GET
/usageYour API usage
GET /usage
Requires the businesses scope.
Your key's own request rollups per endpoint: requests and error counts over the window. Days are Asia/Jerusalem calendar days.
Query parameters
| Name | Type | Description | |
|---|---|---|---|
days | integer | Window in days (default 30, max 365). | |
latency | string | Send 0 to skip the latency figures (latency is then null). One of: 0. |
Response
200: { "ok": true, … }
| Field | Type | Description | |
|---|---|---|---|
ok | boolean | required | |
days | integer | required | |
usage | array of object | required | |
daily | array of object | The same window as a series: one row per day (Israel time) that had traffic. | |
latency | object | Response times from the request log, which is kept for fewer days than the rollup: the figures cover the last days of the window the log still holds. | |
latency.count | integer | Requests the figures were computed over. | |
latency.p50 | integer | Median, ms. | |
latency.p95 | integer | 95th percentile, ms. | |
latency.max | integer | ms | |
latency.days | integer | The part of the window the log covers. |
Errors: 400, 401, 403, 404, 429. Every failure answers { "error", "reason", "hint", "param", "doc_url", "retryable", "request_id" }: branch on reason, and read hint for what to change (https://docs.billos.co.il/errors.md).