Can I set a service to run in its own container thru Powershell? ie.
What is the Powershell equivalent of sc config <service name> type= own
You can change service type property on the corresponding WMI object.
# get service
$s = (Get-WmiObject win32_service -filter "Name='Fax'")
# change service type
$s.Change($null, $null, 16) # 16 = "Own Process"
Other values you can find here (http://msdn.microsoft.com/en-us/library/aa384901(v=vs.85).aspx)
Make sure that ReturnValue from change method is equal to 0. Meaning of error codes you can find in the link above.
Related
I have a little script where the user enters the script name (the description name) and has the option to modify the properties. This is done with a GUI interface: a textbox and some drop-down lists.
I tried several ways to make it but none of them worked.
Here is how I retrieve the service information:
$thisservice = Get-WmiObject -Class Win32_Service -ComputerName $Server | Where { $_.displayName -eq $serviceBox.text }
I look at $server computer and match up the "service display name" to the value contained in the textbox field: $_.displayName -eq $serviceBox.text. This works well, if I list the data I get exactly what is in the service console: .name, .description, .state, .startmode, .displayname
I store the service name in a variable:
$servicename = $thisservice.name
I know you can not start a service unless it is on "Manual" or "auto" so first I set it to auto. Here I tryed to different ways but none worked:
$servicename.ChangeStartMode("Auto")
and
Set-Service $servicename -StartupType Auto
and then I start the service (but this doesn't work either even if I preset the state-mode on Auto form the service manager):
$servicename.startservice()
Also the $servicename.stopservcie() fails.
I tried this on 2 different machines, on both I have admin rights, I tryed running the script with Admin mode, same result. I even tried it directly from PS console and I couldn't change the service state!
What am I doing wrong?
Your problem is here: $servicename = $thisservice.name
By doing that you set $servicename to be a string, not the service object. Therefore you can't call service methods on it any more.
You should be able to just do:
$thisservice.ChangeStartMode("Automatic")
$thisservice.StartService()
I am having problems with a powershell script.
I wrote a script that would search for a windows service with a specific name, and it would Stop or Start this service.
It works when I run it on a server which I log into with a service account that I know that can access the service console. However when it runs off of my build server, the script is no longer able to find the services. I tried giving the service account that runs script the same privaledges as the other service account but that doesn't seem to work.
[System.ServiceProcess.ServiceController]$service = Get-Service -Name $ServiceName -ComputerName $Remoteserver -ErrorAction SilentlyContinue
That is the line that is not longer able to find the service. What am I doing wrong. Is there a way to impersonate a user that can find the service? Any help would be appreciated.
You could try supplying the credentials of the service account using the -Credential parameter. However, since you imply that it used to work with the account that runs the script remotely and no longer does, I think a more likely culprit is that $ServiceName used to only match one service on the target computer, and now there is another service whose name matches that string. If more than one service matches the -Name parameter, Get-Service returns an array of ServiceController objects.
Try running it without ErrorAction -SilentlyContinue. If you get the following error message, then that's what's happening:
Cannot convert the "System.Object[]" value of type "System.Object[]" to type "System.ServiceProcess.ServiceController".
If you get a different error message, please add the full error message to the question.
I'm trying to retrieve some parameters from xenstore using WMI (specifically, I was hoping to use this script to change a VM IP address after it's created).
According to this article, it seems like I should just be able to do something like:
From the xenserver CLI:
xe vm-param-set uuid=e66660e9-85e1-1f99-3229-1dfa7d1065a8 xenstore-data:data/TempValue=test
then in a powershell script:
$base = gwmi -n root\wmi -cl CitrixXenStoreBase
$sid = $base.AddSession("MyNewSession")
$session = gwmi -n root\wmi -q "select * from CitrixXenStoreSession where SessionId=$($sid.SessionId)"
$output = $session.GetValue("data/TempValue").value
log "$output"
But that doesn't seem to retrieve the value that I expect.
One thing I noticed was if I set the value from a powershell script, it seems to consistently retrieve the value when I run the previous script:
$base = gwmi -n root\wmi -cl CitrixXenStoreBase
$sid = $base.AddSession("MyNewSession")
$session = gwmi -n root\wmi -q "select * from CitrixXenStoreSession where SessionId=$($sid.SessionId)"
$session.SetValue("data/TempValue","This is a string")
It seems to retain the set value across sessions, but when I go back to the CLI and attempt to find the value, I get nothing:
xe vm-param-list uuid=e66660e9-85e1-1f99-3229-1dfa7d1065a8 | grep TempValue
So what it boils down to is that I'd like to either:
Know how to retrieve a xenstore parameter in a WMI script after executing the xe vm-param-set command.
Know how to set a parameter in the xenserver CLI in the same way that $session.SetValue works in the above example.
Nevermind, looks like this was user error on my end. I was setting the values after the VM was already started. Looks like the parameters have to be set before the VM starts (or the VM should be restarted).
I am trying to understand the mechanism behind powershell comand.
I see that get-service return a service object that name is one of its fields, so you can filter by name (using where, select...).
I am trying to understand why when i run this:
Get-Service *sql*
PowerShell assume that the argument is filter by name? Where can i see it in the command description?
PowerShell-3.0
Try this:
Get-Help Get-Service
You'll notice that the default parameter set will accept the first argument as the -Name parameter.
You are effectively doing this:
Get-Service -Name *sql*
I'm attempting to use powershell to access a remote registry like so:
$reg = [Microsoft.Win32.RegistryKey]::OpenRemoteBaseKey("LocalMachine", $server)
$key = $reg.OpenSubkey($subkeyPath)
Depending on some factors that I'm not yet able to determine I either get
Exception calling "OpenSubKey" with "1" argument(s): "Requested registry access is not allowed."
Or
System.UnauthorizedAccessException: Attempted to perform an unauthorized operation.
at Microsoft.Win32.RegistryKey.Win32ErrorStatic(Int32 errorCode, String str)
at Microsoft.Win32.RegistryKey.OpenRemoteBaseKey(RegistryHive hKey, String machineName)
It seems pretty clear that this is because the user I'm running the powershell script as doesn't have the appropriate credentials to access the remote registry. I'd like to be able to supply a set of credentials to use for the remote registry access, but I can find no documentation anywhere of a way to do this. I'm also not clear on exactly where to specify which users are allowed to access the registry remotely.
Just thought I'd add my answer to anyone with this problem as well. It seems there is no way to add Credentials using RemoteRegistry. You can however use WMI to query a remote registry using alternative credentials as follows:
$reg = Get-WmiObject -List -Namespace root\default -ComputerName RemotePC -Credential "Domain\User" | Where-Object {$_.Name -eq "StdRegProv"}
From here you can call standard Registry methods. The below example will return the operating system.
$HKLM = 2147483650
$reg.GetStringValue($HKLM,"SOFTWARE\Microsoft\Windows NT\CurrentVersion","ProductName").sValue
Hope this helps someone :)
Are you running remote registry service? It is disabled by default and that must be causing the issue. Check the status of this service on all remote machines you are trying to access.
I couldn't comment directly on bentaylr's entry above, but I've taken what he contributed and added PSCredentials creation (figured out from here) to allow you to hard code credentials into the script.
Peace of mind disclaimer: Be careful when using plaintext credentials in a script. In my case, I'm using generic credentials on machines I'm launching. Depending on your case, you might consider creating an encrypted credential file to store the password in (see link above).
The credentials you use would need to be able to access the registry if you were logged into that user on the machine you are targeting.
$user = "Domain\Username"
$pass = ConvertTo-SecureString "Password" -AsPlainText -Force
$cred = New-Object -TypeName System.Management.Automation.PSCredential -ArgumentList $user,$pass
$reg = Get-WmiObject -List -Namespace root\default -ComputerName $server -Credential $cred | Where-Object {$_.Name -eq "StdRegProv"}
$HKLM = 2147483650
$value = $reg.GetStringValue($HKLM,"Software\Microsoft\.NetFramework","InstallRoot").sValue
$key.OpenSubKey($subkeyName) opens the subkey in write protected mode,
$key.OpenSubKey($subkeyName,$true) opens it in writable mode
Therefore after $key.OpenSubKey($subkeyName,$true) you should be able to create a new subkey or value
If you try the same thing after $key.OpenSubKey($subkeyName) you will get "UnauthorizedAccessException"
PS C:\>$regKey.OpenSubKey
OverloadDefinitions
Microsoft.Win32.RegistryKey OpenSubKey(string name, **bool Writable**)
try
PS C:\>$key.OpenSubKey($subkeyName,**$true**)
http://msdn.microsoft.com/en-us/library/xthy8s8d%28v=vs.110%29.aspx
Came looking for the answer to your question, but in a little googling this morning I noticed that the first parameter is a type rather than a String... hope this helps:
$machine = "<Machine Name Goes Here>"
$type = [Microsoft.Win32.RegistryHive]::LocalMachine
$regkey = [Microsoft.Win32.RegistryKey]::OpenRemoteBaseKey($type,$machine)
$subkey = $regKey.OpenSubKey($key)
foreach ($sub in $regKey.GetSubKeyNames()){$sub}
I wanted to first thank all for answers above really helpful, wanted to add that you can use Get-Credential command to collect credentials without having to hard code it in your script. I have written using the above suggestions into my script the following code and query:
$userCredentials = Get-Credential -Credential <domain\username>
$objReg = Get-WmiObject -List -Namespace root\default -ComputerName $server -Credential $userCredentials | Where-Object{$_.Name -eq "StdRegProv"}
$subKeyNames = $objReg.EnumKey($HKLM,"SOFTWARE\Microsoft\Updates\Microsoft .Net Framework 4.5.1").sNames
The above code returns all sub key names in the specified key so that I can determine installed updates other than OS which have been applied to a server. If you want to determine all collection possibilities with the $objReg variable then run:
$objReg | Get-Member
You will see a list of all possible queries which can be performed against the registry. Hope this helps!