Configure a Docker Container Check

Monitor Docker container health, status, and restarts

Written By Erdinc Akay

Last updated 25 days ago

Configure a Docker Container Check

A Docker Container Check monitors running Docker containers on a device. It warns when containers unexpectedly stop, become unhealthy, or restart excessively.

Prerequisites

  • You need the Monitoring Check Management permission.
  • The target device must have a connected agent with Docker installed and the Docker daemon running.
  • The agent needs access to the Docker socket.

Steps

  1. Go to Devices and open the device.
  2. Click the Checks tab → Add Check → Docker Container.
  3. Decide which containers to monitor using the Monitor all containers toggle (on by default). When the toggle is on, the check monitors every container on the device and the Excluded containers field appears so you can exclude specific container names. When the toggle is off, the Container names field appears and you list the specific containers to monitor. Both fields accept container names only (not IDs) and support wildcards: * matches any characters, ? matches a single character (for example, web-*).
  4. Configure the alert behavior:
FieldDefaultDescription
Warn on stopped containersYesWarns if a container is not running
Critical on stopped containersNoCritical if a container is not running
Warn on unhealthy containersYesWarns if a container's health check fails
Max restart count5Warns if a container's restart count exceeds this value
  1. Click Add Check.

What the check reports

The check reports the status of each monitored container (running, stopped, restarting), the health status (for example healthy, unhealthy, starting, or none if the container does not declare a health check), and the restart count.

Tips

  • Set Critical on stopped containers to Yes for production containers that must always be running.
  • The Max restart count threshold catches containers in crash loops. A container that restarts 5 times likely has a configuration or dependency problem.
  • If you run short-lived containers (e.g. cron jobs), monitor specific long-lived containers by name instead of using "all containers".
  • If the check shows a warning with no containers listed, the agent either could not reach the Docker daemon or no container matched your name filters — check that Docker is running and review your container names and exclusions.

After the first check interval, the result appears in the device's Checks tab.

See also: Docker Container Check Reference