Hi all,
Our team has access to a server with 8 terabytes of storage available on a mounted drive. We connect to this server via SSH to download data and run our machine learning workflows.
I’m currently setting up DVC to use a specific directory on the server (e.g., /mnt/raid/datasets
) as a remote storage location. The idea is to treat this directory like a local DVC remote, allowing us to push and pull data using DVC, and use Git alongside it to track dataset changes.
Here’s the issue I’m running into:
- I set up a local DVC remote pointing to
/mnt/raid/datasets
- I create a new data file, then run
dvc add
and commit the generated.dvc
file with Git - However, when I run
dvc push
, the data does not appear in the remote directory - Similarly,
dvc pull
does not bring data into my working directory from the remote
Am I misunderstanding how local DVC remotes are supposed to work? Or is there something I might be missing in the configuration or usage?