Skip to main content

Examples

This section provides examples of automation definitions in Port.

Send a Slack/Teams message when a service becomes unhealthy​

Automation definition​

By using the ENTITY_UPDATED trigger type, we can run custom logic whenever an entity of a specific type is updated.

For example, the following definition will cause a message to be sent whenever a service entity's status property becomes Unhealthy:

{
  "identifier": "serviceUnHealthy",
  "title": "Service Health Changed",
  "trigger": {
    "type": "automation",
    "event": {
      "type": "ENTITY_UPDATED",
      "blueprintIdentifier": "service"
    },
    // The condition below checks the service's status property after the update
    // The automation will only run for services whose status becomes "Unhealthy"
    "condition": {
      "type": "JQ",
      "expressions": [
        ".diff.after.properties.status == \"Unhealthy\""
      ],
      "combinator": "and"
    }
  },
  "invocationMethod": {
    "type": "GITHUB",
    "org": "github-org-name",
    "repo": "github-repo-name",
    "workflow": "slack-teams-notify-unhealthy-service.yaml",
    // workflowInputs is the payload to be passed to the GitHub workflow upon every execution
    // In this example, we pass the updated service's identifier
    "workflowInputs": {
      "service_name": "{{ .event.context.entityIdentifier }}"
    },
    "reportWorkflowStatus": true
  },
  "publish": true
}

Backend - GitHub workflow​

The slack-teams-notify-unhealthy-service.yaml workflow will contain the logic to send a Slack/Teams message.

Prerequisite - set up webhooks

The workflow requires a Slack webhook URL and/or a Microsoft Teams webhook URL to send the message.

Slack:

  1. To set up a Slack webhook, follow the instructions here.
  2. Once you have the webhook URL, add it as a secret in your GitHub repository named SLACK_WEBHOOK_URL.

Microsoft Teams:

  1. To set up a Microsoft Teams webhook, follow the instructions here.
  2. Once you have the webhook URL, add it as a secret in your GitHub repository named TEAMS_WEBHOOK_URL.
slack-teams-notify-unhealthy-service.yaml
name: Notify when service becomes unhealthy

on:
  workflow_dispatch:
    inputs:
      # Note that the input is the same as the payload (workflowInputs) defined in the automation
      service_name:
        description: 'The unhealthy service's name'
        required: true
        type: string

jobs:
  send_message:
    runs-on: ubuntu-latest
    steps:
      - name: Send message to Slack
        env:
          SLACK_WEBHOOK_URL: ${{ secrets.SLACK_WEBHOOK_URL }}
        run: |
          curl -X POST -H 'Content-type: application/json' --data '{"text":"The service ${{ inputs.service_name }} has become unhealthy."}' $SLACK_WEBHOOK_URL
      
      - name: Send message to Microsoft Teams
        env:
          TEAMS_WEBHOOK_URL: ${{ secrets.TEAMS_WEBHOOK_URL }}
        run: |
          curl -H 'Content-Type: application/json' -d '{"text": "The service ${{ inputs.service_name }} has become unhealthy."}' $TEAMS_WEBHOOK_URL

Terminate an ephemeral environment when its TTL is expired​

Automation definition​

By using the TIMER_EXPIRED trigger type, we can run custom logic whenever a timer property expires.

The following definition will cause a webhook to be triggered whenever the ttl property expires on an environment entity:

{
  "identifier": "ttlEphemeralEnvironment",
  "title": "Terminate ephemeral environment",
  "trigger": {
    "type": "automation",
    "event": {
      "type": "TIMER_PROPERTY_EXPIRED",
      "blueprintIdentifier": "environment",
      "propertyIdentifier": "ttl"
    }
  },
  "invocationMethod": {
    "type": "WEBHOOK",
    "url": "https://myWebhookUrl.com",
    // Under "body" we specify the payload to be passed to the webhook upon every execution
    // In this example, we pass the id of the entity whose TTL has expired and the run id
    "body": {
      "entityId": "{{ .event.context.entityIdentifier }}",
      "runId": "{{ .run.id }}"
    },
  },
  "publish": true
}

Backend - Webhook​

Since the webhook implementation is entirely up to you, it can be used to terminate the environment, clean up resources, send a notification to the relevant team, and anything else that you want to happen as part of the termination process.
The run id can be used to interact with the execution in Port - send log messages and/or update the execution's status.