Managing containers: ps, logs, exec, stop

Day-to-day work with containers is mostly these commands.

Once containers are running you need to see and control them.

docker ps lists running containers; docker ps -a includes stopped ones. Each has an ID (any unique prefix works) and a name.
docker logs api prints what the main process wrote to stdout and stderr. -f follows, --tail 100 limits, --since 10m filters by time. This is why containerized apps should log to stdout rather than to files.
docker exec -it api sh runs an additional process in a running container, here a shell, so you can poke around. Use bash if the image has it. exec only works while the container is running.
docker stop api sends SIGTERM to the main process, waits (10 seconds by default, -t to change), then sends SIGKILL. Apps should handle SIGTERM to shut down gracefully.
docker kill api sends SIGKILL immediately.
docker start api / docker restart api start a stopped container again, keeping its writable layer.
docker rm api deletes a stopped container; docker rm -f api stops and deletes it.
docker inspect api dumps full JSON details: state, exit code, IP address, mounts, env. Use --format to pick fields.
docker stats shows live CPU and memory per container.

Containers that stop keep existing until removed, so docker ps -a grows over time. docker container prune removes all stopped containers in one go.

Example

docker run -d --name api -p 8000:8000 python:3.13 python -m http.server
docker ps                         # running containers
docker logs -f --tail 20 api      # follow output (Ctrl+C stops following)
docker exec -it api bash          # shell inside; exit to leave
docker exec api ls /              # one-off command
docker inspect api --format '{{.State.Status}} {{.NetworkSettings.IPAddress}}'
docker stop api                   # SIGTERM, then SIGKILL after 10 s
docker ps -a --filter name=api    # Exited (...)
docker start api && docker restart api
docker rm -f api
docker container prune            # remove all stopped containers

Remember

ps lists; logs shows stdout/stderr; exec runs extra processes inside
stop = SIGTERM, then SIGKILL after a grace period
Stopped containers persist until docker rm or prune
inspect --format pulls any detail out of the JSON

Watch out for

Expecting docker logs to show files the app writes to /var/log — it only captures stdout and stderr.

Check yourself

What does docker stop do first?
Which command opens a shell inside an already running container named db?

Official docs: docs.docker.com/reference/cli/docker/container/exec/

Example

Remember

Watch out for

Check yourself

Builds on

Unlocks