The IR-mapping/ListAllEnvironments script

The aim of this script is to find and list all the machines that have an integration runtime client installed. It uses the Azure Rest APIs to iterate through all subscriptions available to the user.

When the script is done, a list of all inte nodes is exported in a CSV-file.

The CSV-file

The file contains the following data:

• Subscription name
• Datafactory name
• IR-Name (Integration Runtime name)
• Node Name: The name of the node in Azure. Does not have to be the same as machine name
• Machine Name: The name of the server where the IR runtime client is installed.
• Status: Online or Offline
• Autoupdate: On means configured and you should be good to go
• Version Status:
  • UpToDate: No problems.
  • Expiring: the installed version will soon be out of support.
  • Expired: the installed version is not supported (good luck).
• Registered: The local date and time when the client was installed.
• Last Used: The local date and time when the client was last accessed.

What you need to look out for is expired versions in combination with active online nodes.

Consider removing/uninstalling inactive (last used) nodes.

The script

In order to execute the script, you need to login with valid Azure credentials (usually your adm-account). You will only see data factories located in subscriptions you are allowed to (at least) Read. You also need to use PowerShell version 7.x. This does not work on Windows PowerShell (5.1).

Login will open a browser window and ask for your login, as normal. After login you can close the tab.

The script feeds back names of subscriptions and factores.

...
Looking in AzureSubscription1
Looking in Azure-Prod
Found node:  vmappne24
Looking in Azure-Test
Found node:  integration-test-001
Found node:  vmappne22
...

When the script is done a file called Integration Runtime Nodes.csv is saved in the same directory.

Technical about the script

It uses the Azure Rest APIs and loops like this:

Get Subscriptions

• For each subscription (Get all data factories)
  • For each Datafactory (Get all integration runtimes)
    • For each Integration Runtime (Get all Nodes and status)
      • For each Node (extract node info from Status response)

With each call (except the first one) the URL for the next call is constucted using the id property.

Here is an example of the response for getting all Data Factories in the DataAnalyticsDev subscription:

{
    "value": [
        {
            "id": "/subscriptions/<subscriptionID>/resourceGroups/<resourcegroupname>/providers/Microsoft.DataFactory/factories/<datafactoryname>",
            "name": "<datafactoryname>",
            "type": "Microsoft.DataFactory/factories",
            "location": "eastus",
            "identity": {
                "principalId": "<GUID>",
                "tenantId": "<GUID>",
                "type": "SystemAssigned"
            }
        }
        ... More items
    ]
}

The ID-property can be used to get the next call; getting all the integration runtimes. Simply add /integrationRuntimes and the API-version ID at the end.

Only Self Hosted nodes are listed as those are the only ones we really care about.