S3-Compatible Storage Setup Guide

The S3-Compatible provider connects Blober to storage services that support the common Amazon S3 protocol operations Blober uses. You give it the service's endpoint URL and access keys, so this one provider covers the long tail of S3-compatible services, self-hosted servers, and local test setups.

Path format:

bucket-name/path/to/file.ext

Most services with an S3-compatible API for common bucket and object operations work. They fall into these groups:

• Cloud object storage: Alibaba OSS, Tencent COS, Huawei OBS, IBM COS, Oracle OCI, Google Cloud Storage, Yandex, and more.
• Object storage specialists: Storj, IDrive e2, Cubbit, Impossible Cloud, Seagate Lyve Cloud, Synology C2, MEGA S4, Tebi, Tigris, and more.
• Hosting and VPS storage: Linode/Akamai, Vultr, American Cloud, OVHcloud, Hetzner, Scaleway, Contabo, Exoscale, IONOS, and more.
• Self-hosted and open source: MinIO (the common local-testing server), Ceph / RADOS Gateway, SeaweedFS, Garage, Zenko, Versity Gateway, and more.
• Enterprise and on-premises: NetApp StorageGRID, Dell ECS, Cloudian, Pure FlashBlade, Hitachi HCP, Scality, Zadara, and more.
• Any self-hosted or on-premise S3 server on your own network, or a server on your own machine for local testing (e.g. http://localhost:9000).

For the full roster with endpoint formats and setup notes, see The Complete List of S3-Compatible Storage Providers.

S3-compatible does not mean feature-identical to AWS S3. Blober targets browsing, uploading, downloading, deleting, and normal copy workflows. Advanced behavior such as object lock, lifecycle policies, server-side copy, checksums, and metadata can differ by provider.

• ✅ Browse buckets and objects
• ✅ Upload files (including large files)
• ✅ Download files
• ✅ Delete objects
• ✅ Copy/move objects
• ✅ Both addressing styles: virtual-hosted (https://my-bucket.s3.example.com) and path-style (http://localhost:9000/my-bucket)
• ✅ Single-bucket or full-account access

Enter your endpoint URL in one of two fields. The field you choose sets the addressing style:

• Virtual-host endpoint URL is for virtual-hosted-style S3 addressing, common with hosted providers. The bucket name is added to the hostname as a subdomain (so https://s3.example.com becomes https://my-bucket.s3.example.com). Enter the host only.
• Path-style endpoint URL is for path-style S3 addressing, common with MinIO, Ceph, and many self-hosted or local servers. The bucket name is added to the URL path (so http://localhost:9000 becomes http://localhost:9000/my-bucket). Use this field if uploads or browsing fail against your own server.

Include the scheme (https://, or http:// for a local server) and enter only the host, without a bucket or path. Typical values are https://s3.us-east-1.example.com (virtual-hosted) and http://localhost:9000 (path-style).

Find it in your provider's console, under access keys or service accounts.

The console shows it once, when you create the access key, so store it somewhere safe.

• Default: empty (browse every bucket the keys can list)
• Enter a bucket to limit Blober to that bucket. This is required if your keys are restricted and cannot list all buckets, which is common for migration keys.

• Default: us-east-1
• Most S3-compatible servers ignore the region, so the default is usually fine. Set it only if your provider documents a specific region. Some providers accept only the region of the endpoint you entered.

• Default: STANDARD
• The storage tier for uploaded files. STANDARD is the safest default for ordinary S3-compatible uploads. Some providers also offer tiers like STANDARD_IA or GLACIER, or their own class names; set one only if your provider documents it. Applies when this provider is the destination.

1. Start MinIO (for a quick local test, the default keys are minioadmin / minioadmin).
2. In Blober, add an S3-Compatible source or destination and enter:
   • Path-style endpoint URL: http://localhost:9000 (or your server's URL)
   • Access Key ID and Secret Access Key: your MinIO credentials
   • Bucket name: optional; leave it empty to see all buckets
3. Browse to confirm the connection.

1. Create access keys in the provider's console.
2. Copy the S3 endpoint URL from their docs (e.g. https://s3.region.provider.com).
3. In Blober, put the endpoint in the Virtual-host endpoint URL field, enter the keys, and browse.

American Cloud's A2 object storage expects path-style addressing^[1].

1. In the American Cloud console, go to Storage > Object storage and open your storage unit. Its S3 Access panel shows the endpoint and the keys^[2].
2. In Blober, add an S3-Compatible source or destination and enter:
   • Path-style endpoint URL: https://a2-west.americancloud.com (or the endpoint the panel shows)
   • Access Key ID and Secret Access Key: the keys from the panel
3. Browse to confirm the connection.

STRG is built into Blober. Select STRG as the source or destination instead of S3-Compatible. The form asks only for the Access Key ID and the Secret Access Key, and Blober finds the region of each bucket. The STRG setup guide has the steps.

Filebase documents both virtual-hosted and path-style S3 addressing ^[3]:

• Virtual-host endpoint URL: https://s3.filebase.io gives object URLs like https://<bucket-name>.s3.filebase.io/<object-key>.
• Path-style endpoint URL: https://s3.filebase.io gives object URLs like https://s3.filebase.io/<bucket-name>/<object-key>.

Use the addressing style your Filebase workflow or toolchain expects; both point at the same S3 service host.

Use the Path-style endpoint URL field instead of the virtual-host one. MinIO, Ceph, most on-premise servers, and some hosted providers such as American Cloud require path-style addressing.

Your keys may be scoped to one bucket and not allowed to list all buckets. Enter that bucket in the Bucket name field.

Set the Region field to the region of your endpoint. Some providers reject requests signed for a different region. They also report a bucket in another region as not found. For that bucket, add another S3-Compatible source or destination with that region's endpoint.

Check the endpoint URL again. It needs the scheme (https://, or http:// for a local server) followed by the service host and nothing else: no bucket and no object path.

• rclone S3 backend: a reference list of S3-compatible providers, their endpoints, and quirks
• S3 path-style vs virtual-hosted addressing: background for the two endpoint fields above
• Ceph Object Gateway S3 API: the S3 compatibility reference for a common self-hosted server