Uptime monitors
Use uptime monitors to check responses from public URLs and TCP ports on a schedule. Web performs the checks, so the target does not need a Collector. Docker service response checks come from the Docker Collector and appear in that Docker 환경 (Docker environment); you do not need to add them again as uptime monitors.
Add a monitor
On the 업타임 (Uptime) screen, click 추가 (Add), fill in the fields below, and click 추가 (Add).
| Field | Description |
|---|---|
| 이름 (Name) | Shown on screen and in alerts |
| 체크 방식 (Check type) | HTTP or TCP. It cannot be changed later |
| URL | For HTTP, the full address to check (e.g. https://api.example.com/health) |
| 호스트 (Host), 포트 (Port) | For TCP, the host and port to connect to (default 443) |
| 주기 (초) (Interval, seconds) | How often to check. Default 30, minimum 5 |
| 기대 상태 코드 (Expected status code) | An HTTP response passes when its code matches this value. Default 200 |
| 타임아웃 (초) (Timeout, seconds) | A check fails if there is no response in this time. Default 5 seconds |
An HTTP check passes when its response code matches the configured expected status code; a TCP check passes when the connection succeeds.
Internal addresses can't be checked
For security, localhost, private networks (10.0.0.0/8, 172.16.0.0/12, 192.168.0.0/16), and link-local addresses (169.254.0.0/16, including cloud metadata) are rejected. Monitor internal services with the Docker Collector instead.
Status and management
The 업타임 (Uptime) screen shows only the uptime monitors you add there. Each target shows its HTTP or TCP method and check address. View Docker Collector response checks and container state in Docker 환경 (Docker environments).
Counts for 장애 (down), 정상 (healthy), 확인 대기 (pending), and 일시정지 (paused) appear at the top. Click a count to filter by that state. Compare monitor names, check addresses, 24-hour uptime, and average response times in the list.
The list and detail screens refresh automatically every 30 seconds by default. Check the last successful refresh time, turn automatic refresh off, or click 새로고침 (Refresh) to refresh manually. Automatic requests pause in hidden browser tabs and resume with a refresh when the tab becomes visible. If a refresh fails, the screen keeps previously loaded data and labels it as prior information.
Both the grid and cards show 최근 체크 (Recent checks): up to 10 check results and the latest outcome. Hover over the check strip, or focus it and use the arrow keys, to inspect an exact timestamp and the full message. An ongoing outage shows its duration below the status; hover over or focus the duration to inspect its start time. Click the target name for response trends and 90-day history.
Use 일시정지 (Pause) to stop checks and 재개 (Resume) to restart them. Use 수정 (Edit) or 삭제 (Delete) on the detail screen to manage the monitor.
The detail screen's 응답 시간 (Response time) chart defaults to the last 6 hours. Select 1H, 6H, or 24H for averages in 2-, 5-, or 20-minute buckets. Response times average successful checks only. The summary shows the selected period's average, total checks, and failed checks.
Hover over the chart, or focus it and use the left and right arrow keys, to inspect a bucket's average and check/failure counts. Confirmed outages have a shaded background. Missing collection data breaks the line and the tooltip shows 수집 데이터 없음 (No collection data). Missing data is not represented as an outage or a 0ms response.
Getting outage alerts
An uptime monitor that goes down sends nothing unless an alert rule matches it. A rule with the 헬스체크 (Health check) category and an empty 대상 (Target) applies to every uptime monitor and Docker service.