Debug containerized apps

The Container Tools extension provides more support for debugging applications within containers, such as scaffoldinglaunch.json configurations for attaching a debugger to applications running within a container.

The Container Tools extension provides adocker debug configuration provider that manages how VS Code will launch an application and/or attach a debugger to the application in a running container. This provider is configured via entries withinlaunch.json, with configuration being specific to each application platform supported by the provider.

The Container Tools extension currently supports debuggingNode.js,Python, and.NET applications within containers.

Requirements

Scaffolding or pasting a launch configuration intolaunch.json isnot sufficient to build and debug a container. To successfully run a container launch configuration, you must have:

• A Dockerfile.
• docker-build anddocker-run tasks intasks.json.
• A launch configuration that invokes these tasks.

We recommend using theContainers: Add Docker Files to Workspace... command to create these items, if none of these assets already exist. If you already have a functional Dockerfile, we recommend using theContainers: Initialize for container debugging command to scaffold a launch configuration and container-related tasks.

Node.js

More information about debugging Node.js applications within containers can be found atDebug Node.js within a container.

Examplelaunch.json configuration for debugging a Node.js application:

{  "configurations": [    {      "name":"Containers: Node.js Launch",      "type":"docker",      "request":"launch",      "preLaunchTask":"docker-run: debug",      "platform":"node"    }  ]}

Python

More information about debugging Python applications within containers can be found atDebug Python within a container.

Examplelaunch.json configuration for debugging a Python application:

{  "configurations": [    {      "name":"Containers: Python - Django",      "type":"docker",      "request":"launch",      "preLaunchTask":"docker-run: debug",      "python": {        "pathMappings": [          {            "localRoot":"${workspaceFolder}",            "remoteRoot":"/app"          }        ],        "projectType":"django"      }    }  ]}

.NET

You can choose between two ways of building and debugging your project within containers:

• With .NET SDK: If you are familiar withMSBuild or want to containerize your project without a Dockerfile, this is the recommended choice.

  Note: This option is only available for .NET SDK 7 and above and uses thedotnet publish command to build the image.

• With a Dockerfile: If you prefer customizing your project with aDockerfile, choose this option.

For more details about these two options, refer toDebug .NET within containers.

Examplelaunch.json configuration for debugging a .NET application usingDockerfile:

{  "version":"0.2.0",  "configurations": [    {      "name":"Containers: .NET Launch",      "type":"docker",      "request":"launch",      "preLaunchTask":"Run Container",      "netCore": {        "appProject":"${workspaceFolder}/project.csproj"      }    }  ]}

Configuration reference

dockerServerReadyAction object properties

node object properties

These properties are the same as those described in theVS Code documentation for attaching a debugger to Node.js applications. All properties passed in thenode object will be passed on to the Node.js debug adaptor, even if not specifically listed below.

python object properties

netCore object properties

Properties passed in thenetCore object are generally passed on to the .NET debug adaptor, even if not specifically listed below. The complete list of debugger properties is in theOmniSharp VS Code extension documentation.

Next steps

Read on to learn more about:

• Debugging Node.js within containers
• Debugging Python within containers
• Debugging .NET within containers
• Debugging with Docker Compose
• Troubleshooting