Powershell Code:
#
# LUNAR DUST - home grown server/service monitor - sends out email when services defined in related .CSV config file are down
# Author: Neal Walters - Nov 2013
# (Lunar Dust is a play on words of the monitor name "Solar Wind")
#
[string[]] $users = "john@abc.com",'"fred@abc.com" # List of users to email your report to (separate by comma)
$fromemail = "FromEmail@abc.com"
$SMTPserver = "YourEmailRelayServer.com" #enter your own SMTP server DNS name / IP address here
$YNTraceSuccess = "N" #setting to "Y" will create more trace/debug to the ServiceTestTrace.txt file, setting to "N" only shows servers/services that were down
$TraceFilename = "D:\scripts\ServiceMonitorTrace.txt"
$LocalServerName = "MyServerName" #used to determine if we should check remote server or not on GetService
set-item trustedhosts $LocalServerName
#Get arguments flexibly, in either order.
#one parms is a "Y" or "N" to indicate to send an email, even when no servers are in error status
#another optional parm is the filename of the CSV to read, if omitted, a default filename is used.
if ($args.Length -gt 0)
{
if ($args[0] -eq "Y" -or $args[0] -eq "N")
{
$IsEmailOn = $args[1]
}
if ($args[0].Length -gt 4)
{
$csvFilename = $args[0]
}
}
if ($args.Length -gt 1)
{
if ($args[1] -eq "Y" -or $args[1] -eq "N")
{
$IsEmailOn = $args[1]
}
if ($args[1].Length -gt 4)
{
$csvFilename = $args[0]
}
}
if ([string]::IsNullOrEmpty($csvFilename))
{
$csvFilename = "D:\Scripts\ServerMonitorConfig.csv"
Write-Host "Setting csfFileName=$csvFileName"
}
Write-Host "csvFilename=$csvFilename"
$csv = Import-Csv $csvFilename -Header @("IsActive","Environment","Category","ServerName","ServiceName","Criticality")
$HTMLMessage=""
$TextMessage=""
$HTMLMessage="<h2>Server/Service Status</h2><table border='1'><tr><th>Environment</th><th>Category</th><th>ServerName</th><th>ServiceName</th><th>Status</th><th>Process Started Date/Time</th><th>UserName</th></tr>"
$CriticalErrorCount = 0
$ErrorCount = 0
$ServerCount = 0
foreach ($line in $csv) {
if ($line.IsActive -eq "Active")
{
$reportStatus = ""
$ServerCount = $ServerCount + 1
#$Service = (get-service -Name $line.ServiceName -ComputerName $line.ServerName)
#this is slower than above, but it gives us the processId which we can use to find out what time the service/process started
write-host "Verifying: " $line.ServerName $line.ServiceName
$myDate = Get-Date
if ($YNTraceSuccess = "Y")
{
Add-Content $TraceFilename "$myDate TRC01 $($line.ServerName) $($line.ServiceName)"
}
$error.clear() #clear any prior errors, otherwise same error may repeat over-and-over in trace
if ($LocalServerName -eq $line.ServerName)
{
# see if not using -ComputerName on local computer avoids the "service not found" error
Add-Content $TraceFilename "$myDate TRCW1 using local computer "
$Service = (get-wmiobject win32_service -filter "name = '$($line.ServiceName)'")
}
else
{
Add-Content $TraceFilename "$myDate TRCW2 using remote computer $($line.ServerName) not eq $LocalServerName"
$Service = (get-wmiobject win32_service -ComputerName $line.ServerName -filter "name = '$($line.ServiceName)'")
}
if ($error -ne $null)
{
Write-Host "----> $($error[0].Exception) "
Add-Content $TraceFilename "$myDate TRCE1 $($error[0].Exception)"
}
if ($Service -eq $null)
{
$reportStatus = "Service Not Found: name = '$($line.ServiceName)'"
$trColor = "Yellow"
$ErrorCount = $ErrorCount + 1
$CriticalErrorCount = $CriticalErrorCount + 1
$CreationDate = "NA"
Write-Host "----> $reportStatus "
Add-Content $TraceFilename "$myDate TRC02 $reportStatus"
}
else
{
#Write-Host "Service Exists"
#$status = $Service.Status
#if ($status -eq "Running") #this was the check when using get-service instead of get-wmiobject win32_service)
$reportStatus = $Service.State
if ($Service.Started -eq "True") {
#$reportStatus = "Up"
$trColor = "White"
# when service is running, then we can lookup the ProcessId to get the Userid and CreationDate (Time the service was started)
$ServicePID = $Service.ProcessID
#Write-Host "Process id: $ServicePID"
$ProcessInfo = Get-WmiObject -Class Win32_Process -ComputerName $line.ServerName -Filter "ProcessID='$ServicePID'" -ea 0
$CreationDate = $ProcessInfo | % { $_.ConvertToDateTime( $_.CreationDate )}
}
else
{
Write-Host "Down Service.Started=$($Service.Started) "
Write-Host "Status=$($Service.Status) State=$($Service.State)"
#$reportStatus = $Service.State
$trColor = "Orange"
$ErrorCount = $ErrorCount + 1
if ($line.Criticality -eq "Error")
{
#switch from orange to yellow background
$trColor = "Yellow"
$CriticalErrorCount = $CriticalErrorCount + 1
}
#Write-Host "down status=$status" #result was empty string or null
$CreationDate = "NA"
}
}
#Write-Host "test=$reportStatus"
$TextMessage += "$($line.Environment) $($line.Category) $($line.ServerName) $($line.ServiceName) $reportStatus $CreationDate $($Service.StartName)`r`n"
#build the TR and TD Cells of the HTML Table
$HTMLMessage += "<tr bgcolor='$trColor'>"
$HTMLMessage += "<td>$($line.Environment)</td>"
$HTMLMessage += "<td>$($line.Category)</td>"
$HTMLMessage += "<td>$($line.ServerName)</td>"
$HTMLMessage += "<td>$($line.ServiceName)</td>"
$HTMLMessage += "<td>$reportStatus</td>"
$HTMLMessage += "<td>$CreationDate</td>"
$HTMLMessage += "<td>$($Service.StartName)</td>"
$HTMLMessage += "</tr>`r`n"
}
else
{
Write-Host "Skipping InActive " $line.ServerName $line.ServiceName
}
}
Write-Host '------'
#$HTMLMessage = "<h3>Critical Server Count=$CriticalErrorCount Total Error Count=$ErrorCount</h3>" + $HTMLMessage + "</table>" + "<h3>Yellow is critical, Orange is not critical.</h3>"
$HTMLMessage = "<h3>Server-Count=$ServerCount Critical-Count=$CriticalErrorCount Total-Error-Count=$ErrorCount</h3>$HTMLMessage</table><h3>Yellow is critical, Orange is not critical.</h3>"
$mydate = Get-Date
Write-Host "Date=$myDate"
$HTMLMessage = "<h3>$mydate</h3>$HTMLMessage"
Write-Host $TextMessage
Write-Host "`r`n Server-Count=$ServerCount Critical-Count=$CriticalErrorCount Total-Error-Count=$ErrorCount"
Write-Host "Date=$myDate"
$emailSubject = "QT Service Email was Requested"
if ($CriticalErrorCount -gt 0)
{
$emailSubject = "Critical QT Server Down Alert: There are $CriticalErrorCount critical services down"
}
#always send email when one 1 or more critical errors are found, or when the $IsEmailOn parm is set to "Always" which sends email regardless of error count
if ($CriticalErrorCount -gt 0 -or $IsEmailOn -eq "Always")
{
send-mailmessage -from $fromemail -to $users -subject $emailSubject -BodyAsHTML -body $HTMLMessage -priority High -smtpServer $SMTPserver
Write-Host "Alert Email Sent with Subject=$emailSubject"
Add-Content $TraceFilename "$myDate TRC99 Alert Email Sent with Subject=$emailSubject"
}
Create a CSV like this, with a list of the servers to monitor.
Column 1
Column 2 is an arbitrary environment name. Column 3 is a category of server functionality (could potentially be used to route email to the group in charge of those types of servers, e.g. IIS, BizTalk, SQL…)
Column 3 is Server/Machine Name.
Column 4 is the Windows Service name. Use “LanmanServer” just to know if the machine is up and running.
Column 5 indicates whether this should cause a critical error or just a warning (when the service is down).
"Active","PROD","SQL","SqlServer01","MSSQLSERVER","Error" "Active","PROD","SQL","SqlServer01","SQLSERVERAGENT","Error" "Active","PROD","SQL","SqlServer01","MSDTC","Error","Error" "Active","PROD","BizTalk","BizTalkServer01","MSDTC","Error" "Active","QA","BizTalk","BizTalkServer03","LanmanServer","Error" "Active","QA","BizTalk","BizTalkServer03","MSDTC","Error" "Active","QA","BizTalk","BizTalkServer03","BTSSvc$Application_Default","Error"
Schedule a .bat or .cmd file such as follows, pointing to the desired CSV as defined above.
powershell -command "& 'D:\Scripts\ServerMonitor.ps1'" d:\Scripts\ServerMonitorConfig.csv