Skip to main content
OneBucket stores nothing itself — your Endpoints (the object stores you already have: AWS S3, Azure Blob, GCS, Wasabi, on-prem, edge) hold the data. A policy maps one OneBucket™ to one or more Peer Buckets at your Endpoints, so reads and writes signed with your credentials know where to go. This is a one-time setup in your OneBucket console, and it applies account-wide — across every cluster your account uses. Once deployed, any S3 client or MCP connector can use your storage — see the Quickstart.
1

Create your credentials

On the Credentials page, generate (or import) your Access Key ID / Secret Access Key pair — the identity your S3 clients and connectors sign requests with.
Secret credentials are shown only once — copy or download them right away.
2

Add an Endpoint

On the Endpoints page, register a backend: a friendly name, its URL(s), and protocol (S3 or Azure). Then add the Endpoint’s own credentials — an existing Access Key ID / Secret Access Key (S3) or Account Name (Azure) — so OneBucket can reach it.
3

Create a policy

On the Policies page, create a OneBucket™ (* matches any Peer Bucket) and map it to one or more Peer Buckets at your Endpoints — select an existing bucket or enter a new name and OneBucket creates it. Optionally rename with a prefix/suffix (e.g. backup-*). Set the write Quorum — how many Peer Buckets with write permission must succeed — and select the active credentials the policy applies to.
4

Deploy

Deploy your policies to put them into effect — a deployment is the collection of policies that runs, and only one deployment runs at a time. Policies without at least one active Core Access Key, or without at least one active Sync Peer Bucket, are ignored.

​
Next

Quickstart

Read and write your first object over S3 or MCP.