Mounting Google Drive with larger chunk size

I’m using Duplicacy with the Docker image saspus/duplicacy-web.

I’ve configured it a while ago with my Google Drive (gsuite) account and it’s been working fine until recently, where I hit the maximum of ca. 400K files.

So I figured I’d increase the chunk size to something larger than the default setting but I can’t for the life of me figure out how to create the gdrive repository with a different chunk size, while using this web-ui docker image.

Can anyone guide me thru a way to do this?

Move your datastore out of Shared/Team drive into your My Drive.

Can’t seem to find the limitations that exist on the “personal” drive. Do you know?

I don’t think there are any. Only the quota on total size (which may or may not be enforced) and daily ingress.

I might just try that out then. Regardless, it would be nice to have larger chunks just to keep the amount of files a bit lower. Isn’t the a way to do it?

EDIT: Also just noticed (great job me) you’re the one who created the docker container. Thanks for that! :slight_smile:

You can specify chunk sizes when you initialize the storage. Since for the storage to be copy compatible chunk size parameters shall be the same among other things you would not be able to copy backup to that storage.

So it seems your only way is to start a new backup into a new repository.

However increasing chunk size has its downsides… like higher storage space overhead.

Glad it’s useful. It was something I put together quickly for myself, a way to easily deploy duplicacy to Synology dsm, but suddenly other people started using it… so now I feel I can’t just abandon it and keep supporting it :slight_smile:

1 Like

Yeah, I’ll probably just start over, which is fine. I just can’t really figure out how to make the initialize the storage in cli (on the host machine) and then use it with the web ui in the docker container.

Oh, that’s simple. Just create a storage normally in the cli, passing all the parameters as you wish.

Then go through the motion of creating the same storage in UI — because the folders, passwords, and tokens provided would result in access to the storage that is already initialized all the rest of the parameters ( that you would have passed but could not) would be ignored and instead fetched from the existing config file.

1 Like

It really was simple, no idea why I overcomplicated it in my head.

Myeah, there’s probably a good reason for the default values. But since I got Gsuite and thus “unlimited” space (for now) the storage space overhead shouldn’t be that much of a concern. I just need to keep the amount of API calls pretty low, I guess, and - if I use shared drive - the amount of chunks as well.

Any idea if there’s a best practive as for avg and max size?

@saspus since I started over with what was meant to be a new empty repository, I haven’t been able to do checks. It fails with “missing chunks”. Thing is, I used the same repository name as previously which is where the confusing comes from, I think. By any chance, do you know how I can make duplicacy-web disregard previous chunks and threat the repository as a new one?

Confirm that the repository is actually new; I.e. in the new folder on the target and you haven’t accidentally simply linked to an old one. Also try deleting the cache folder (see Temporary Location or something like that in duplicacy settings). Nuke everything in that folder.