Connect S3-compatible storage

Use this profile for any provider that speaks the S3 API but has no dedicated Plume profile. You supply the endpoint yourself.

1. Create credentials

1. Create a bucket with your provider.
2. Create an access key and secret that can list, read, write and delete objects in it.
3. Find the provider’s S3 endpoint URL and region in its documentation.
4. Check whether the provider expects path-style requests. Many self-hosted stores do.

2. Fill in the setup wizard

Open Media → Plume and choose S3-compatible.

Field	Where to find it
Bucket	The bucket name from your provider.
Region	The region your provider documents. Some providers accept any string.
Custom Endpoint	The full base URL, for example https://s3.example.com.
Access Key	The access key ID from your provider.
Secret Key	The secret from your provider.
Session Token (optional)	Only if your provider issues temporary credentials.
Path Prefix (optional)	A folder inside the bucket, for example media.
Custom Domain (optional)	The public address for files, if different from the endpoint.

Path-style requests are not set for you here. Under Show advanced connection fields on the Settings tab, turn on Use Path Style if your provider needs it.

3. Test and offload

Open Tools and select Test Connection. A working setup reports that the connection succeeded and write access is verified. Then run Offload existing library under Bulk media jobs, or queue files from the command line:

wp nimbus sync --limit=500
wp nimbus process-pending --limit=500

sync marks attachments as pending. process-pending offloads them. See bulk migration for details.

Notes

• If the endpoint is a Cloudflare R2 URL, Plume switches to R2 behavior. Use the Cloudflare R2 page instead.
• Plume does not set object ACLs. Make sure the bucket or your CDN serves the files publicly.
• If Test Connection fails, read the “Suggested checks” under the result. They name the likely cause.

Tip: If the test fails with a signature error, try toggling path-style requests first.