Docker volumes explained: Named volumes, bind mounts and how to manage them

In Docker, data within a container is not persistent. When a container is restarted, stopped, or removed, the data inside it is lost — the next container starts from a clean state. For databases, stateful applications, and development workflows, this is a problem. Docker volumes solve it by storing data outside the container lifecycle, on the host filesystem or in a Docker-managed location.

This guide covers the three volume types, how to create and manage them, and when to use each.

Named volumes

Named volumes are managed entirely by Docker and stored under /var/lib/docker/volumes. They are the recommended approach for persistent application data because they are portable, easy to back up, and work across container restarts and removals.

Create a named volume and mount it to a container:

1docker run -it -v demo:/usr/share busybox

Or using the recommended --mount syntax, which is more explicit and easier to read:

1docker run -it --mount type=volume,source=demo,target=/usr/share busybox

List all volumes on your system:

1docker volume ls

Expected output:

1DRIVER    VOLUME NAME
2local     demo

The volume persists after the container exits. Create a new container and mount the same volume — the data will still be there.

Bind mounts

A bind mount links a specific path on your host machine directly into the container. You control exactly where the data lives. This is the standard pattern for development workflows where you want live code or configuration changes reflected in the container immediately.

1docker run -it -v /home:/temp busybox
2

Or with --mount:

1docker run -it --mount type=bind,source=/home,target=/temp busybox

Navigate to the mounted directory inside the container and create a file:

1cd /temp
2touch test
3exit

The file is now visible at /home/test on your host machine. Changes made on the host are reflected inside the container and vice versa.

Anonymous volumes

If you do not specify a name or host path, Docker creates an anonymous volume with a randomly generated hash as its name. The data is still persisted but you cannot easily reference the volume again by name.

1docker run -it -v /home busybox

Anonymous volumes are useful for temporary scratch space where you do not need to access the data after the container exits.

Managing volumes

Inspect a volume

See details including the mount point on the host, creation date, and driver:

1docker volume inspect demo

Expected output:

1[
2    {
3        "CreatedAt": "2024-01-15T10:23:45Z",
4        "Driver": "local",
5        "Mountpoint": "/var/lib/docker/volumes/demo/_data",
6        "Name": "demo",
7        "Scope": "local"
8    }
9]

Remove a volume

1docker volume rm demo

This permanently deletes the volume and all data inside it. The volume must not be in use by any container.

Remove all unused volumes

1docker volume prune

This removes all volumes not currently attached to a container. Add -f to skip the confirmation prompt. Run docker volume ls first to confirm what will be removed — this action is irreversible.

When to use which