Now that you have created an image, you can run a container.
Run the container. This installs the latest version of the agent, configures it, and runs the agent. It targets the Default pool of a specified Azure DevOps or Azure DevOps Server instance of your choice:
```shell
docker run -e VSTS_ACCOUNT=<Azure DevOps instance> -e VSTS_TOKEN=<PAT token> -e VSTS_AGENT=mydockeragent cgot/linux-dockeragent:latest
```
Optionally, you can control the pool and agent work directory by using additional environment variables.
| Environment variable | Description |
|---|---|
| VSTS_ACCOUNT | The URL of the Azure DevOps or Azure DevOps Server instance. |
| VSTS_TOKEN | Personal Access Token (PAT) with Agent Pools (read, manage) scope, created by a user who has permission |
| VSTS_AGENT | Agent name (default value: the container hostname). |
| VSTS_POOL | Agent pool name (default value: Default). |
| VSTS_WORK | Work directory (default value: _work). |
Find all details here: https://github.com/MicrosoftDocs/azure-devops-docs/blob/master/docs/pipelines/agents/docker.md
Ubuntu 22.04.
- git : 2.44.0
- openjdk : 18
- dotnet : 6.0, 8.0
- node : 21.6.2
- docker : 25.0.3
- docker-compose : 2.24.6
- yarn
- lerna