Move OneDrive to Wasabi via External Rclone

When Microsoft 365 data sets are large, moving them through a laptop can be slow and unreliable. Running Rclone on a cloud VM (EC2, GCE, or any Linux host) and driving it from RcloneView keeps traffic in the data center, avoids home-network bottlenecks, and makes browser-less auth possible.

You will:

Run rclone rcd on a remote server as an external engine.
Open a new RcloneView window that connects to that external Rclone.
Add OneDrive and Wasabi remotes (including headless/CLI-only auth paths).
Copy, sync, or schedule jobs from OneDrive to Wasabi without touching local bandwidth.

Part 1. Deploy Rclone on a Server (External Rclone)

Launch a small Linux VM (for example, t3.medium on AWS or equivalent).
Open TCP 5572 to your IP or tunnel over SSH.
Install Rclone:

curl https://rclone.org/install.sh | sudo bash

Start the remote-control daemon with auth:

rclone rcd --rc-addr=0.0.0.0:5572 --rc-user=admin --rc-pass=admin --log-level INFO

From your laptop, confirm it is reachable:

curl -u admin:admin -X POST "http://<SERVER-IP>:5572/rc/noop"

A {} response means the daemon is ready for RcloneView.

👉 Need a refresher? See Run Rclone on AWS EC2.

Part 2. Open a New RcloneView Window

Each RcloneView window can pair with a different Rclone instance.

Launch RcloneView.
Click New Window from the Home menu.
A second window opens; this one will connect to the external Rclone you just launched.

👉 Learn more: Using Multiple Rclone Connections with New Window.

Part 3. Connect RcloneView to the External Rclone

In the new window:

Open Settings -> Connection Manager -> New Connection.
Fill in:

Field	Value
Display Name	`external-rclone`
Remote Address	`http://<SERVER-IP>:5572`
Username	`admin`
Password	`admin`

Click Test Connection -> Save -> Connect. The status bar should show you are connected to the external daemon.

Part 4. Add Wasabi and OneDrive Remotes (Inside the External Rclone)

All remotes you create now live inside the external Rclone process and are shared by the connected RcloneView window.

➕ Add Wasabi (S3-Compatible)

Go to the Remote tab -> + New Remote.
Choose S3 / Wasabi.
Enter Access Key, Secret Key, and the endpoint for your region (for example s3.ap-northeast-2.wasabisys.com).
Save the remote (e.g., name it wasabi-prod).

👉 Details: How to Add Wasabi Remote.

➕ Add OneDrive (Works Without a Local Browser)

Click + New Remote again and select OneDrive.
If the server has a browser, complete the standard OAuth flow in RcloneView.
If the server is headless or cannot open a browser, follow the headless/CLI method: generate the token on a device with a browser and paste it into the server-side config.

👉 Step-by-step headless flow: Add Microsoft remotes from EC2/headless. After saving, you should see two remotes listed in the Explorer: OneDrive and Wasabi.

Part 5. Transfer or Sync to Wasabi

Method A: One-Time Copy/Sync

In the Explorer, open OneDrive on one side and Wasabi on the other.
Select the source folder on OneDrive and the destination bucket/folder on Wasabi.
Click Sync (makes destination match source) or use Copy -> for a simple copy.
Optionally run Dry Run first, then Run to start. Progress appears in the Transfer tab.

Method B: Save or Schedule Jobs

Configure a copy/sync as above, then choose Save to Jobs in the dialog.
Open Job Manager to re-run the saved job anytime.
To automate, enable Schedule in Job Manager (PLUS feature) and set the desired cadence.
Check Job History for logs and outcomes.

👉 More on jobs and scheduling:

Quick Tips for Faster Wasabi Uploads

Keep the external VM in the same region as Wasabi when possible to reduce latency.
For large objects, higher --transfers and --s3-upload-concurrency values can improve throughput; adjust gradually based on CPU and network.
Use Dry Run before destructive syncs to confirm what will change.

✅ Wrap-Up

By hosting Rclone as a remote daemon and steering it from a dedicated RcloneView window, you get reliable OneDrive -> Wasabi migrations without local bottlenecks. Headless auth flows keep you covered when no browser is available, and jobs/schedules make repeat runs effortless.