Deploy an app

Deploy apps using the Darcy Portal UI or terminal commands via edgectl

Prerequisites #

To deploy your custom app, you will need a Darcy Cloud account with at least one node accessible and ONLINE. You will need a basic understanding of how to structure app YAML for Darcy Cloud.

Deploy an App using Darcy Cloud Portal #

1. Go to the Darcy Cloud project page
2. Select Apps
3. Click on + DEPLOY APP
4. Drag n’ drop a YAML file into the box

View your Application Details #

Click on your app in the list of application to see its details.

The appl details page contains information about its microservices , routes, the nodes involved in running the microservices and its YAML Definition.

Darcy Cloud Portal allows you to edit your app YAML right in the browser. More information about the Application YAML files can be found here .

Everything shown above using Darcy Cloud Portal can also be achieved using our CLI, edgectl .

Deploy an App using edgectl #

Once we have an edge node in our project, we can deploy applications to run on those nodes.

To deploy an app, use edgectl deploy app. We can run the following to learn how it works:

edgectl deploy app --help

The deploy app command requires a path to a YAML file to be specified. See the YAML for our demo apps: Heart Rate Demo and Darcy AI Explorer .

The YAML file defines everything about our edge application that is required to deploy it. For example:

edgectl deploy app ./edgeworx-heart-rate.yaml

You can use edgectl commands to view details of the app.

$ edgectl get apps
NAME                 DESCRIPTION  STATUS
edgeworx-heart-rate               running

$ edgectl get app edgeworx-heart-rate
NAME                 DESCRIPTION  STATUS
edgeworx-heart-rate               running

If you need more detail on an app, use the --json flag, e.g. edgectl get app edgeworx-heart-rate --json: (output abbreviated for clarity):

{
  "description": "",
  "id": 1,
  "isActivated": true,
  "name": "edgeworx-heart-rate",
  "projectUUID": "b75676cb-ae3f-4808-a992-0811e077d783",
  "routes": [
    {
      "from": "monitor",
      "name": "monitor-to-viewer",
      "to": "viewer"
    }
  ],
  "status": "running"
}

Note that an app consists of one or more microservices (containers). Use edgectl get microservices:

$ edgectl get microservices --app edgeworx-heart-rate
NAME     STATUS   IMAGE                                        NODE
monitor  RUNNING  edgeworx/healthcare-heart-rate:arm-v1        raspberrypie1c0fe7939883f228946
viewer   RUNNING  edgeworx/healthcare-heart-rate-ui-arm:1.0.0  raspberrypie1c0fe7939883f228946

To view the details of the monitor microservice in the edgeworx-heart-rate app:

edgectl get microservice --app edgeworx-heart-rate monitor

Container Logs #

You can use the edgectl logs command to view the container logs for a microservice. This is effectively equivalent to executing docker logs CONTAINER_ID. The logs command takes one argument, APP_NAME/MICROSERVICE_NAME. For example:

$ edgectl logs edgeworx-heart-rate/monitor --tail 5
[Thu Feb 24 2022 20:52:49] [LOG]   test-mode = true, generating mock sensor data..
[Thu Feb 24 2022 20:52:54] [LOG]   Retrieving heart rate sensor reading
[Thu Feb 24 2022 20:52:54] [LOG]   test-mode = true, generating mock sensor data..
[Thu Feb 24 2022 20:52:59] [LOG]   Retrieving heart rate sensor reading
[Thu Feb 24 2022 20:52:59] [LOG]   test-mode = true, generating mock sensor data..

Note that you can use the typical logs arguments, such as --tail 100 or --follow.

The following section will go into detail about application YAML files.