since there will always only be one id for each repository initialized?

You could have multiple IDs from multiple backup roots in one storage (I have) or backup multiple IDs from different computers into the same storage.

For a given backup root on a particular computer, you would indeed in most cases use one backup ID for the root and then use include/exclude filters to select your files. The one possible exception would be if you have a subset of that backup root that you want to backup much more often. Then you could create a second backup ID for that subset and backup into the same storage. Having the same files in two different backup IDs may seem like a lot of duplication of data, but in reality Duplicacy’s deduplication feature would ensure that (in most cases) already existing file chunks of those files would be reused.

[gchen]

@sshanky

it tells me that the storage has not been initialized.

The error actually says the repository has not been initialized. You just need to run the init command from the new directory with any backup ID and then list -all should work.

[sshanky]

This makes sense. I must have interchanged the terminology “repository” with “storage” (at first they sound the same until you learn the meaning of “repository”). Thanks

[sshanky]

OK, I understand. This makes more sense. There’s a lot of strategy that has to go into setting it up
Thanks