# Read a check's results

See a check's uptime, response times, outages and latest results, per location.

Replace _check-id_ with the check's id.

## Prerequisites

- Permission for `monitoring:GetCheckResults` on the check,
  `rn:ews::<organization id>:monitoring:check/<check id>`. Without it, the API answers `404` "Check
  not found", and the console's check page shows an access-denied notice in place of the results.
- In the console, also `monitoring:DescribeChecks` to open **Uptime**, and on the check to open its
  page (without it the check's page is not found).

## Read the results

**Console**
1. Open **Monitoring**, then **Uptime**, and choose the check's row.
2. Under **Response time**, choose a range: **1 hour**, **24 hours** (the default), **7 days** or
   **30 days**. The tiles show the uptime of the range, **Runs that succeeded**,
   **Average response** and **95th percentile**, over a chart of the response time.
3. **Locations** shows the newest result from each location among the latest 50 results.
4. **Outages** lists each outage with when it started and ended, how long it lasted and its cause.
5. **Latest results** lists the 50 newest results in the range, one per run and location, with
   where it ran from, its status, time and detail.

**CLI**
```sh
dcloud monitoring checks results check-id --start 2026-11-01T00:00:00Z --limit 100
dcloud monitoring checks outages check-id
```

**API**
```http
GET /monitoring/api/checks/check-id/results?start=2026-11-01T00:00:00Z&limit=100
```

```http
GET /monitoring/api/checks/check-id/outages
```

## Result

The results answer covers `start` (a day before `end` by default) to `end` (now by default):

- `summary`: the results (one per run and location, so the console's **Runs that succeeded** counts
  each location's answer), how many succeeded, the average and 95th-percentile response time of the
  successful ones, and
  `uptimePercent`, the share of the window, from the check's creation on, not covered by an outage.
  Time while the check was paused counts as up.
- `series`: one point per period that has results (a period without any is left out), with the
  results, the successful ones, and the average and maximum response time of the successful ones.
  The period (`periodSeconds`) is a whole number of minutes chosen so the window holds about 300
  periods, and at least one minute.
- `locations`: the newest result from each location among the raw results returned, so a location
  whose results are older than the newest `limit` is missing, and `limit=0` returns none.
- `results`: up to `limit` raw results (50 by default, at most 500), newest first, with the status
  code, response time, error class and detail, attempts and location.

The outage log lists up to 50 outages (at most 500), newest first, with `startedAt`, `endedAt`
(`null` while open), `cause`, the locations that confirmed it, and what was sent when it went down
and when it came back.

## Limits

- A window over 30 days is refused with `400` "At most 30 days at once; results are kept that
  long". Raw results are kept 30 days; outages as long as the check exists.

## Related

- [How uptime checks work](how-uptime-checks-work.md)
- `monitoring.checks.results` · `dcloud monitoring checks results`
