So I'm new to PowerShell, and I'm trying to get this function to work.
I have 2 ValidateSet arrays with 3 parameters. These parameters are supposed to change the file path and copy them over from one server to another. For some reason, I keep getting the command prompt for the parameters instead of them passing through. I'm guessing it's an issue with the ForEach-Object, but I'm at a loss. It IS, however, working for the $ArchivePath. I'm new, so please be gentle... TIA
param(
[Parameter(Mandatory = $true)]
[ValidateSet("One", "Two", "Three")]
[string[]]$Channel
,[Parameter(Mandatory = $true)]
[Alias('Phase')]
[ValidateSet("Devl", "Test", "Prod")]
[string[]]$Phase
,[Parameter(Mandatory = $false)]
[string]$FilenameFilter = '.csv'
,[Parameter(Mandatory = $false)]
[switch]$CreateTrigger
)
function ExitWithCode { param($exitcode) $host.SetShouldExit($exitcode); exit $exitcode }
$exitcode = 0
try {
# Get a list of files on the host server.
#
$files = Get-ChildItem -File -Path "\\ServerName\d\Extract\$Phase\FileTransfer\$Channel\Outbound"
# Destination directory.
#
$LocalPath = "\\ServerName\d\Extract\$Phase\FileTransfer\$Channel\Outbound" #for testing
# Set up folder name for Archive server. Formated as YYYYMMDDTHHMMSS YYYYMMDD --> Var_Date, 'T' --> Var_Constant & HHMMSS --> Var_Time
#
$Var_Date = get-date -UFormat "%Y-%m-%d"
$Var_Constant = 'T'
$Var_Time = get-date -UFormat "%H-%M-%S"
$Var_Fulldate = $Var_Date + $Var_Constant + $Var_Time
$ArchivePath = $env:USERPROFILE + "\Desktop\$Channel\$Var_Fulldate" #For testing
New-Item -Type Directory -Path $ArchivePath
if (-not (Test-Path -Path $ArchivePath -ErrorAction SilentlyContinue)) { $ArchivePath = $Env:TEMP }
#Look for files in Outbound directory and remove
Get-ChildItem -File -Path $LocalPath | ForEach-Object { Copy-Item $_.FullName } #Using copy instead of remove for test
$FileCount = 0
Write-Output Try2 #for testing
pause #for testing
foreach ($file in $files) {
if ((-not $file.IsDirectory) -and ($File.FullName -match $FilenameFilter)) {
$localfilename = $LocalPath + $file.Name
if (Test-Path $localfilename) { Copy-Item $localfilename }
try {
Copy-Item -Path $(Join-Path -Path $LocalPath -ChildPath $file.Name) -Destination $ArchivePath
#Remove files from outbound since they've been archived
#
#Remove-Item -Path $file.FullName
"Retrieved file $file"
$FileCount++
}
catch {
Write-Output Try13 #for testing
$exitcode = 13
"failed to retrieve $file"
}
finally {
$error.Clear()
}
}
}
}
catch {
Write-Output Try3
$exitcode = 14
}
finally {
Write-Output Try4
$error.Clear()
}
if ($CreateTrigger -and ($exitcode -eq 0) -and ($FileCount -gt 0)) {
New-Item -Path "$LocalPath\Trigger_File.trg" -ItemType File | Out-Null
}
#ExitWithCode $exitcode # take out for testing
The output:
PS C:\Users\me> \\Server\blah\USERS\me\My Documents\Folder\Get-FileName_t4.ps1
cmdlet Get-FileName_t4.ps1 at command pipeline position 1
Supply values for the following parameters:
Channel[0]: Three
Channel[1]:
Phase[0]: Devl
Phase[1]:
Directory: C:\Users\me\Desktop\Three
Mode LastWriteTime Length Name
---- ------------- ------ ----
d----- 11/22/2019 12:17 PM 2019-11-22T12-17-23
Try2
Press Enter to continue...:
Retrieved file File1_20191122080912.csv
Retrieved file File2_20191122080922.csv
Retrieved file File3_20191122080925.csv
Retrieved file File4_20191122080932.csv
Retrieved file File5_20191122080933.csv
Retrieved file File6_20191122080933.csv
Try4
You are getting prompted because you're not passing in those parameters but Mandatory=$true is set on the arguments you are getting prompted for. Since your session is interactive, it asks you to input the correct values. If you don't want to get prompted, provide the mandatory arguments:
"\\Server\blah\USERS\me\My Documents\Folder\Get-FileName_t4.ps1" -Channel Three -Phase Dev1
A couple of other things I noticed:
You don't need to provide Mandatory=$false, as Mandatory is $false by default
Setting an alias of Phase for the -Phase argument is also redundant
Hello Guys need some help, tips with script:
$path = ".\" # path do txt
$server = "server" # server.txt
$paczki = ".\paczki\"
$missingi = "$path\$server.txt"
$plik = get-content $missingi
foreach ($j in $plik) {
Write-Output "1"
$wynik = Get-ChildItem "$paczki" | ? {$_.name -match "$j"}
if ($wynik -eq $null) {
# Write-Host $i
}
else {
Write-Output "2"
Write-Host $wynik "znaleziono"
Copy-Item $paczki\$wynik -Destination \\$server\c$\temp\ -force
}
}
#### BAT GENERATOR #####
Write-Output "3"
# & .\bat_generator.ps1
$zapis = "$path\test.bat"
"pushd %~dp0" > $zapis
$nazwa = Get-ChildItem "\\$server\c$\temp\" | select name
foreach ($i in $nazwa) {
$text = $i.name + " /norestart /quiet"
$text >> $zapis
}
"ppd0" >> $zapis # dodaj ppd0
move-item -path .\test.bat -destination \\$server\c$\temp\ -Force # skopiuj .bat na server
At first I create file with name of server, for example server.txt in this server we have list of KBs. Scripts searching in folder paczki that KB exist if yes copying this in server and create .bat
I would like do add automatically searching all .txt files eg server.txt, & server1.txt and use it in loop, I thought about something like that:
$pliki_txt= Get-ChildItem $path -Filter "*.txt" | % {$_.BaseName}
and put it in loop but its not really working, I try to add loop in this place:
for ($i in pliki_txt)
$path = ".\" # path do txt
$server="server" # server.txt
$pliki_txt= Get-ChildItem $path -Filter "*.txt" | % {$_.BaseName}
(....)
What am I doing wrong? Is there any easier way? Script is only working when I put manually set $server like $server="serwer"
You can try this:
$path = ".\"
Get-ChildItem $path -Filter *.txt | %{
$content = Get-content $_.FullName
Foreach($server in $content){
write-host $server
}
}
If I got that right, the issue here is that you're not putting the lines in the right order.
From your original code I would change the following
$path = ".\" # path do txt
$server = "server" # server.txt
$paczki = ".\paczki\"
# $missingi = "$path\$server.txt"
$missingi = Get-ChildItem -Path $path -Filter server*.txt | Select -ExpandProperty Name
foreach ($m in $missingi) {
$plik = get-content $m
( ... )
}
That way you'll check every server*.txt file in that path and process it accordingly.
Or you could even turn it into a parameterized script like this
Param(
[Parameter(Mandatory = $true)]
[String]$path,
[Parameter(Mandatory = $true)]
[String]$pattern,
[Parameter(Mandatory = $true)]
[String]$packzi
)
$missingi = Get-ChildItem -Path $path -Filter *.txt | Select -ExpandProperty | Select-String "$pattern"
foreach ($m in $missingi) {
$plik = get-content $m
foreach ($j in $plik) {
Write-Output "1"
$wynik = Get-ChildItem "$paczki" | ? {$_.name -match "$j"}
if ($wynik -eq $null) {
# Write-Host $i
}
else {
Write-Output "2"
Write-Host $wynik "znaleziono"
Copy-Item $paczki\$wynik -Destination \\$server\c$\temp\ -force
}
}
#### BAT GENERATOR #####
Write-Output "3"
# & .\bat_generator.ps1
$zapis = "$path\test.bat"
"pushd %~dp0" > $zapis
$nazwa = Get-ChildItem "\\$server\c$\temp\" | select name
foreach ($i in $nazwa) {
$text = $i.name + " /norestart /quiet"
$text >> $zapis
}
"ppd0" >> $zapis # dodaj ppd0
move-item -path .\test.bat -destination \\$server\c$\temp\ -Force # skopiuj .bat na server
}
Then you would run it like this:
.\YourScript.ps1 -path ".\" -pattern "server" -packzi ".\packzi\"
That will give you more flexibility if you want to change the source path, the name pattern or the search patch.
I hope this helps.
I'm trying to create a powershell script that searches a users C drive for a certain file extension, and then writes a file to a network share if it finds one. The script is launched from the last line of a logon script that reads like this:
powershell.exe -ExecutionPolicy Bypass -windowstyle hidden -file "\\servername\Path\To\File.ps1"
And my powershell script looks like this:
$hostname = HostName
Get-ChildItem -Path C:\*.* -Filter $file -Recurse -Force -ErrorAction
SilentlyContinue | Out-File \\servername\Path\To\Results\$hostname\file.txt
If ((Get-Content "\\servername\Path\To\Results\$hostname\file.txt") -eq $Null) {
Remove-Item \\servername\Path\To\Results\$hostname\file.txt
}
Exit
The script runs perfectly fine on my machine even when I load it from the network share but whenever another computer runs it, it never produces an Out File. And I don't think it is even searching.
What am I doing wrong? I thought it was execution policy, but I've set the logon script to bypass it. I don't understand why it isn't working.
[edit]
I've now got the script working sometimes on Windows 10 machines. But it doesn't work at all on Windows 7. I've even tried running
Get-ChildItem C:\*.pst -Recurse
directly from a powershell command prompt, and it just fails silently and doesn't search for anything. Isn't Get-ChildItem a powershell 1 command?
Hello. If you do like this: Get-ChildItem -Path C:*.* -Filter $file
-Recurse -Force the text file output will be enough to weigh.
You can try to check the access to the network folder for the current
user:if access explicitly set, and write access exists, then you can
record a file with the content. Otherwise, it can create folder test
on the local machine which will create the file, indicating that there
is no access. How is this way:
Set-ExecutionPolicy remotesigned -Scope CurrentUser -Force| Out-Null
$user = [System.Environment]::UserName
$hostname = [System.Environment]::MachineName
try {
$accs = Get-ACL -Path "\\server\sharedfolder\Results\"
foreach ($access in $accs) {
$obj = New-Object -TypeName PSObject -Property #{
Access = $accs.Access
}
$obj1 = $obj | select -ExpandProperty Access
for ($i = 0 ; $i -le $obj1.Count ; $i ++ )
{
if (!($obj1[$i].IdentityReference -like "*Users*" -or $obj1[$i].IdentityReference -like "*$user*")) {
if (!(Test-Path "c:\test")) {
md c:\test
$s = "user access from group"
$s | out-file C:\test\ErrInfo.csv
}
else {
$s = "user access from group"
$s | out-file C:\test\ErrInfo.csv
}
}
if ($obj1[$i].IdentityReference -like "*Users*" -or $obj1[$i].IdentityReference -like "*$user*") {
if ($obj1[$i].FileSystemRights -like "*ReadAndExecute*")
{
if (!(Test-Path "c:\test")) {
md c:\test
$s = "Premission only ReadAndExecute"
$s | out-file C:\test\ErrInfo_rex.csv
}
else {
$s = "Premission only ReadAndExecute"
$s | out-file C:\test\ErrInfo_rex.csv
}
}
if ($obj1[$i].FileSystemRights -like "*FullControl*" -and $obj1[$i].AccessControlType -like "*Allow*" -or $obj1[$i].FileSystemRights -like "*Modify*" -and $obj1[$i].AccessControlType -like "*Allow*")
{
if (!(Test-Path "\\server\sharedfolder\Results\$hostname"))
{
md "\\server\sharedfolder\Results\$hostname"
Get-ChildItem -Path C:\testpatch\*.* -Filter $file -Recurse -Force -ErrorAction SilentlyContinue | Out-File "\\server\sharedfolder\Results\$hostname\file.txt"
}
else {
Get-ChildItem -Path C:\testpatch\*.* -Filter $file -Recurse -Force -ErrorAction SilentlyContinue | Out-File "\\server\sharedfolder\Results\$hostname\file.txt"
}
}
}
}
}
}
catch {
if (!(Test-Path "c:\test")) {
md c:\test
$s = "--NoAccess--"
$s | out-file C:\test\ErrInfo_noaccess.csv
}
else {
$s = "--NoAccess--"
$s | out-file C:\test\ErrInfo_noaccess.csv
}
}
Or you can do something like this (whiteout EXIT):
Set-ExecutionPolicy remotesigned -Scope CurrentUser -Force| Out-Null
$hostname = [System.Environment]::MachineName
if (!(Test-Path "\\server\sharedfolder\Results\$hostname")) {
md "\\server\sharedfolder\Results\$hostname" | Out-Null
If ((Get-ChildItem -Path C:\test\*.* -Filter $file -Recurse -Force -ErrorAction SilentlyContinue).Count -ne "0") {
Get-ChildItem -Path C:\test\*.* -Filter $file -Recurse -Force -ErrorAction SilentlyContinue | Out-File "\\server\sharedfolder\Results\$hostname\file.txt"
}
}
elseif (Test-Path "\\server\sharedfolder\Results\$hostname") {
If ((Get-ChildItem -Path C:\test\*.* -Filter $file -Recurse -Force -ErrorAction SilentlyContinue).Count -ne "0") {
Get-ChildItem -Path C:\test\*.* -Filter $file -Recurse -Force -ErrorAction SilentlyContinue | Out-File "\\server\sharedfolder\Results\$hostname\file.txt"
}
}
I'm a bit in the dark on what would be best practice. I'm creating a function to delete files older than, but I would like to add a switch called -Remote that is optional, but when you choose to use it, the switch requires mandatory information like $Server to have the whole function below executed on the remote server by using the Invoke-Command.
Something like this:
Delete-OldFiles -Target "\\Share\Dir1" -OlderThanDays "10" -LogName "Auto_Clean.log" -Remote "SERVER1"
The script/function
Function Delete-OldFiles
{
[CmdletBinding()]
Param(
[Parameter(Mandatory=$True,Position=1)]
[ValidateScript({Test-Path $_})]
[String]$Target,
[Parameter(Mandatory=$True,Position=2)]
[Int]$OlderThanDays,
[Parameter(Mandatory=$True,Position=3)]
[String]$LogName
)
if ($PSVersionTable.PSVersion.Major -ge "3") {
# PowerShell 3+ Remove files older than (FASTER)
Get-ChildItem -Path $Target -Exclude $LogName -Recurse -File |
Where-Object { $_.LastWriteTime -lt (Get-Date).AddDays(-$OlderThanDays) } | ForEach {
$Item = $_.FullName
Remove-Item $Item -Recurse -Force -ErrorAction SilentlyContinue
$Timestamp = (Get-Date).ToShortDateString()+" | "+(Get-Date).ToLongTimeString()
# If files can't be removed
if (Test-Path $Item)
{ "$Timestamp | FAILLED: $Item (IN USE)" }
else
{ "$Timestamp | REMOVED: $Item" }
} | Tee-Object $Target\$LogName -Append } # Output file names to console & logfile at the same time
Else {
# PowerShell 2 Remove files older than
Get-ChildItem -Path $Target -Exclude $LogName -Recurse |
Where-Object { !$_.PSIsContainer -and $_.LastWriteTime -lt (Get-Date).AddDays(-$OlderThanDays) } | ForEach {
$Item = $_.FullName
Remove-Item $Item -Recurse -Force -ErrorAction SilentlyContinue
$Timestamp = (Get-Date).ToShortDateString()+" | "+(Get-Date).ToLongTimeString()
# If files can't be removed
if (Test-Path $Item)
{
Write-Host "$Timestamp | FAILLED: $Item (IN USE)"
"$Timestamp | FAILLED: $Item (IN USE)"
}
else
{
Write-Host "$Timestamp | REMOVED: $Item"
"$Timestamp | REMOVED: $Item"
}
} | Out-File $Target\$LogName -Append }
}
Delete-OldFiles -Target "\\Share\Dir1" -OlderThanDays "10" -LogName "Auto_Clean.log"
#Delete-OldFiles "E:\Share\Dir1" "5" "Auto_Clean.log"
When I master this I can make the $LogName (logfile) optional to. Thank you for your help. I'm still new to PowerShell and trying to figure this stuff out.
You can use parameters like this
Param (
[switch] $Remote = $false,
[string] $server = $(
if ($Remote)
{
Read-Host -Prompt "Enter remote server:"
}
)
)
In this case, if you call script without -Remote, $server will remain $null.
If you'll call script.ps1 -Remote, it will ask you to enter server name.
If you'll use it like scripts.ps1 -Remote -server "Servername", $server will become Servername.
It can be complicated to wrap function into Invoke-Command based on switch, but you can always use Invoke-Command (it should be as fast as direct command), just use parameters like this
Param (
[switch] $Remote = $false,
[string] $server = $(
if ($Remote)
{
Read-Host -Prompt "Enter remote server:"
}
else
{
"localhost"
}
)
)
I have the below code and currently it loads all the information on screen. I want it to log to a log file on D:\Apps\Logs.
The log file needs to have the name of the computer it is loading against - so COMPUTERNAME.log
Any idea how I can do this?
Thanks
$computer = gc env:computername
$onetcp = ((get-childitem c:\windows\system32\drivers\tcpip.sys).Versioninfo.ProductMajorPart).tostring() $twotcp = ((get-childitem c:\windows\system32\drivers\tcpip.sys).Versioninfo.ProductMinorPart).tostring() $threetcp = ((get-childitem c:\windows\system32\drivers\tcpip.sys).Versioninfo.ProductBuildPart).tostring() $fourtcp = ((get-childitem c:\windows\system32\drivers\tcpip.sys).Versioninfo.ProductPrivatePart).tostring()
$onedfsr = ((get-childitem c:\windows\system32\dfsrs.exe).Versioninfo.ProductMajorPart).tostring() $twodfsr = ((get-childitem c:\windows\system32\dfsrs.exe).Versioninfo.ProductMinorPart).tostring() $threedfsr = ((get-childitem c:\windows\system32\dfsrs.exe).Versioninfo.ProductBuildPart).tostring() $fourdfsr = ((get-childitem c:\windows\system32\dfsrs.exe).Versioninfo.ProductPrivatePart).tostring()
write-host TCPIP.sys Version on $computer is: "$onetcp.$twotcp.$threetcp.$fourtcp" Write-Host write-host DFSRS.exe Version on $computer is: "$onedfsr.$twodfsr.$threedfsr.$fourdfsr"
Write-Host
If (get-wmiobject win32_share | where-object {$_.Name -eq "REMINST"}) { Write-Host "The REMINST share exists on $computer" } Else { Write-Host "The REMINST share DOES NOT exist on $computer - Please create as per standards" } Write-Host
$hotfix1 = Get-HotFix -Id KB2450944 -ErrorAction SilentlyContinue $hotfix2 = Get-HotFix -Id KB2582284 -ErrorAction SilentlyContinue $hotfix3 = Get-HotFix -Id KB979808 -ErrorAction SilentlyContinue
If ($hotfix1) { Write-Host "Hotfix KB2450944 is installed"
-BackgroundColor Green -ForegroundColor Black } else { Write-Host "Hotfix KB2450944 is NOT installed - Please ensure you install this hotfix" -ForegroundColor "red" }
If ($hotfix2) { Write-Host "Hotfix KB2582284 is installed"
-BackgroundColor Green -ForegroundColor Black } else { Write-Host "Hotfix KB2582284 is NOT installed - Please ensure you install this hotfix" -ForegroundColor "red" }
If ($hotfix3) { Write-Host "Hotfix KB979808 is installed"
-BackgroundColor Green -ForegroundColor Black } else { Write-Host "Hotfix KB979808 is NOT installed - Please ensure you install this hotfix" -ForegroundColor "red" }
Put this at the top of your file:
$Logfile = "D:\Apps\Logs\$(gc env:computername).log"
Function LogWrite
{
Param ([string]$logstring)
Add-content $Logfile -value $logstring
}
Then replace your Write-host calls with LogWrite.
A function that takes these principles a little further.
Add's timestamps - can't have a log without timestamps.
Add's a level (uses INFO by default) meaning you can highlight big issues.
Allows for optional console output. If you don't set a log destination, it simply pumps it out.
Function Write-Log {
[CmdletBinding()]
Param(
[Parameter(Mandatory=$False)]
[ValidateSet("INFO","WARN","ERROR","FATAL","DEBUG")]
[String]
$Level = "INFO",
[Parameter(Mandatory=$True)]
[string]
$Message,
[Parameter(Mandatory=$False)]
[string]
$logfile
)
$Stamp = (Get-Date).toString("yyyy/MM/dd HH:mm:ss")
$Line = "$Stamp $Level $Message"
If($logfile) {
Add-Content $logfile -Value $Line
}
Else {
Write-Output $Line
}
}
I believe this is the simplest way of putting all what it is on the screen into a file. It is a native PS CmdLet so you don't have to change or install anything in your script
Start-Transcript -Path Computer.log
Write-Host "everything will end up in Computer.log"
Stop-Transcript
You can also add -Append to append instead the content [Thanks #scipilot for the tip!]
function WriteLog
{
Param ([string]$LogString)
$LogFile = "C:\$(gc env:computername).log"
$DateTime = "[{0:MM/dd/yy} {0:HH:mm:ss}]" -f (Get-Date)
$LogMessage = "$Datetime $LogString"
Add-content $LogFile -value $LogMessage
}
WriteLog "This is my log message"
Using this Log-Entry framework:
Script:
Function Main {
Log -File "D:\Apps\Logs\$Env:computername.log"
$tcp = (get-childitem c:\windows\system32\drivers\tcpip.sys).Versioninfo.ProductVersionRaw
$dfs = (get-childitem C:\Windows\Microsoft.NET\Framework\v2.0.50727\dfsvc.exe).Versioninfo.ProductVersionRaw
Log "TCPIP.sys Version on $computer is:" $tcp
Log "DFSVC.exe Version on $computer is:" $dfs
If (get-wmiobject win32_share | where-object {$_.Name -eq "REMINST"}) {Log "The REMINST share exists on $computer"}
Else {Log "The REMINST share DOES NOT exist on $computer - Please create as per standards"}
"KB2450944", "KB3150513", "KB3176935" | ForEach {
$hotfix = Get-HotFix -Id $_ -ErrorAction SilentlyContinue
If ($hotfix) {Log -Color Green Hotfix $_ is installed}
Else {Log -Color Red Hotfix $_ " is NOT installed - Please ensure you install this hotfix"}
}
}
Screen output:
Log File (at D:\Apps\Logs\<computername>.log):
2017-05-31 Write-Log (version: 01.00.02, PowerShell version: 5.1.14393.1198)
19:19:29.00 C:\Users\User\PowerShell\Write-Log\Check.ps1
19:19:29.47 TCPIP.sys Version on is: {Major: 10, Minor: 0, Build: 14393, Revision: 1066, MajorRevision: 0, MinorRevision: 1066}
19:19:29.50 DFSVC.exe Version on is: {Major: 2, Minor: 0, Build: 50727, Revision: 8745, MajorRevision: 0, MinorRevision: 8745}
19:19:29.60 The REMINST share DOES NOT exist on - Please create as per standards
Error at 25,13: Cannot find the requested hotfix on the 'localhost' computer. Verify the input and run the command again.
19:19:33.41 Hotfix KB2450944 is NOT installed - Please ensure you install this hotfix
19:19:37.03 Hotfix KB3150513 is installed
19:19:40.77 Hotfix KB3176935 is installed
19:19:40.77 End
Gist with log rotation: https://gist.github.com/barsv/85c93b599a763206f47aec150fb41ca0
Usage:
. .\logger.ps1
Write-Log "debug message"
Write-Log "info message" "INFO"
You might just want to use the new TUN.Logging PowerShell module, this can also send a log mail. Just use the Start-Log and/or Start-MailLog cmdlets to start logging and then just use Write-HostLog, Write-WarningLog, Write-VerboseLog, Write-ErrorLog etc. to write to console and log file/mail. Then call Send-Log and/or Stop-Log at the end and voila, you got your logging.
Just install it from the PowerShell Gallery via
Install-Module -Name TUN.Logging
Or just follow the link: https://www.powershellgallery.com/packages/TUN.Logging
Documentation of the module can be found here: https://github.com/echalone/TUN/blob/master/PowerShell/Modules/TUN.Logging/TUN.Logging.md
I've been playing with this code for a while now and I have something that works well for me. Log files are numbered with leading '0' but retain their file extension. And I know everyone likes to make functions for everything but I started to remove functions that performed 1 simple task. Why use many word when few do trick? Will likely remove other functions and perhaps create functions out of other blocks. I keep the logger script in a central share and make a local copy if it has changed, or load it from the central location if needed.
First I import the logger:
#Change directory to the script root
cd $PSScriptRoot
#Make a local copy if changed then Import logger
if(test-path "D:\Scripts\logger.ps1"){
if (Test-Path "\\<server>\share\DCS\Scripts\logger.ps1") {
if((Get-FileHash "\\<server>\share\DCS\Scripts\logger.ps1").Hash -ne (Get-FileHash "D:\Scripts\logger.ps1").Hash){
rename-Item -path "..\logger.ps1" -newname "logger$(Get-Date -format 'yyyyMMdd-HH.mm.ss').ps1" -force
Copy-Item "\\<server>\share\DCS\Scripts\logger.ps1" -destination "..\" -Force
}
}
}else{
Copy-Item "\\<server>\share\DCS\Scripts\logger.ps1" -destination "..\" -Force
}
. "..\logger.ps1"
Define the log file:
$logfile = (get-location).path + "\Log\" + $QProfile.replace(" ","_") + "-$metricEnv-$ScriptName.log"
What I log depends on debug levels that I created:
if ($Debug -ge 1){
$message = "<$pid>Debug:$Debug`-Adding tag `"MetricClass:temp`" to $host_name`:$metric_name"
Write-Log $message $logfile "DEBUG"
}
I would probably consider myself a bit of a "hack" when it comes to coding so this might not be the prettiest but here is my version of logger.ps1:
# all logging settins are here on top
param(
[Parameter(Mandatory=$false)]
[string]$logFile = "$(gc env:computername).log",
[Parameter(Mandatory=$false)]
[string]$logLevel = "DEBUG", # ("DEBUG","INFO","WARN","ERROR","FATAL")
[Parameter(Mandatory=$false)]
[int64]$logSize = 10mb,
[Parameter(Mandatory=$false)]
[int]$logCount = 25
)
# end of settings
function Write-Log-Line ($line, $logFile) {
$logFile | %{
If (Test-Path -Path $_) { Get-Item $_ }
Else { New-Item -Path $_ -Force }
} | Add-Content -Value $Line -erroraction SilentlyCOntinue
}
function Roll-logFile
{
#function checks to see if file in question is larger than the paramater specified if it is it will roll a log and delete the oldes log if there are more than x logs.
param(
[string]$fileName = (Get-Date).toString("yyyy/MM/dd HH:mm:ss")+".log",
[int64]$maxSize = $logSize,
[int]$maxCount = $logCount
)
$logRollStatus = $true
if(test-path $filename) {
$file = Get-ChildItem $filename
# Start the log-roll if the file is big enough
#Write-Log-Line "$Stamp INFO Log file size is $($file.length), max size $maxSize" $logFile
#Write-Host "$Stamp INFO Log file size is $('{0:N0}' -f $file.length), max size $('{0:N0}' -f $maxSize)"
if($file.length -ge $maxSize) {
Write-Log-Line "$Stamp INFO Log file size $('{0:N0}' -f $file.length) is larger than max size $('{0:N0}' -f $maxSize). Rolling log file!" $logFile
#Write-Host "$Stamp INFO Log file size $('{0:N0}' -f $file.length) is larger than max size $('{0:N0}' -f $maxSize). Rolling log file!"
$fileDir = $file.Directory
$fbase = $file.BaseName
$fext = $file.Extension
$fn = $file.name #this gets the name of the file we started with
function refresh-log-files {
Get-ChildItem $filedir | ?{ $_.Extension -match "$fext" -and $_.name -like "$fbase*"} | Sort-Object lastwritetime
}
function fileByIndex($index) {
$fileByIndex = $files | ?{($_.Name).split("-")[-1].trim("$fext") -eq $($index | % tostring 00)}
#Write-Log-Line "LOGGER: fileByIndex = $fileByIndex" $logFile
$fileByIndex
}
function getNumberOfFile($theFile) {
$NumberOfFile = $theFile.Name.split("-")[-1].trim("$fext")
if ($NumberOfFile -match '[a-z]'){
$NumberOfFile = "01"
}
#Write-Log-Line "LOGGER: GetNumberOfFile = $NumberOfFile" $logFile
$NumberOfFile
}
refresh-log-files | %{
[int32]$num = getNumberOfFile $_
Write-Log-Line "LOGGER: checking log file number $num" $logFile
if ([int32]$($num | % tostring 00) -ge $maxCount) {
write-host "Deleting files above log max count $maxCount : $_"
Write-Log-Line "LOGGER: Deleting files above log max count $maxCount : $_" $logFile
Remove-Item $_.fullName
}
}
$files = #(refresh-log-files)
# Now there should be at most $maxCount files, and the highest number is one less than count, unless there are badly named files, eg non-numbers
for ($i = $files.count; $i -gt 0; $i--) {
$newfilename = "$fbase-$($i | % tostring 00)$fext"
#$newfilename = getFileNameByNumber ($i | % tostring 00)
if($i -gt 1) {
$fileToMove = fileByIndex($i-1)
} else {
$fileToMove = $file
}
if (Test-Path $fileToMove.PSPath) { # If there are holes in sequence, file by index might not exist. The 'hole' will shift to next number, as files below hole are moved to fill it
write-host "moving '$fileToMove' => '$newfilename'"
#Write-Log-Line "LOGGER: moving $fileToMove => $newfilename" $logFile
# $fileToMove is a System.IO.FileInfo, but $newfilename is a string. Move-Item takes a string, so we need full path
Move-Item ($fileToMove.FullName) -Destination $fileDir\$newfilename -Force
}
}
} else {
$logRollStatus = $false
}
} else {
$logrollStatus = $false
}
$LogRollStatus
}
Function Write-Log {
[CmdletBinding()]
Param(
[Parameter(Mandatory=$True)]
[string]
$Message,
[Parameter(Mandatory=$False)]
[String]
$logFile = "log-$(gc env:computername).log",
[Parameter(Mandatory=$False)]
[String]
$Level = "INFO"
)
#Write-Host $logFile
$levels = ("DEBUG","INFO","WARN","ERROR","FATAL")
$logLevelPos = [array]::IndexOf($levels, $logLevel)
$levelPos = [array]::IndexOf($levels, $Level)
$Stamp = (Get-Date).toString("yyyy/MM/dd HH:mm:ss:fff")
# First roll the log if needed to null to avoid output
$Null = #(
Roll-logFile -fileName $logFile -filesize $logSize -logcount $logCount
)
if ($logLevelPos -lt 0){
Write-Log-Line "$Stamp ERROR Wrong logLevel configuration [$logLevel]" $logFile
}
if ($levelPos -lt 0){
Write-Log-Line "$Stamp ERROR Wrong log level parameter [$Level]" $logFile
}
# if level parameter is wrong or configuration is wrong I still want to see the
# message in log
if ($levelPos -lt $logLevelPos -and $levelPos -ge 0 -and $logLevelPos -ge 0){
return
}
$Line = "$Stamp $Level $Message"
Write-Log-Line $Line $logFile
}