Monitor bucket usage

Documentation Index

Fetch the complete documentation index at: https://docs.rafftechnologies.com/llms.txt

Use this file to discover all available pages before exploring further.

_{Updated May 8, 2026} Every bucket has a Metrics tab that shows live usage and the numbers your bill is based on. It’s the first place to look when you’re verifying storage growth, watching request volume after a launch, or sanity-checking egress before the month-end settlement.

​

Open the Metrics tab

In Object Storage → Buckets, click into a bucket. The bucket detail page has five tabs across the top — Objects, Properties, Permissions, Metrics, Management. Click Metrics. The page is organized into three sections — Storage Metrics, Request Metrics, and Data Transfer — each pulled live from the Ceph RGW usage API.

​

Storage Metrics

Current state of the bucket — these are point-in-time figures, not month-to-date. Total Size is the headline number for storage billing. The first 100 GB across all your buckets is covered by the $7/month subscription; anything above accrues at $0.07/GB/month.

​

Request Metrics

GET, PUT, and DELETE operation counts for the current calendar month. Resets on the 1st. Request counts are not billed today — Raff doesn’t charge per-request. The numbers are here for visibility: spotting a runaway script, verifying CDN cache-hit ratios, or sizing capacity before a launch.

​

Data Transfer

Inbound and outbound bytes for the current month, also resetting on the 1st. Egress is the only data-transfer line item that ever charges. Same-region traffic between Raff VMs / Kubernetes and Object Storage doesn’t go over the public internet, so it doesn’t show up in this counter and doesn’t bill — see Bandwidth. When you’re approaching 1 TB of public egress in a month, this is the page that shows it.

​

How fresh are the numbers?

Metrics are read from the Ceph RGW usage API and refreshed when you load the tab. Storage figures update within a few seconds of an upload or delete. Request and data-transfer counters lag by a short interval (usually under a minute) — they’re aggregated server-side, not streamed. Reload the tab to pull the latest values; nothing is cached on the dashboard side.

​

Per-account vs per-bucket

The Metrics tab is per bucket. To see usage rolled up across every bucket in your account — total storage, total egress, what you’ll be billed for at month-end — open Object Storage → Overview.

​

When the numbers don’t match what you expect

A few common gotchas before you open a support ticket:

• Storage size jumped after a delete — versioning is on, so the deleted object became a non-current version (still counted). Empty old versions from the Objects tab → Versions view.
• Egress higher than your traffic — public-bucket workloads (asset hosting, static sites) count every public read; if a CDN is in front, check whether it’s actually caching.
• Request count keeps climbing while idle — something has access keys and is polling. Rotate the keys and watch Request Metrics drop.
• Total Operations far higher than GET + PUT + DELETE — bucket-level calls (HeadBucket, GetBucketAcl, GetBucketCors, etc.) count toward Total Operations but aren’t broken out individually. SDKs do these on every connection setup.

​

Related