How to Stop and Remove Docker Containers

If you don’t already have Docker installed, you can grab it from the official website: Docker Download.

To follow along with this article, once installed, pull the Ubuntu image from the Docker hub so we can use the image with the containers that we will manipulate using the command:

For other Docker commands, check out our Docker cheat sheet.

To run a simple container based on the Ubuntu image we just pulled, run the command below:

After running the container, you’ll be inside the container’s shell (in this example, a Bash shell). You can interact with it just like a regular Linux system.

When you’re done, you can exit the container by typing exit.

Let’s take a look at the command options for docker run.

This command is used to start a container, and the syntax looks like this:

In our example, the IMAGE we used was ubuntu , the -it OPTIONS we specified make the container interactive and allocate a terminal. The COMMAND we supplied was the command to run inside the terminal of the running container.

You can find out more about the options available to the Docker run command over on the official Docker reference pages.

To see a list of running containers, you can use the docker ps command. This command provides information about running containers, including their names, IDs, and status.

Learn more: How to List Docker Containers

To stop a running Docker container, use the docker stop command and provide either the container name or the container ID. You can find these from the information supplied in the docker ps command.

When you run docker stop, the main process inside the container will receive SIGTERM signal, giving the container a chance to shut down gracefully. If the container doesn’t stop within 10 seconds (default timeout), Docker sends SIGKILL. 

The first signal can be changed with the STOPSIGNAL instruction in the container’s Dockerfile or the --stop-signal option to docker run.

Here’s the full syntax of the docker stop command:

You can also use the option -s to change the signal sent to the container, and -t to supply the number of seconds to wait before killing the container. 

For example, as shown above, the container ID of our running Ubuntu container is ccfac1f88d1b:

Run docker ps again to check that the container is stopped.

To forcefully stop a Docker container immediately without a timeout, you can use the docker kill command. The main process inside the container is sent the SIGKILL signal (default), or the signal specified with the --signal option, which can terminate the process abruptly.

When you run docker kill, you will notice the container is terminated much quicker than when running docker stop. This is because the container is not gracefully terminated, and it does not have a chance to perform any cleanup tasks.

The syntax is similar to the docker stop command:

Be aware that this can potentially lead to data corruption or unexpected behavior in the container.

Because of this,docker kill should be used after command with caution, usually after trying docker stop first to allow the container to shut down cleanly and only resort to docker kill when the container is unresponsive or cannot be stopped by normal means.

To stop all running Docker containers with a single command rather than individually, you can use a for loop in the shell:

This command works by first listing the container IDs of all running containers (docker ps -q), then passing those IDs to docker stop, which stops each one. If no containers are running, the command has no effect.

The docker stop command supports two additional options that control how containers are stopped:

• -t, --time: Specifies the timeout (in seconds) before forcibly killing the container. By default, Docker waits 10 seconds after sending SIGTERM before issuing SIGKILL. Example: docker stop -t 5 my_container waits 5 seconds before force-killing.
• -s, --signal: Defines which signal to send instead of the default SIGTERM. This allows customized shutdown behavior if the container handles different signals. Example: docker stop -s SIGINT my_container.

The docker ps command shows only currently running containers by default. If your container is not listed there, it’s not running. However, it might still exist in a stopped state. To see both running and stopped containers, use:

Look for your container in the list. If it appears with a STATUS like Exited (0) or Exited (1) followed by a timestamp, it has stopped.

To remove Docker containers after stopping them, use the docker rm command followed by the container ID or name. 

The container must be stopped before removal. If you attempt to remove a running container, Docker will return an error unless you include the -f (force) flag, which forcibly stops and deletes it.

 You can also run it like this:

The additional -a option in the docker ps -a -q command lists all containers, including stopped ones. After running the command, a list of the container IDs that have been removed will be shown on the console:

Similar to the docker kill command, if required, you can use the docker rm command with the --kill option to force the removal of a running container, using SIGKILL.

Removing all stopped containers

To remove all stopped Docker containers, you can also use:

This command deletes all containers with a Exited or Created status. The -f flag (or --force) skips the confirmation prompt.

Alternatively, to be more explicit and see the container IDs being removed:

Both commands only remove stopped containers, not running ones.

In Docker Compose, you can stop and remove containers associated with your Compose project using the docker-compose down command. You can execute the docker-compose down command from the directory where your docker-compose.yml file is located.

Below we have a docker-compose.yaml file to create a simple Ubuntu container:

Let’s run docker compose up to create the container:

And then run docker compose down to remove it:

This command not only stops the containers but also removes:

• Containers for services defined in the Compose file
• Networks defined in the networks section of the Compose file
• The default network, if one is used

Networks and volumes defined as external are never removed.

To remove named volumes declared in the “volumes” section of the Compose file and anonymous volumes attached to containers, you can use the -v option.

In Docker Swarm, containers are managed by services, so you cannot stop individual containers directly. Depending on your intent, you can use docker service or docker node commands to scale down or remove the service that manages them.

1. To reduce the number of running containers to zero: docker service scale <service_name>=0
2. To completely stop and remove all containers associated with the service: docker service rm <service_name>
3. If you want to stop all tasks on a node, to prevent the node from receiving tasks and stop running ones: docker node update --availability drain <node_name>

Avoid stopping containers manually via docker stop, as Swarm will restart them to maintain the declared state.

Both commands will terminate containers. The key difference is that docker stop allows for a graceful shutdown by sending a SIGTERM signal, while docker kill forces immediate termination with a SIGKILL signal.

You should generally prefer docker container stop when possible to ensure a clean shutdown of your containers, but docker container kill can be useful when a container is not responding or needs to be stopped forcefully. This distinction mirrors typical Linux signal handling and is important for maintaining data integrity and process stability.

Manipulating containers on the command line and knowing how to start, stop, kill, and remove containers on demand with the available options is a great way to start learning and experimenting with Docker. Knowing the differences between the stop and kill commands, as explained above, is valuable, especially to avoid any issues when running in production.