This repository contains the source code for Heimdall's Backend, Frontend (AKA Heimdall Lite), OHDF Converters, and InSpecJS.

• Heimdall
  • Contents
  • Demos
    • Video
    • Hosted
      • These demos are only intended to show the functionality of Heimdall, please do not upload any sensitive data to them.
      • Released Previews
      • Current Development Master Branch Preview
  • Heimdall (Lite) vs Heimdall with Backend (Server)
    • Heimdall-Lite
    • Heimdall with Backend (Server)
    • Features
    • Use Cases
  • Getting Started / Installation
    • Heimdall Lite
      • Running via npm
      • Running via Docker
    • Heimdall Server - Docker
      • Setup Docker Container (Clean Install)
      • Updating Docker Container
      • Stopping the Container
      • Helm Chart
      • Running via Cloud.gov
  • External Data Sources (Interfaces)
    • AWS S3
    • Splunk
    • Tenable.SC
  • API Usage
  • For Developers
    • How to Install
    • Debugging Heimdall Server
    • Developing Heimdall Lite Standalone
    • Lint and fix files
    • Compile and minify the frontend and backend for production
    • Run tests
      • Run Cypress End to End Tests
    • Creating a Release
  • Versioning and State of Development
  • Contributing, Issues and Support
    • Contributing
    • Issues and Support

Heimdall Lite | Heimdall Server   

Heimdall Lite   

Heimdall Server   

There are two ways to deploy the MITRE Heimdall application - Heimdall-Lite and the full Heimdall with Backend Server. Both share the same frontend but have been produced to meet different needs and use-cases.

As a single-page javascript app - you can run Heimdall-Lite from any web-server, a secured S3 bucket or directly via GitHub Pages (as it is here). Heimdall-Lite gives you the ability to easily review and produce reports about your InSpec run, filter the results for easy review and hot-wash, print out reports, and much more.

Heimdall with Backend, or Heimdall Server runs the same front end as Heimdall-Lite, but is supported with a backend database to store persistent data overtime.

Heimdall Lite is published to npmjs.org and is available here.

To run Heimdall Lite locally, just use the npm built-in utility npx:

npx @mitre/heimdall-lite

If you use this tool often and want to have it installed locally, use the following command:

npm install -g @mitre/heimdall-lite

Then, any subsequent npx @mitre/heimdall-lite will use the local version and load much more quickly.

It is also possible to run Heimdall-Lite using Docker, using the following command:

docker run -d -p 8080:80 mitre/heimdall-lite:release-latest

You can then access Heimdall-Lite at http://localhost:8080.

If you would prefer to run the bleeding edge version of Heimdall-Lite, replace mitre/heimdall-lite:release-latest with mitre/heimdall-lite:latest.

Given that Heimdall requires at least a database service, we use Docker and Docker Compose to provide a simple deployment experience. This process will also deploy an NGINX webserver in front of Heimdall to handle TLS.

Heimdall's frontend container image is distributed on DockerHub, and on Iron Bank.

1. Install Docker

2. Download and extract the most recent Heimdall release from our releases page. Alternatively, you can clone this repository and navigate to the heimdall2 folder.

3. Navigate to the base folder where docker-compose.yml is located

4. By default Heimdall will generate self-signed certificates that will last for 7 days. For production use, place your certificate files in ./nginx/certs/ with the names ssl_certificate.crt and ssl_certificate_key.key respectively. For development use, you can use the default generated certificates which means you do not need to put any certificate files in the ./nginx/certs/ folder.

NGINX Configuration Note: You can configure NGINX settings by changing values in the nginx/conf/default.conf file.

5. Run the following commands in a terminal window from the Heimdall source directory. For more information on the .env file, visit Environment Variables Configuration.

   # For Linux or Mac
   ./setup-docker-env.sh
   
   # For Windows
   ./setup-docker-env.bat

6. Heimdall might need certificates to access the open internet or internal resources (ex. an LDAP server). Please convert any certificates into PEM files and place them in ./certs/ where they will be automatically ingested. Alternatively, you can place a shell script that will retrieve those certs in that directory, and modify the command attribute underneath the certs service in the docker-compose.yml to run that script.

# Below is an example of what may be in the ./certs directory, including any scripts or certificates.
# ./certs/
# ├── dodcerts.sh
# └── my_certificates.pem

# For the given example, the ./docker-compose.yml should look like the following:
services:
  ...
  certs:
    ...
    command: sh -c "sh /etc/pki/ca-trust/source/anchors/dodcerts.sh && update-ca-trust && tail -f /dev/null"
    # NOTE: The `command` attribute only needs to know about scripts not any particular certificates.
    ...
  ...

To make the docker-compose.yml aware of additional scripts, add sh /etc/pki/ca-trust/source/anchors/NAME_OF_SCRIPT.sh && to the beginning of the section in quotes. NOTE: The script should make sure to place the certs within /etc/pki/ca-trust/source/anchors/ since it will be run from the container, not the host.

7. Start Heimdall Server by running this command:

docker-compose up

8. Navigate to https://127.0.0.1. You should see the application's login page. (Note that if you used the option to generate your own self-signed certs, you will get warnings about them from your browser.)

Starting with version 2.5.0, Heimdall on Docker uses SSL by default. Place your certificate files in ./nginx/certs/ with the names ssl_certificate.crt and ssl_certificate_key.key respectively.

A new version of the docker container can be retrieved by running:

docker-compose pull
docker-compose up -d

This will fetch the latest version of the container, redeploy if a newer version exists, and then apply any database migrations if applicable. No data should be lost by this operation.

From the source directory you started from run:

docker-compose down

https://github.com/mitre/heimdall2-helm

Cloud.gov is a FEDRAMP moderate Platform-as-a-Service (PaaS). This repository includes a sample manifest.yml.example file ready to be pushed and run the latest version of Heimdall2 as a container. Make a copy of the example file and update the key values as appropriate. $ cp manifest.yml.example manifest.yml

1. Setup a cloud.gov account - https://cloud.gov/docs/getting-started/accounts/

2. Install the cf-cli - https://cloud.gov/docs/getting-started/setup/

3. Run the following commands in a terminal window from the Heimdall source directory.

$ cd ~/Documents/Github/Heimdall2
$ cf login -a api.fr.cloud.gov  --sso 
# Follow the link to copy the Temporary Authentication Code when prompted

4. Setup a demo application space

$ cf target -o sandbox-rename create-space heimdall2-rename

5. Create a PostgreSQL database

# Update manifest.yml file to rename application and database key name
$ cf marketplace
$ cf create-service aws-rds medium-psql heimdall2-rename
$ cf create-service-key heimdall2-db-rename heimdall2-db-test-key
$ cf push

You should be returned the URL for your new test instance to navigate to.

Note: This is only for demonstration purposes, in order to run a production level federal/FISMA system. You will need to contact the cloud.gov program and consult your organization's security team (for risk assessment and an Authority to Operate).

Heimdall currently provides connectivity to the following services for importing and visualizing scans:

• AWS S3
• Splunk
• Tenable.SC

For detail information on how to setup and connect to an AWS S3 bucket see the Heimdall Interface Connection - AWS S3 Wiki

For detail information on how to setup and connect to an Splunk instances (logical or virtual) see the Heimdall Interface Connection - Splunk Wiki

For detail information on how to setup and connect to an Tenable.SC instance see the Heimdall Interface Connection - Tenable.SC Wiki

API usage only works when using Heimdall Enterprise Server (AKA "Server Mode").

Heimdall API documentation is being compiled and it is located in this wiki page. In the meantime here are quick instructions for uploading evaluations to Heimdall Server.

# To use API Keys, ensure you have set the API_KEY_SECRET environment variable. To create a secret run: openssl rand -hex 33
# Create an API key using the Heimdall frontend (within the edit user profile modal) and upload an evaluation with the following command
curl -F "data=@<Path to Evaluation File>" -F "filename=<Filename To Show in Heimdall>" -F "public=true/false" -F "evaluationTags=<tag-name>,<another-tag-name>..." -H "Authorization: Api-Key apikeygoeshere" "http://localhost:3000/evaluations"
# You can upload multiple files at once (up to 100)
curl -F "data=@<Path to first evaluation File>" -F "data=@<Path to second evaluation File>" ... -F "public=true/false" -F "evaluationTags=<tag-name>,<another-tag-name>..." -H "Authorization: Api-Key apikeygoeshere" "http://localhost:3000/evaluations"

If you would like to change Heimdall to your needs, you can use Heimdall's 'Development Mode' to ease the development process. The benefit to using this mode is that it will automatically rebuild itself and use those changes as soon as you make them. Please note that you should not run development mode when deploying Heimdall for general usage.

1. Install system dependencies with your system's package manager. NodeJS is required and can be installed via your system's package manager, or an alternative method if desired. Documented below is the installation via your system's package manager.

   Ubuntu:

   • # grab nodesource for recent version of nodejs
     sudo curl -sL https://deb.nodesource.com/setup_18.x -o /tmp/nodesource_setup.sh
     sudo bash /tmp/nodesource_setup.sh
     
     # use apt to install dependencies
     sudo apt install postgresql nodejs git
     sudo apt install nano                        # recommended installation
     sudo npm install -g yarn

   NOTES

   • The installation scripts setup_XX.x are no longer supported and are not needed anymore, as the installation process is straightforward for any RPM and DEB distro.

   • See the Debian and Ubuntu based distributions nodesource for nodejs supported version and additional installation information

   OSX:

   • brew install postgresql node@18 git      
     brew install nano                        # recommended installation
     sudo npm install -g yarn

   WINDOWS:

   • Install Node.js via MSI Installer

     • Download the node release 18.xx installer (msi) from the nodejs site
     • Open and run (double-click) the .msi file, the installation process begins, follow the installation instructions
     • Node.js offers you options to install tools for native modules, we recommend checking the Automatically install the necessary tools check box.
     • Verify the Node and npm version

     node --version 
     npm --version

   • Install Yarn via MSI Installer

     • Download the Yarn installation file from GitHub
     • Open and run the installation file, follow the installation instructions
     • Run the following command in the PowerShell to verify the installation:

     yarn --version

2. Clone this repository:

   • git clone https://github.com/mitre/heimdall2

3. Run the PostgreSQL server:

   Ubuntu:

   • # Switch to the OS postgres user
     sudo -u postgres -i
     
     # Start the Postgres terminal
     psql postgres
     
     # Create the database user
     CREATE USER <username> with encrypted password '<password>';
     ALTER USER <username> CREATEDB;
     \q
     
     # Switch back to your original OS user
     exit

   OSX:

   • # Start the postgres server corresponding to your installation method
     pg_ctl -D /opt/homebrew/var/postgres start
     # Alternatively, you may find postgres in another location like the following:
     pg_ctl -D /usr/local/var/postgres start
     # Brew method
     brew services start postgresql@13
     
     # Start the Postgres terminal
     psql postgres
     
     # Create the database user
     CREATE USER <username> with encrypted password '<password>';
     ALTER USER <username> CREATEDB;
     \q
     
     # Switch back to your original OS user
     exit

   WINDOWS:

   • Start the postgres server base on the installation method
     • Starting Postgres Server Using net start

       net start postgresql-[x32 or x64]-[version]

     • Starting Postgres Server Using pg_ctl

       pg_ctl -D "C:\[path-to-postgres-installation]\PostgreSQL\[version]\data" start

     • Starting Postgres Server Using Services Manager
       • Press the win key + R to launch the Run window.
       • Type services.msc and hit the OK button to open the Services Manager:
       • Search for Postgresql-[x32 or x64]-[version], select the service, and hit the Start/play button to start
   • Create the database user
     • Recommend using pgAdmin and follow instruction listed here
     • Open a postgres shell terminal (path to postgres executable directory must be set)

       # Start the terminal
       psql -U postgres  
       # Create the database user
       CREATE USER <username> with encrypted password '<password>';
       ALTER USER <username> CREATEDB;
       \q

4. Install project dependencies:

   • cd heimdall2
     yarn install      # you may need to run yarn install --registry https://registry.npmjs.org

5. Edit your apps/backend/.env file using the provided setup-dev-env.sh or setup-dev-env.bat script. Make sure to set a DATABASE_USERNAME and DATABASE_PASSWORD that match what you set for the PostgresDB in step 3.

You can also open the apps/backend/.env file in a text editor and set additional optional configuration values. For more info on configuration values see Environment Variables Configuration.

6. Create the database:

   • # Windows
     yarn backend sequelize-cli-windows db:create
     yarn backend sequelize-cli-windows db:migrate
     yarn backend sequelize-cli-windows db:seed:all
     
     # All other OSs
     yarn backend sequelize-cli db:create
     yarn backend sequelize-cli db:migrate
     yarn backend sequelize-cli db:seed:all

7. Start Heimdall:

   • yarn start:dev

This will start both the frontend and backend in development mode, meaning any changes you make to the source code will take effect immediately. Please note we already have a Visual Studio Code workspace file you can use to organize your workspace.

If you are using Visual Studio Code, it is very simple to debug this application locally. First open up the Visual Studio Code workspace and ensure the Node debugger Auto Attach feature in Visual Studio Code is enabled. Next, open the integrated Visual Studio Code terminal and run:

yarn backend start:debug

Visual Studio Code will then automatically attach a debugger and stop and any breakpoints you place in the application.

If you only want to make changes to the frontend (heimdall-lite) use the following command:

yarn frontend start:dev

To validate and lint your code run:

yarn run lint

yarn build

To test your code to make sure everything still works:

# Run Frontend Vue Tests
yarn frontend test
# Run Backend Nest Tests (see note)
yarn backend test:ci-cov

NOTE: The Backend Nest Tests will remove (BULKDELETE) all entries in the configured PostgreSQL server for the following tables:

• EvaluationTags
• Evaluations
• Users
• GroupEvaluations
• Groups
• GroupUsers

The application includes an End-to-End (E2E) frontend and Backend tests (built using the cypress.io framework). The E2E tests performed is to validate that Heimdall Server is running as intended. In order to run these tests, a running instance of the application is required.

CYPRESS_TESTING=true yarn start:dev
CYPRESS_BASE_URL=http://localhost:8080 yarn test:ui:open

The first command will start an instance of Heimdall Server and exposes additional routes required to allow the tests to run. The second will open the Cypress UI which will run the tests any time code changes are made.

Note: This action requires appropriate privileges on the repository to perform.

The steps to create a release are now on the wiki.

This project uses the Semantic Versioning Policy

Please feel free to look through our issues, make a fork and submit PRs and improvements. We love hearing from our end-users and the community and will be happy to engage with you on suggestions, updates, fixes or new capabilities.

Please feel free to contact us by opening an issue on the issue board, or, at inspec@mitre.org should you have any suggestions, questions or issues. If you have more general questions about the use of our software or other concerns, please contact us at opensource@mitre.org.