Docker restart policies

Topic: Containers core

Set container restart policy with --restart (no, on-failure, always, unless-stopped). Use always or unless-stopped for long-running services so they come back after reboot or crash. Use this when you want containers to restart automatically.

Intent: How-to

Quick answer

  • docker run --restart=always image. Options: no (default), on-failure[:max], always, unless-stopped. always and unless-stopped start on daemon start.
  • unless-stopped: restart except if user stopped the container. on-failure: only restart on non-zero exit; optional max count.
  • In compose: restart: always (or on-failure, unless-stopped). Ensures services survive host reboot when Docker starts.

Prerequisites

Steps

  1. Choose policy

    Use always or unless-stopped for services. Use on-failure for batch jobs with retry limit.

  2. Run with policy

    docker run -d --restart=unless-stopped myimage. Or in compose: restart: unless-stopped.

  3. Verify

    Kill container or reboot; confirm container restarts after Docker starts.

Summary

Set —restart=always or unless-stopped for services so they restart on failure or host reboot.

Prerequisites

Steps

Step 1: Choose policy

Pick always, unless-stopped, or on-failure with optional max.

Step 2: Run with policy

Use —restart in docker run or restart in compose.

Step 3: Verify

Kill or reboot; confirm restart behavior.

Verification

  • Container restarts as expected after crash or reboot.

Troubleshooting

No restart — Check policy and Docker daemon running. Restart loop — Fix container entrypoint or fix cause of exit.

Next steps

Continue to