---
url: /howto/installing-teamscale/installing-with-docker.md
description: >-
  This guide shows you how to setup and run Teamscale with Docker and Docker
  Compose.
---

# How to install Teamscale with Docker using Docker Compose

We recommend running Teamscale with Docker using Docker Compose.
The images are available [on Docker Hub](https://hub.docker.com/r/cqse/teamscale/tags).
Please note that we do not provide a *latest* tag.

\[\[toc]]

## Prerequisites

To run Teamscale with Docker, you have to install [Docker](https://www.docker.com) and the [Docker Compose Plugin](https://docs.docker.com/compose/install/).

:::warning Minimal Version
Teamscale requires at least version `20.10.13` of Docker.
Run `docker version` to check the installed version.
:::

On Linux, please ensure that `docker` is on your `PATH` and can be run from your command line.
**You must run all `docker` commands as `root`.**
Either use a `root` shell or prefix the commands with `sudo`.

Run `docker version` and verify that no errors are printed on the command line.

:::tip Use WSL 2 under Windows for Better Performance
When using Docker on Windows, running Docker from the Windows CLI has [worse performance](https://www.docker.com/blog/docker-desktop-wsl-2-best-practices/) than running it from within the Windows Subsystem for Linux 2.
Within WSL 2, mounting folders from the Windows host's file system, e.g. `/mnt/c/`, also has worse performance than mounting a WSL 2 path.
Thus, we recommend that you run Docker from within your WSL 2 Linux distribution of choice and store all Teamscale-related files inside the WSL Linux.
:::

## Docker Compose Setup

You can use one of two reference Docker Compose setups, depending on your use case:

### Evaluation Setup

If you want to run Teamscale on your local machine, the fastest path is to use the [evaluation setup](https://github.com/cqse/teamscale-docker-evaluation).
It runs a single Teamscale instance and is suitable for trying out Teamscale locally.
It is not recommended to deploy this setup to a server or use it for a production instance.

### Production Setup

If you are deploying Teamscale for production use, use the [production setup](https://github.com/cqse/teamscale-docker-deployment).
It runs NGINX as a reverse proxy and two Teamscale instances.
The second instance acts as a [shadow instance](../../../howto/updating-teamscale/shadow-instance/#feature-version-updates-with-a-shadow-instance) during feature-version updates to avoid downtime.
