Edit

Conditional statements and loops in a Configuration

  • Article

You can make your Configuration more dynamic by using PowerShell flow-control keywords. This article shows you how you can use conditional statements and loops to make your Configuration more dynamic. Combining conditional statements and loops with parameters and Configuration Data allows you more flexibility and control when compiling your Configuration.

Just like a function or a script block, you can use any PowerShell language feature within a Configuration. The statements you use will only be evaluated when you call your Configuration to compile a .mof file. The examples below show scenarios to demonstrate concepts. Conditional statements and loops are more often used with parameters and configuration Data.

In this example, the Service resource block retrieves the current state of a service at compile time to generate a .mof file that maintains its current state.

Note

Using dynamic Resource blocks will preempt the effectiveness of Intellisense. The PowerShell parser cannot determine if the values specified are acceptable until the Configuration is compiled.

Configuration ServiceState
{
    # It is best practice to explicitly import any resources used in your Configurations.
    Import-DSCResource -Name Service -Module PSDesiredStateConfiguration
    Node localhost
    {
        Service Spooler
        {
            Name = "Spooler"
            State = $(Get-Service -Name 'spooler').Status
            StartType = $(Get-Service -Name 'spooler').StartType
        }
    }
}

Additionally, you could create a Service resource block for every service on the current machine using a foreach loop.

Configuration ServiceState
{
    # It is best practice to explicitly import any resources used in your Configurations.
    Import-DSCResource -Name Service -Module PSDesiredStateConfiguration
    Node localhost
    {
        foreach ($service in $(Get-Service))
        {
            Service $service.Name
            {
                Name = $service.Name
                State = $service.Status
                StartType = $service.StartType
            }
        }
    }
}

You could also create a Configuration only for machines that are online by using an if statement.

Configuration ServiceState
{
    # It is best practice to explicitly import any resources used in your Configurations.
    Import-DSCResource -Name Service -Module PSDesiredStateConfiguration

    foreach ($computer in @('Server01', 'Server02', 'Server03'))
    {
        if (Test-Connection -ComputerName $computer)
        {
            Node $computer
            {
                Service "Spooler"
                {
                    Name = "Spooler"
                    State = "Started"
                }
            }
        }
    }
}

Note

The dynamic resource blocks in the above examples reference the current machine. In this instance, that would be the machine you are authoring the Configuration on, not the target Node.

Summary

In summary, you can use any PowerShell language feature within a Configuration.

This includes things like:

  • Custom Objects
  • Hashtables
  • String manipulation
  • Remoting
  • WMI and CIM
  • ActiveDirectory objects
  • and more...

Any PowerShell code defined in a Configuration is evaluated at compile time, but you can also place code in the script containing your Configuration. Any code outside of the Configuration block is executed when you import your Configuration.

See also