Event ID 129: Disk I/O Timeout

# Search for Event ID 129 and related storage events
$events = Get-WinEvent -FilterHashtable @{
    LogName = 'System'
    Id = 129, 153, 51, 11, 15, 9
    StartTime = (Get-Date).AddDays(-7)
} -ErrorAction SilentlyContinue | Sort-Object TimeCreated -Descending

Write-Host "=== Storage Events (Last 7 Days) ===" -ForegroundColor Cyan
$events | Group-Object Id | Select-Object @{N='EventID';E={$_.Name}}, Count |
    Format-Table -AutoSize

# Show recent Event ID 129 details
Write-Host "`n=== Recent Event ID 129 Details ===" -ForegroundColor Yellow
$events | Where-Object {$_.Id -eq 129} | Select-Object -First 10 |
    ForEach-Object {
        Write-Host "Time: $($_.TimeCreated)"
        Write-Host "Device: $($_.Properties[0].Value)"0].Value)"
        Write-Host "---"
    }

# Check for correlation with Exchange events
$exchangeEvents = Get-WinEvent -FilterHashtable @{
    LogName = 'Application'
    ProviderName = 'MSExchangeIS', 'ESE'
    Level = 2, 3  # Error, Warning
    StartTime = (Get-Date).AddDays(-7)
} -ErrorAction SilentlyContinue

Write-Host "`n=== Exchange Storage-Related Events ===" -ForegroundColor Yellow
$exchangeEvents | Where-Object {$_.Message -match "I/O|disk|storage|write|read"} |
    Select-Object TimeCreated, Id, Message | Select-Object -First 10 |
    Format-Table -AutoSize -Wrap

# List all disks and their details
Get-Disk | Select-Object Number, FriendlyName, SerialNumber, Size, PartitionStyle,
    OperationalStatus, HealthStatus | Format-Table -AutoSize

# Get physical disk details
Get-PhysicalDisk | Select-Object FriendlyName, SerialNumber, MediaType, Size,
    HealthStatus, OperationalStatus, SpindleSpeed | Format-Table -AutoSize

# Map disks to volumes and Exchange databases
Get-Volume | Where-Object {$_.DriveLetter} | ForEach-Object {
    $vol = $_
    $partition = Get-Partition -DriveLetter $vol.DriveLetter -ErrorAction SilentlyContinue
    $disk = if ($partition) { Get-Disk -Number $partition.DiskNumber } else { $null }

    [PSCustomObject]@{
        DriveLetter = $vol.DriveLetter
        Label = $vol.FileSystemLabel
        SizeGB = [math]::Round($vol.Size/1GB,2)
        DiskNumber = $partition.DiskNumber
        DiskFriendlyName = $disk.FriendlyName
        HealthStatus = $disk.HealthStatus
    }
} | Format-Table -AutoSize

# Check which databases are on which volumes
Get-MailboxDatabase | Select-Object Name,
    @{N='DbVolume';E={(Split-Path $_.EdbFilePath.PathName).Substring(0,2)}},
    @{N='LogVolume';E={(Split-Path $_.LogFolderPath.PathName).Substring(0,2)}} |
    Format-Table -AutoSize

# Real-time disk latency monitoring
$server = $env:COMPUTERNAME
$duration = 60  # seconds
$interval = 5   # seconds

Write-Host "Monitoring disk latency for $duration seconds..." -ForegroundColor Cyan

$counters = @(
    "\$server\PhysicalDisk(*)Avg. Disk sec/Read",
    "\$server\PhysicalDisk(*)Avg. Disk sec/Write",
    "\$server\PhysicalDisk(*)Current Disk Queue Length",
    "\$server\PhysicalDisk(*)% Disk Time"
)

$samples = Get-Counter -Counter $counters -SampleInterval $interval -MaxSamples ($duration/$interval)

# Analyze results
$results = @{}
foreach ($sample in $samples) {
    foreach ($cs in $sample.CounterSamples) {
        if ($cs.InstanceName -ne "_total") {
            $key = "$($cs.InstanceName)_$($cs.Path.Split('')[-1])"$cs.Path.Split('')[-1])"
            if (-not $results[$key]) { $results[$key] = @() }
            $results[$key] += $cs.CookedValue
        }
    }
}

Write-Host "`n=== Disk Latency Summary ===" -ForegroundColor Yellow
foreach ($key in $results.Keys | Sort-Object) {
    $avg = ($results[$key] | Measure-Object -Average).Average
    $max = ($results[$key] | Measure-Object -Maximum).Maximum

    $color = "Green"
    if ($key -match "sec/" -and $max -gt 0.05) { $color = "Yellow" }
    if ($key -match "sec/" -and $max -gt 0.1) { $color = "Red" }

    Write-Host ("{0}: Avg={1:N4}, Max={2:N4}"1:N4}, Max={2:N4}" -f $key, $avg, $max) -ForegroundColor $color
}

Write-Host "`nNote: Values for 'sec/Read' and 'sec/Write' are in seconds." -ForegroundColor Cyan
Write-Host "Target: < 0.020 (20ms), Alert: > 0.050 (50ms), Critical: > 0.100 (100ms)"020 (20ms), Alert: > 0.050 (50ms), Critical: > 0.100 (100ms)"

# Check MPIO configuration
$mpioFeature = Get-WindowsFeature -Name Multipath-IO

if ($mpioFeature.Installed) {
    Write-Host "MPIO Feature: Installed" -ForegroundColor Green

    # Get MPIO devices
    Get-MSDSMAutomaticClaimSettings
    Get-MPIOAvailableHW

    # Check paths
    mpclaim -s -d

    # Check load balance policy
    Get-MSDSMGlobalDefaultLoadBalancePolicy

} else {
    Write-Host "MPIO Feature: Not Installed" -ForegroundColor Yellow
    Write-Host "For SAN storage, MPIO should be installed and configured"
}

# Check HBA information
Get-InitiatorPort | Select-Object PortAddress, ConnectionType, OperationalStatus | Format-Table

# For iSCSI
if (Get-Service -Name MSiSCSI -ErrorAction SilentlyContinue) {
    $iscsiTargets = Get-IscsiTarget
    Write-Host "`n=== iSCSI Targets ===" -ForegroundColor Cyan
    $iscsiTargets | Format-Table NodeAddress, IsConnected

    # Check iSCSI sessions
    Get-IscsiSession | Select-Object TargetNodeAddress, IsConnected, IsPersistent,
        NumberOfConnections | Format-Table
}

# WARNING: This masks the symptom, not the cause
# Only use while actively working on storage fix

# Check current timeout value (default is 30 seconds)
$currentTimeout = (Get-ItemProperty "HKLM:SYSTEMCurrentControlSetServicesDisk").TimeOutValue
Write-Host "Current disk timeout: $currentTimeout seconds"

# Increase timeout to 60 seconds (registry change)
Set-ItemProperty -Path "HKLM:SYSTEMCurrentControlSetServicesDisk" -Name "TimeOutValue" -Value 60

# For SAN storage, the timeout may be in the HBA driver settings
# Check vendor documentation for your specific HBA

# Verify the change
Get-ItemProperty "HKLM:SYSTEMCurrentControlSetServicesDisk" | Select-Object TimeOutValue

Write-Host "`nIMPORTANT:" -ForegroundColor Yellow
Write-Host "- This requires a reboot to take effect"
Write-Host "- This is a TEMPORARY measure while fixing storage"
Write-Host "- Extended timeout means longer hangs when issues occur"
Write-Host "- Always investigate and fix the root cause!"

# Check current storage driver versions
Get-WmiObject Win32_PnPSignedDriver |
    Where-Object {$_.DeviceClass -eq "DiskDrive" -or $_.DeviceClass -eq "SCSIAdapter" -or $_.DeviceName -match "HBA|Fibre|iSCSI"} |
    Select-Object DeviceName, DriverVersion, DriverDate | Format-Table -AutoSize

# Check storport miniport drivers
Get-WmiObject Win32_PnPSignedDriver |
    Where-Object {$_.InfName -match "disk|storport"} |
    Select-Object DeviceName, DriverVersion | Format-Table

# Driver update checklist:
Write-Host "=== Driver Update Checklist ===" -ForegroundColor Yellow
Write-Host "1. HBA drivers - Download from vendor (QLogic, Emulex, etc.)"
Write-Host "2. Storage firmware - Check SAN vendor support portal"
Write-Host "3. Windows storage drivers - Windows Update or vendor"
Write-Host "4. Multipath DSM - From storage vendor"
Write-Host ""
Write-Host "After updates:"
Write-Host "- Test in non-production first"
Write-Host "- Schedule maintenance window for production"
Write-Host "- Have rollback plan ready"

# Check for pending driver updates in Windows Update
Get-WindowsUpdateLog  # Creates readable log from ETL files

# Plan migration to new storage
# Best practices for Exchange storage:

Write-Host "=== Exchange Storage Requirements ===" -ForegroundColor Cyan
Write-Host ""
Write-Host "Performance Requirements:"
Write-Host "  - Database reads: < 20ms latency"
Write-Host "  - Database writes: < 20ms latency"
Write-Host "  - Log writes: < 10ms latency"
Write-Host "  - IOPS: Plan for peak, not average"
Write-Host ""
Write-Host "Recommended Storage Types:"
Write-Host "  1. All-flash SAN (best performance)"-flash SAN (best performance)"
Write-Host "  2. Hybrid SAN with SSD tier"
Write-Host "  3. Direct-attached SSD (excellent for small deployments)"-attached SSD (excellent for small deployments)"
Write-Host "  4. NVMe storage (highest performance)"
Write-Host ""
Write-Host "Migration Steps:"
Write-Host "  1. Provision new storage with required capacity"
Write-Host "  2. Run JetStress to validate performance"
Write-Host "  3. Create new database on new storage"
Write-Host "  4. Move mailboxes to new database"
Write-Host "  5. Decommission old database"

# For DAG environments - add new database copies
# New-MailboxDatabase -Name "DB_NewStorage" -Server EXCH01 -EdbFilePath "S:DBDB.edb" -LogFolderPath "S:Logs"-Name "DB_NewStorage" -Server EXCH01 -EdbFilePath "S:DBDB.edb" -LogFolderPath "S:Logs"
# Then move mailboxes:
# Get-Mailbox -Database "OldDatabase" | New-MoveRequest -TargetDatabase "DB_NewStorage"-Database "OldDatabase" | New-MoveRequest -TargetDatabase "DB_NewStorage"