Performing networking tasks

This sample only applies to Windows platforms.

Because TCP/IP is the most commonly used network protocol, most low-level network protocol administration tasks involve TCP/IP. In this section, we use PowerShell and WMI to do these tasks.

Listing IP addresses for a computer

To get all IP addresses in use on the local computer, use the following command:

Get-CimInstance -Class Win32_NetworkAdapterConfiguration -Filter IPEnabled=$true |
    Select-Object -ExpandProperty IPAddress

Since the IPAddress property of a Win32_NetworkAdapterConfiguration object is an array, you must use the ExpandProperty parameter of Select-Object to see the entire list of addresses.

10.0.0.1
fe80::60ea:29a7:a233:7cb7
2601:600:a27f:a470:f532:6451:5630:ec8b
2601:600:a27f:a470:e167:477d:6c5c:342d
2601:600:a27f:a470:b021:7f0d:eab9:6299
2601:600:a27f:a470:a40e:ebce:1a8c:a2f3
2601:600:a27f:a470:613c:12a2:e0e0:bd89
2601:600:a27f:a470:444f:17ec:b463:7edd
2601:600:a27f:a470:10fd:7063:28e9:c9f3
2601:600:a27f:a470:60ea:29a7:a233:7cb7
2601:600:a27f:a470::2ec1

Using the Get-Member cmdlet, you can see that the IPAddress property is an array:

Get-CimInstance -Class Win32_NetworkAdapterConfiguration -Filter IPEnabled=$true |
    Get-Member -Name IPAddress
   TypeName: Microsoft.Management.Infrastructure.CimInstance#root/cimv2/Win32_NetworkAdapterConfiguration

Name      MemberType Definition
----      ---------- ----------
IPAddress Property   string[] IPAddress {get;}

The IPAddress property for each network adapter is actually an array. The braces in the definition indicate that IPAddress isn't a System.String value, but an array of System.String values.

Listing IP configuration data

To display detailed IP configuration data for each network adapter, use the following command:

Get-CimInstance -Class Win32_NetworkAdapterConfiguration -Filter IPEnabled=$true

The default display for the network adapter configuration object is a very reduced set of the available information. For in-depth inspection and troubleshooting, use Select-Object or a formatting cmdlet, such as Format-List, to specify the properties to be displayed.

In modern TCP/IP networks you are probably not interested in IPX or WINS properties. You can use the ExcludeProperty parameter of Select-Object to hide properties with names that begin with "WINS" or "IPX".

Get-CimInstance -Class Win32_NetworkAdapterConfiguration -Filter IPEnabled=$true |
    Select-Object -ExcludeProperty IPX*,WINS*

This command returns detailed information about DHCP, DNS, routing, and other minor IP configuration properties.

Pinging computers

You can perform a simple ping against a computer using by Win32_PingStatus. The following command performs the ping, but returns lengthy output:

Get-CimInstance -Class Win32_PingStatus -Filter "Address='127.0.0.1'"

A more useful form for summary information a display of the Address, ResponseTime, and StatusCode properties, as generated by the following command. The Autosize parameter of Format-Table resizes the table columns so that they display properly in PowerShell.

Get-CimInstance -Class Win32_PingStatus -Filter "Address='127.0.0.1'" |
    Format-Table -Property Address,ResponseTime,StatusCode -Autosize
Address   ResponseTime StatusCode
-------   ------------ ----------
127.0.0.1            0          0

A StatusCode of 0 indicates a successful ping.

You can use an array to ping multiple computers with a single command. Because there is more than one address, use the ForEach-Object to ping each address separately:

'127.0.0.1','localhost','bing.com' |
  ForEach-Object -Process {
    Get-CimInstance -Class Win32_PingStatus -Filter ("Address='$_'") |
      Select-Object -Property Address,ResponseTime,StatusCode
  }

You can use the same command format to ping all the addresses on a subnet, such as a private network that uses network number 192.168.1.0 and a standard Class C subnet mask (255.255.255.0)., Only addresses in the range of 192.168.1.1 through 192.168.1.254 are legitimate local addresses (0 is always reserved for the network number and 255 is a subnet broadcast address).

To represent an array of the numbers from 1 through 254 in PowerShell, use the expression 1..254. A complete subnet ping can be performed by adding each value in the range to a partial address in the ping statement:

1..254| ForEach-Object -Process {
  Get-CimInstance -Class Win32_PingStatus -Filter ("Address='192.168.1.$_'") } |
    Select-Object -Property Address,ResponseTime,StatusCode

Note that this technique for generating a range of addresses can be used elsewhere as well. You can generate a complete set of addresses in this way:

$ips = 1..254 | ForEach-Object -Process {'192.168.1.' + $_}

Retrieving network adapter properties

Earlier, we mentioned that you could retrieve general configuration properties using the Win32_NetworkAdapterConfiguration class. Although not strictly TCP/IP information, network adapter information such as MAC addresses and adapter types can be useful for understanding what's going on with a computer. To get a summary of this information, use the following command:

Get-CimInstance -Class Win32_NetworkAdapter -ComputerName .

Assigning the DNS domain for a network adapter

To assign the DNS domain for automatic name resolution, use the SetDNSDomain method of the Win32_NetworkAdapterConfiguration. The Query parameter of Invoke-CimMethod takes a WQL query string. The cmdlet calls the method specified on each instance returned by the query.

$wql = 'SELECT * FROM Win32_NetworkAdapterConfiguration WHERE IPEnabled=True'
$args = @{ DnsDomain = 'fabrikam.com'}
Invoke-CimMethod -MethodName SetDNSDomain -Arguments $args -Query $wql

Filtering on IPEnabled=True is necessary, because even on a network that uses only TCP/IP, several of the network adapter configurations on a computer aren't true TCP/IP adapters. they're general software elements supporting RAS, VPN, QoS, and other services for all adapters and thus don't have an address of their own.

Performing DHCP configuration tasks

Modifying DHCP details involves working with a set of network adapters, just as the DNS configuration does. There are several distinct actions you can perform using WMI.

Finding DHCP-enabled adapters

To find the DHCP-enabled adapters on a computer, use the following command:

Get-CimInstance -Class Win32_NetworkAdapterConfiguration -Filter "DHCPEnabled=$true"

To exclude adapters with IP configuration problems, you can retrieve only IP-enabled adapters:

Get-CimInstance -Class Win32_NetworkAdapterConfiguration -Filter "IPEnabled=$true and DHCPEnabled=$true"

Retrieving DHCP properties

Because DHCP-related properties for an adapter generally begin with DHCP, you can use the Property parameter of Format-Table to display only those properties:

Get-CimInstance -Class Win32_NetworkAdapterConfiguration -Filter  "IPEnabled=$true and DHCPEnabled=$true" |
  Format-Table -Property DHCP*

Enabling DHCP on each adapter

To enable DHCP on all adapters, use the following command:

$wql = 'SELECT * from Win32_NetworkAdapterConfiguration WHERE IPEnabled=True and DHCPEnabled=False'
Invoke-CimMethod -MethodName ReleaseDHCPLease -Query $wql

Using the filter statement IPEnabled=True and DHCPEnabled=False avoids enabling DHCP where it's already enabled.

Releasing and renewing DHCP leases on specific adapters

Instances of the Win32_NetworkAdapterConfiguration class has ReleaseDHCPLease and RenewDHCPLease methods. Both are used in the same way. In general, use these methods if you only need to release or renew addresses for an adapter on a specific subnet. The easiest way to filter adapters on a subnet is to choose only the adapter configurations that use the gateway for that subnet. For example, the following command releases all DHCP leases on adapters on the local computer that are obtaining DHCP leases from 192.168.1.254:

$wql = 'SELECT * from Win32_NetworkAdapterConfiguration WHERE DHCPServer="192.168.1.1"'
Invoke-CimMethod -MethodName ReleaseDHCPLease -Query $wql

The only change for renewing a DHCP lease is to use the RenewDHCPLease method instead of the ReleaseDHCPLease method:

$wql = 'SELECT * from Win32_NetworkAdapterConfiguration WHERE DHCPServer="192.168.1.1"'
Invoke-CimMethod -MethodName RenewDHCPLease -Query $wql

Note

When using these methods on a remote computer, be aware that you can lose access to the remote system if you are connected to it through the adapter with the released or renewed lease.

Releasing and renewing DHCP leases on all adapters

You can perform global DHCP address releases or renewals on all adapters by using the Win32_NetworkAdapterConfiguration methods, ReleaseDHCPLeaseAll and RenewDHCPLeaseAll. However, the command must apply to the WMI class, rather than a particular adapter, because releasing and renewing leases globally is performed on the class, not on a specific adapter. The Invoke-CimMethod cmdlet can call the methods of a class.

Invoke-CimMethod -ClassName Win32_NetworkAdapterConfiguration -MethodName ReleaseDHCPLeaseAll

You can use the same command format to invoke the RenewDHCPLeaseAll method:

Invoke-CimMethod -ClassName Win32_NetworkAdapterConfiguration -MethodName RenewDHCPLeaseAll

Creating a network share

To create a network share, use the Create method of Win32_Share:

Invoke-CimMethod -ClassName Win32_Share -MethodName Create -Arguments @{
    Path = 'C:\temp'
    Name = 'TempShare'
    Type = [uint32]0 #Disk Drive
    MaximumAllowed = [uint32]25
    Description = 'test share of the temp folder'
}

This is equivalent to the following net share command on Windows:

net share tempshare=c:\temp /users:25 /remark:"test share of the temp folder"

To call a method of a WMI class that takes parameters you must know what parameters are available and the types of those parameters. For example, you can list the methods of the Win32_Class with the following commands:

(Get-CimClass -ClassName Win32_Share).CimClassMethods
Name          ReturnType Parameters                                   Qualifiers
----          ---------- ----------                                   ----------
Create            UInt32 {Access, Description, MaximumAllowed, Name…} {Constructor, Implemented, MappingStrings, Stati…
SetShareInfo      UInt32 {Access, Description, MaximumAllowed}        {Implemented, MappingStrings}
GetAccessMask     UInt32 {}                                           {Implemented, MappingStrings}
Delete            UInt32 {}                                           {Destructor, Implemented, MappingStrings}

Use the following command to list the parameters of the Create method.

(Get-CimClass -ClassName Win32_Share).CimClassMethods['Create'].Parameters
Name            CimType Qualifiers                                  ReferenceClassName
----            ------- ----------                                  ------------------
Access         Instance {EmbeddedInstance, ID, In, MappingStrings…}
Description      String {ID, In, MappingStrings, Optional}
MaximumAllowed   UInt32 {ID, In, MappingStrings, Optional}
Name             String {ID, In, MappingStrings}
Password         String {ID, In, MappingStrings, Optional}
Path             String {ID, In, MappingStrings}
Type             UInt32 {ID, In, MappingStrings}

You can also read the documentation for Create method of the Win32_Share class.

Removing a network share

You can remove a network share with Win32_Share, but the process is slightly different from creating a share, because you need to retrieve the specific instance to be removed, rather than the Win32_Share class. The following example deletes the share TempShare:

$wql = 'SELECT * from Win32_Share WHERE Name="TempShare"'
Invoke-CimMethod -MethodName Delete -Query $wql

Connecting a Windows-accessible network drive

The New-PSDrive cmdlet can create a PowerShell drive that's mapped to a network share.

New-PSDrive -Name "X" -PSProvider "FileSystem" -Root "\\Server01\Public"

However, drives created this way are only available to PowerShell session where they're created. To map a drive that's available outside of PowerShell (or to other PowerShell sessions), you must use the Persist parameter.

New-PSDrive -Persist -Name "X" -PSProvider "FileSystem" -Root "\\Server01\Public"

Note

Persistently mapped drives may not be available when running in an elevated context. This is the default behavior of Windows UAC. For more information, see the following article: