Skip to content

Latest commit

 

History

History
121 lines (94 loc) · 5.38 KB

README.md

File metadata and controls

121 lines (94 loc) · 5.38 KB
Raw

EDC Data Dashboard

Please note: This repository does not contain production-grade code and is only intended for demonstration purposes.

EDC Data Dashboard is a dev frontend application for EDC Management API.

Documentation

Developer documentation can be found under docs/developer, where the main concepts and decisions are captured as decision records.

Generate client code for EDC REST APIs

  1. [optional] copy the current version of EDC's management-api.yaml file to openapi/. There is one checked in, so this is not required.
  2. in a shell execute 
    docker run --rm -v "${PWD}:/local" openapitools/openapi-generator-cli generate -i /local/openapi/management-api.yaml -g typescript-angular -o /local/src/modules/edc-dmgmt-client/
    This re-generates the service and model classes.

Please note that some of the client classes were edited manually after generation. When regenerating the classes for the API update be careful especially not to overwrite service constructor methods using the generator!

Running the frontend locally

Should you want to run the frontend on your development machine, you'll have to configure some backend values. Those are stored in app.config.json, and by default contain the following:

{
  "managementApiUrl": "{{managementApiUrl}}",
  "catalogUrl": "{{catalogUrl}}",
  "storageAccount": "{{account}}",
  "storageExplorerLinkTemplate": "storageexplorer://v=1&accountid=/subscriptions/{{subscriptionId}}/resourceGroups/{{resourceGroup}}/providers/Microsoft.Storage/storageAccounts/{{account}}&subscriptionid={{subscriptionId}}&resourcetype=Azure.BlobContainer&resourcename={{container}}",
}

Substitute the values as necessary:

  • apiKey: enter here what your EDC instance expects in th x-api-key header
  • catalogUrl: prepend your connector URL, e.g. http://localhost, assuming your catalog endpoint is exposed at port 8181, which is the default
  • managementApiUrl: prepend your connector URL, e.g. http://localhost, assuming your IDS endpoint is exposed at port 9191
  • storageAccount: insert the name of an Azure Blob Storage account to which the connector has access, otherwise data transfers won't work.

Be extra careful NOT to commit those changes, as they might leak potentially sensitive information!!!

As some extra safety consider running git udpate-index --assume-unchanged src/assets/config/app.config.json before changing this file.

Running a frondend and a connector locally

To test the correct functionality locally you can spin up a local docker compose that will load the data-dashboard service and the connector one. First you need to change the app.config.json this way:

{
  ...
  "managementApiUrl": "http://consumer-connector/management",
  "catalogUrl": "http://consumer-connector/management",
  ...
}

Then you can start the docker compose:

docker compose up

The DataDashboard will be available at http://localhost:8080

Running DataDashboard from the host machine

To have a quicker development cycle, you can also run the DataDashboard from the host machine using npm start, sending request against the connector loaded by docker compose. First you need to change the app.config.json this way:

{
  ...
  "managementApiUrl": "http://localhost:4200/management",
  "catalogUrl": "http://localhost:4200/management",
  ...
}

Then start the local DataDashboard:

npm start

The DataDashboard will be available at http://localhost:4200

Deploy to Azure

Create a resource group and container registry:

export RESOURCE_GROUP=edc-data-dashboard
export ACR_NAME=edcdatadashboard
az group create --resource-group $RESOURCE_GROUP --location westeurope -o none
az acr create --resource-group $RESOURCE_GROUP --name $ACR_NAME --sku Standard --location westeurope --admin-enabled -o none

Dockerize the application and push it to the registry by running:

az acr build --registry $ACR_NAME --image edc-showcase/edc-data-dashboard:latest .

The docker image is now ready to be deployed to Azure Container Instances (ACI). The app.config.json file contains configuration which is fetched by the application at startup. This file can be overridden at deployment time by mounting a secret on assets/config. For each deployment you need to provide the corresponding connector backend URL, the storage account name and the API key using this secret. Deploy to ACI using the following command:

export CONNECTOR_DATA_URL=<CONNECTOR_DATA_URL>
export CONNECTOR_CATALOG_URL=<CONNECTOR_CATALOG_URL>
export STORAGE_ACCOUNT=<STORAGE_ACCOUNT>
export API_KEY=<API_KEY>

# deploy to ACI (when prompted for credentials use the username/password as available in Azure Portal: ACR->Access Keys)
az container create --image ${ACR_NAME}.azurecr.io/edc-showcase/edc-data-dashboard:latest \
--resource-group $RESOURCE_GROUP \
--name edc-data-dashboard \
--secrets "app.config.json"="{\"managementApiUrl\": \"$CONNECTOR_DATA_URL\", \"catalogUrl\": \"$CONNECTOR_CATALOG_URL\", \"storageAccount\": \"$STORAGE_ACCOUNT\", \"apiKey\": \"$API_KEY\"}" \
--secrets-mount-path /usr/share/nginx/html/assets/config \
--dns-name-label edc-data-dashboard

Contributing

See how to contribute for details.