Workload Measurement Tool - Web application

This is the webserver application that will present the information stored in the WMT database to the end user.

It is a node.js application using the express web framework.

Prerequisites

• Node v20 (managed using nvm)
• Docker

On OSX (using homebrew):

• brew install nvm

• Follow the instructions in the brew installer output

• Go to Docker get started to install Docker

Install Node version 20

• nvm install 20

Run application locally against Dev environment

• It is possible to get the web application running locally to:
  • authenticate against the dev environment
  • make networks calls to the dev environment APIs
  • integrate with the dev databases
• The below sections describe how to achieve all of the above...

Create a .env file

• Duplicate the .env.template file and rename the duplicated file to .env
• You will notice that in your new .env file you have the properties that the application requires
• You will also notice that the secret values (that are intentionally left out of values.dev.yml for deployments) are also intentionally not included .env.template
• The placeholder values of the properties in .env.template will need to be swapped out for the real secrets
• these secrets are stored in Kubenetes and can be accessed in the hmpps-workload-dev namespace in the following secrets:
  • hmpps-workload
  • rds-history-instance-output
  • rds-live-instance-output
• here is a guide for connecting to the Kubernetes Cluster to access the namespace/secrets

Connect to DEV DBs

• If you have just done the previous section, you may have noticed that in your resulting .env file you have secrets for two databases
• To connect to the DEV databases, we will need to port forward to both of them
• Here is the wiki on how to port forward to dbs in general Access the DEV RDS Database
• This wiki explains how to do a single port forward to a single DB, in our case we will need to do it twice (once for each database):
• So re this command in the wiki:

kubectl \
  -n [your namespace] \
  run [your pod name] \
  --image=ministryofjustice/port-forward \
  --port=5432 \
  --env="REMOTE_HOST=[your database hostname]" \
  --env="LOCAL_PORT=5432" \
  --env="REMOTE_PORT=5432"

• We need to run this twice (once for each database) and make the [your pod name] unique for each one
• I normally set them to:
  • port-forward-pod for the live database
  • port-forward-history-pod for the history database
• Then when you follow the Forward local traffic to the port-forward-pod section in the wiki you will need to do that twice also
• And IMPORTANTLY: for the port-forward-history-pod forward from local port 5433. So this:

kubectl \
-n [your namespace] \
port-forward \
port-forward-pod 5432:5432

kubectl \
-n [your namespace] \
port-forward \
port-forward-history-pod 5433:5432

• Notice this in your .env file:

WMT_HISTORY_DB_PORT=5433

• So, in the locally deployed application, based on this prop we will forward traffic for the history DB o 5433 (hence the second port forward being setup on that port)

Run localstack, manage-users-api and redis docker services locally

• we can run redis locally as a docker container so that we do not need to integrate with the dev environment's redis datasource
• we can run localstack locally to simulate the AWS resources we need locally
• run this command from this repo's root directory:

docker-compose up -d redis localstack hmpps-manage-users-api

• with the above command you will have noticed that we are specifically running the redis localstack containers only. If we were to run the usual docker-compose up -d command then we would run the auth service and other downstream services as local containers which is not our intention here
• just running these containers is good because we limit the amount of mocking and interact with dev services as much as possible
• jump into localstack container and run shell script (to create the localstack AWS infra)

docker exec -it wmt-web-localstack bash
cd /docker-entrypoint-initaws.d
./setup-s3.sh
exit

Start the web application

• In Intellij create a new Configuration by clicking Edit Configuration in the dropdown next to the Run and Debug buttons
• Click the Add new configuration button (the button with the + sign in the top left)
• In the resulting list find npm in the list and click on it
• In the resulting form take the defaults and set the following values:
  • Command = run
  • Scripts = start-local
• Hit Apply button
• Hit OK button
• Now you should see a new start-local run configuration in the dropdown next to the Run and Debug buttons
• You should now be in a position to run (as well as debug) the application and it will use the properties set in your new .env file to fire the networks calls at the dev environment
• Navigate to http://localhost:3000 to see the running application.

Testing

Unit Tests

• To run Unit Tests run the following command:

npm test

• if you want to generate an html report so that you can view any failures vid=sually run this command:

npm run test-generate-report

Integration Tests

To run Integration Tests

• run docker containers

docker-compose up -d

• jump into localstack container and run shell script (to create the localstack AWS infra)

docker exec -it wmt-web-localstack bash
cd /docker-entrypoint-initaws.d
./setup-s3.sh
exit

• run the following command:

npm run integration-test

• if you want to generate an html report so that you can view any failures visually run this command instead:

npm run integration-test-generate-report

E2E Tests

E2E Tests are run using Selenium and webdriver

Run tests using the following commands:

• run docker containers

docker-compose up -d

• jump into localstack container and run shell script (to create the localstack AWS infra)

docker exec -it wmt-web-localstack bash
cd /docker-entrypoint-initaws.d
./setup-s3.sh
exit

• run the following command to start the application and run the e2e tests:

npm run start-dev
npm run test-e2e