I need to periodically scan a folder for new fontfiles to install and delete them afterwards using a powershell script.
During processing I want to skip already installed files and to achieve that I need to resolve the "real" fontname of the provided file.
I figured out everything and it seems to work everything but the file deletion.
The deletion did work until I added the font name resolution using this GlythTypeInterface Object. It seems like the invoked object does "file lock" the fontfile resulting in an UnauthorizedAccessException.
Thats why I tried some garbage collection stuff I found but I can't make it work.
My code so far:
Add-Type -AssemblyName PresentationCore
$FONTS = 0x14
$Path="C:\_fonts_to_install"
$FontItem = Get-Item -Path $Path
$FontList = Get-ChildItem -Path "$FontItem\*" -Include ('*.fon','*.otf','*.ttc','*.ttf')
$objShell = New-Object -ComObject Shell.Application
$objFolder = $objShell.Namespace($FONTS)
$Fontdir = dir $Path
$username = $env:UserName
foreach($File in $FontList) {
$try = $true
$installedFonts = #(Get-ChildItem C:\Users\$username\AppData\Local\Microsoft\Windows\Fonts | Where-Object {$_.PSIsContainer -eq $false} | Select-Object basename)
$fontObject = New-Object -TypeName Windows.Media.GlyphTypeface -ArgumentList $File.fullname
$fontName = $fontObject.Win32FamilyNames.Values
Write-Host $fontName
$fontObject = $null
Remove-Variable fontObject
foreach($font in $installedFonts)
{
if ($font -match $fontName)
{
$try = $false
}
}
if ($try)
{
$objFolder.CopyHere($File.fullname)
}
Write-Host $File
Remove-Item $File -Force -Verbose
}
I'm trying to use this function to monitor BSOD remotely I know I have to add a parameter of computername just not sure where
Function Get-BSODInfo{
try {
Invoke-WebRequest -Uri "https://www.nirsoft.net/utils/bluescreenview.zip" -OutFile "$($ENV:Temp)\bluescreeview.zip"
Expand-Archive "$($ENV:Temp)\bluescreeview.zip" -DestinationPath "$($ENV:Temp)" -Force
Start-Process -FilePath "$($ENV:Temp)\Bluescreenview.exe" -ArgumentList "/scomma `"$($ENV:Temp)\Export.csv`"" -Wait
}
catch {
Write-Host "BSODView Command has Failed: $($_.Exception.Message)"
exit 1
}
$BSODs = get-content "$($ENV:Temp)\Export.csv" | ConvertFrom-Csv -Delimiter ',' -Header Dumpfile, Timestamp, Reason, Errorcode, Parameter1, Parameter2, Parameter3, Parameter4, CausedByDriver | foreach-object { $_.Timestamp = [datetime]::Parse($_.timestamp, [System.Globalization.CultureInfo]::CurrentCulture); $_ }
Remove-item "$($ENV:Temp)\Export.csv" -Force
$BSODFilter = $BSODs | where-object { $_.Timestamp -gt ((get-date).addhours(-24)) }
if (!$BSODFilter) {
write-host "Healthy - No BSODs found in the last 24 hours"
}
else {
write-host "Unhealthy - BSOD found. Check Diagnostics"
$BSODFilter
exit 1
}
}
I made a small Powershell script 'Start.ps1' which launches some websites on Windows startup. When I launch it manually, all configured websites are started in Chrome. However, when I run it with task scheduler, the task runs, but Chrome does not start.
schtasks /create /tn "Start" /sc onstart /delay 0000:30 /rl highest /ru system /tr "powershell.exe -file 'C:\Users\myuser\Google Drive\Home\Start.ps1'"
I've also added some logs in my powershell script so I'm sure the script has started:
Function Write-Log {
Param (
[parameter(Mandatory=$True,HelpMessage='Log output')][string]$Log,
[parameter(Mandatory=$True,HelpMessage='Log severity')][ValidateSet('Debug', 'Info', 'Warning', 'Error', 'Unknown')][string]$Severity,
[parameter(Mandatory=$True,HelpMessage='Log message')][string]$Message
)
$Now = Get-Date -Format 'yyyy-MM-dd HH:mm:ss,fff'
$LocalScriptName = Split-Path -Path $myInvocation.ScriptName -Leaf
If ( $Log -eq 'Verbose' ) {
Write-Verbose -Message ('{0}: {1}: {2}: {3}' -f $Now, $LocalScriptName, $Severity, $Message)
}
ElseIf ( $Log -eq 'Debug' ) {
Write-Debug -Message ('{0}: {1}: {2}: {3}' -f $Now, $LocalScriptName, $Severity, $Message)
}
ElseIf ( $Log -eq 'Output' ) {
Write-Verbose -Message ('{0}: {1}: {2}: {3}' -f $Now, $LocalScriptName, $Severity, $Message)
}
ElseIf ( $Log -match '^(([a-zA-Z0-9]|[a-zA-Z0-9][a-zA-Z0-9\-]*[a-zA-Z0-9])\.)*([A-Za-z0-9]|[A-Za-z0-9][A-Za-z0-9\-]*[A-Za-z0-9])(?::(?<port>\d+))$' -or $Log -match '^(([0-9]|[1-9][0-9]|1[0-9]{2}|2[0-4][0-9]|25[0-5])\.){3}([0-9]|[1-9][0-9]|1[0-9]{2}|2[0-4][0-9]|25[0-5])$' ) {
$IpOrHost = $log.Split(':')[0]
$Port = $log.Split(':')[1]
If ( $IpOrHost -match '^(([0-9]|[1-9][0-9]|1[0-9]{2}|2[0-4][0-9]|25[0-5])\.){3}([0-9]|[1-9][0-9]|1[0-9]{2}|2[0-4][0-9]|25[0-5])$' ) {
$Ip = $IpOrHost
}
Else {
$Ip = [Net.Dns]::GetHostAddresses($IpOrHost)[0].IPAddressToString
}
Try {
$LocalHostname = ([Net.Dns]::GetHostByName((& "$env:windir\system32\hostname.exe")).HostName).tolower()
$JsonObject = (New-Object -TypeName PSObject |
Add-Member -PassThru -NotePropertyName NoteProperty -NotePropertyValue logsource -InputObject $LocalHostname |
Add-Member -PassThru -NotePropertyName NoteProperty -NotePropertyValue hostname -InputObject $LocalHostname |
Add-Member -PassThru -NotePropertyName NoteProperty -NotePropertyValue scriptname -InputObject $LocalScriptName |
Add-Member -PassThru -NotePropertyName NoteProperty -NotePropertyValue logtime -InputObject $Now |
Add-Member -PassThru -NotePropertyName NoteProperty -NotePropertyValue severity_label -InputObject $Severity |
Add-Member -PassThru -NotePropertyName NoteProperty -NotePropertyValue message -InputObject $Message )
If ( $psversiontable.psversion.major -ge 3 ) {
$JsonString = $JsonObject | ConvertTo-Json
$JsonString = $JsonString -replace "`n",' ' -replace "`r",' '
}
Else {
$JsonString = $JsonObject | ConvertTo-Json2
}
$Socket = New-Object -TypeName System.Net.Sockets.TCPClient -ArgumentList ($Ip,$Port)
$Stream = $Socket.GetStream()
$Writer = New-Object -TypeName System.IO.StreamWriter -ArgumentList ($Stream)
$Writer.WriteLine($JsonString)
$Writer.Flush()
$Stream.Close()
$Socket.Close()
}
Catch {
Write-Verbose -Message ("{0}: {1}: Error: Something went wrong while trying to send message to logserver `"{2}`"." -f $Now, $LocalScriptName, $Log)
}
Write-Verbose -Message ('{0}: {1}: {2}: Ip: {3} Port: {4} JsonString: {5}' -f $Now, $LocalScriptName, $Severity, $Ip, $Port, $JsonString)
}
ElseIf ($Log -match '^((([a-zA-Z]:)|(\\{2}\w+)|(\\{2}(?:(?:25[0-5]|2[0-4]\d|[01]\d\d|\d?\d)(?(?=\.?\d)\.)){4}))(\\(\w[\w ]*))*)') {
If (Test-Path -Path $Log -pathType container){
Write-Host ('{0}: {1}: Error: Passed Path is a directory. Please provide a file.' -f $Now, $LocalScriptName)
Exit 1
}
ElseIf (!(Test-Path -Path $Log)) {
Try {
$Null = New-Item -Path $Log -ItemType file -Force
}
Catch {
$Now = Get-Date -Format 'yyyy-MM-dd HH:mm:ss,fff'
Write-Host ("{0}: {1}: Error: Write-Log was unable to find or create the path `"{2}`". Please debug.." -f $Now, $LocalScriptName, $Log)
exit 1
}
}
Try {
('{0}: {1}: {2}: {3}' -f $Now, $LocalScriptName, $Severity, $Message) | Out-File -filepath $Log -Append
}
Catch {
Write-Host ("{0}: {1}: Error: Something went wrong while writing to file `"{2}`". It might be locked." -f $Now, $LocalScriptName, $Log)
}
}
}
$Logfile="c:\Tempo\Start.log"
Write-Log $Logfile Info "Initiating Chrome Site Startup"
Start-Process -FilePath "C:\Program Files (x86)\Google\Chrome\Application\chrome.exe"
Start-Process -FilePath "C:\Program Files (x86)\Google\Chrome\Application\chrome.exe" -ArgumentList "https://outsideit.net"
Start-Process -FilePath "C:\Program Files (x86)\Google\Chrome\Application\chrome.exe" -ArgumentList "https://psychologischezorgopmaat.be"
Start-Process -FilePath "C:\Program Files (x86)\Google\Chrome\Application\chrome.exe" -ArgumentList "https://github.com"
Start-Process -FilePath "C:\Program Files (x86)\Google\Chrome\Application\chrome.exe" -ArgumentList "https://gitlab.com"
Start-Process -FilePath "C:\Program Files (x86)\Google\Chrome\Application\chrome.exe" -ArgumentList "http://localhost:9200"
Start-Process -FilePath "C:\Program Files (x86)\Google\Chrome\Application\chrome.exe" -ArgumentList "http://localhost:5601"
Write-Log $Logfile Info "Finalizing Chrome Site Startup"
How can I make this work in Task Scheduler? I tried running it with SYSTEM and my own user and with the "Run with highest privilege" option.
Always test if the scheduled task is working as intended when you run it manually (Scheduled Tasks - right click your task and Run it). When that works, move on to test if it works as intended when you start up the computer (or whenever, depending on the trigger).
When I imported your task with through cmd with the SYSTEM user account, the default setting was "Run whether user is logged on or not". If I tried to run it manually, the script wouldn't start.
I then changed the user to my account, changed the option to "Run only when the user is logged on", and that fixed the problem for me (many times before, as well). I wasn't able to change this option when using the SYSTEM user.
The goal is to reset the current cmd.exe shell to have the original set of environment variables. This should include deleting current environment variables that were created after the cmd.exe shell started.
The System and User environment variables can be read from the registry. But, the dynamic variables such as ALLUSERSPROFILE, APPDATA, LOGONSERVER, etc. are not in those locations. Where can those be found?
Because of this, the code cannot delete a variable created after the cmd.exe shell was started. This is because it may be one of the dynamic variables.
Put both of these files in the same directory.
=== Do-Environment.bat
#ECHO OFF
SET "TEMPFILE=%TEMP%\do-environment-script.bat"
powershell -NoLogo -NoProfile -File "%~dp0Do-Environment.ps1" >"%TEMPFILE%"
type "%TEMPFILE%"
CALL "%TEMPFILE%"
EXIT /B %ERRORLEVEL%
=== Do-Environment.ps1
$Vars = #{}
$UserVars = 'HKCU:\Environment'
$SystemVars = 'HKLM:\SYSTEM\CurrentControlSet\Control\Session Manager\Environment'
(Get-Item -Path $SystemVars).Property |
ForEach-Object {
$KeyName = $_
$KeyValue = (Get-Item -Path $SystemVars).GetValue($KeyName, $null)
$Vars[$KeyName] = $KeyValue
}
(Get-Item -Path $UserVars).Property |
ForEach-Object {
$KeyName = $_
$KeyValue = (Get-Item -Path $UserVars).GetValue($KeyName, $null)
if (($null -eq $Vars[$KeyName]) -or ($KeyName -ne 'Path')) {
$Vars[$KeyName] = $KeyValue
} else {
$Vars[$KeyName] = $Vars[$KeyName] + ';' + $KeyValue
}
#'SET "{0}={1}"' -f #($KeyName, $KeyValue)
}
$Vars.Keys | Sort-Object | ForEach-Object { 'SET "{0}={1}"' -f #($_, $Vars[$_]) }
I'm trying to divide output into two files with following script:
try {
gci C:\Windows\System32 -r -Force | % {
if (!$_.PsIsContainer) {
$_.FullName;
$file = $_.FullName
}
} > paths.txt
} catch [System.UnauthorizedAccessException] {
$file > errors.txt
}
I'm aware that you can't catch non-terminating scripts with catch [System.UnauthorizedAccessException], but I don't want to use -ErrorAction Stop with catch [System.Management.Automation.ActionPreferenceStopException] either (like here), because I won't get all file paths.
This worked for me
$epath2 = $null
Get-ChildItem -Path C:\Windows\System32 -Recurse -Force -Directory -ErrorVariable epath2 |
foreach {
if ($?) {
Out-File -InputObject $_.FullName -FilePath c:\test\paths.txt -Append
}
}
if ($epath2) {
$epath2 |
foreach {
($_.Exception -split "'")[1] | Out-File -FilePath c:\test\errors.txt -Append
}
}
catch the errors in the -ErrorAction variable and send to errors.txt as a second pass. There may be other errors than access denied but you'll know which folders were showing problems