Quotas¶

A quota bounds how much a bucket may hold. Both dimensions default to unlimited.

Dimension	Limits
`bytes`	Current logical bytes in the bucket
`objects`	Currently visible objects in the bucket

Setting a quota¶

Send both dimensions. Each is either unlimited or a limit with a value:

curl -X PUT https://management.example.com/api/v1/buckets/uploads/quota \
  -H "Authorization: Bearer <your-management-token>" \
  -H "Content-Type: application/json" \
  -d '{"quota":{"bytes":{"mode":"limit","bytes":10737418240},"objects":{"mode":"unlimited"}}}'

That caps the bucket at 10 GiB with no object-count limit.

To remove a quota, set both back to unlimited:

curl -X PUT https://management.example.com/api/v1/buckets/uploads/quota \
  -H "Authorization: Bearer <your-management-token>" \
  -H "Content-Type: application/json" \
  -d '{"quota":{"bytes":{"mode":"unlimited"},"objects":{"mode":"unlimited"}}}'

The bucket record, including its quota, comes back in the response.

Enforcement¶

A write that would take the bucket past either limit is refused. The check happens before the object is committed, so a rejected upload leaves nothing behind.

Enforcement is on current usage. On a versioned bucket, non-current versions do not count toward the quota — which means a bucket can grow on disk while staying inside its quota. Watch both:

record-store storage inspect --endpoint https://management.example.com

See Capacity Planning.

Lowering a quota below current usage¶

This is refused. Setting a quota is itself validated against the bucket's present usage: if the bucket already holds more than the proposed limit, the request fails and the existing quota stays in place.

To shrink a bucket, delete objects first — or add a lifecycle rule and lower the quota once it has run.

What a quota does not do¶

It does not reserve space. Two buckets each capped at 10 GiB on a 15 GiB disk will still fill the disk.
It does not bound in-progress multipart uploads before completion.
It is per bucket. There is no deployment-wide quota.

Use quotas to stop one bucket consuming a shared deployment, and disk monitoring to stop the deployment running out. They answer different questions.