Container Commands

A container is a running instance of an image. This page covers the high-frequency operations for starting, entering, logging, and stopping containers. docker run is the single most-used command.

Start a container (docker run)

docker run is the core command. Here is a table of the most common flags:

Flag	Purpose	Example
-d	Run detached (background)	docker run -d nginx
-p	Port mapping host:container	-p 8080:80
-v	Volume mount host:container	-v $PWD/data:/data
-e	Inject environment variable	-e ROS_DOMAIN_ID=1
--name	Name the container (easier to manage)	--name ros-node
--restart	Restart policy on exit	--restart unless-stopped
--rm	Auto-remove container on exit	docker run --rm ubuntu

A ROS 2 node example, running in the background with port mapping and a restart policy:

Linux

docker run -d \
  --name ros-node \
  -p 11311:11311 \
  -e ROS_DOMAIN_ID=1 \
  --restart unless-stopped \
  ros:humble \
  ros2 run demo_nodes_cpp talker

💡 For interactive image debugging, add -it and override the default command to drop into a shell: docker run -it --rm ubuntu bash.

List running containers

Running containers only:

Linux

docker ps

All containers, including stopped ones:

Linux

docker ps -a

Enter a running container

Open an interactive terminal (the most common troubleshooting step):

Linux

docker exec -it ros-node bash

Run a one-off command without an interactive shell:

Linux

docker exec ros-node ros2 node list

View logs

Follow container logs in real time (-f behaves like tail -f):

Linux

docker logs -f ros-node

Stop and remove

Graceful stop (sends SIGTERM, waits up to 10s by default):

Linux

docker stop ros-node

Force stop (immediate SIGKILL):

Linux

docker kill ros-node

Remove a stopped container:

Linux

docker rm ros-node

tip

The troubleshooting trio

When a container misbehaves, locate the issue in this order:

Linux

docker ps -a          # 1. check status and exit code
docker logs ros-node  # 2. check startup logs
docker exec -it ros-node bash  # 3. enter the container to investigate

Resource usage

Live CPU, memory, and network usage per container:

Linux

docker stats