Update windows-server-mpio-troubleshooting.md

Edit review per CI 8339

Author: przlplx

support/windows-server/backup-and-storage/windows-server-mpio-troubleshooting.md

@@ -1,6 +1,6 @@
 ---
 title: Multipath I/O (MPIO) Troubleshooting Guidance
-description: Describes how to troubleshoot and fx issues in Hyper-V, clustering, and virtualization environments that use the Windows Multipath I/O (MPIO) feature.
+description: Discusses how to troubleshoot and fix issues in Hyper-V, clustering, and virtualization environments that use the Windows Multipath I/O (MPIO) feature.
 ms.date: 12/12/2025
 manager: dcscontentpm
 audience: itpro
@@ -34,12 +34,12 @@ Use this checklist for systematic troubleshooting.
 ### Prepare to troubleshoot
 
 - Back up all the affected systems and data. Test that you can restore the backup.
-- Confirm that you have a maintenance window and change management approval for troubleshooting.
+- Verify that you have change management approval and a maintenance window for troubleshooting.
 - Document all observed issues, error messages, event IDs, and timing.
 
 ### Review the symptoms
 
-- Check the Disk Management and Failover Cluster Manager tools, as well as the settings of any affected virtual machines (VMs). Are disks or volumes missing?
+- Check the Disk Management and Failover Cluster Manager tools. Additionally, check the settings of any affected virtual machines (VMs). Are disks or volumes missing?
 - Are issues isolated to one cluster node, all nodes, or all servers?
 - Check Server Manager. Are storage controller and network paths missing?
 - If you're using third-party DSMs, check their documentation. Are they certified for your operating system version?
@@ -60,19 +60,19 @@ Use this checklist for systematic troubleshooting.
 ### Verify the MPIO and storage configuration
 
 1. To review the device IDs of the storage hardware, follow these steps:
-   1. In the search bar, type *mpiocpl*, and then in the search results, select **MPIO**.
-   1. Select **Discover Multi-Paths**. Review the list of IDs and fix any errors.
+   1. In the search bar, type *mpiocpl*, and then select **MPIO** in the search results.
+   1. Select **Discover Multi-Paths**. Review the list of IDs, and fix any errors.
    > [!NOTE]  
    > You can also use the Windows PowerShell `Get-MPIOAvailableHW` and `Get-MPIOSetting` cmdlets, or the command line `mpclaim -e` and `mpclaim -r -i -a` commands to perform these steps.
-1. To review and update the storage area network (SAN) policy, open a Windows command prompt window and then run the following command:
+1. To review and update the storage area network (SAN) policy, open a Windows Command Prompt window, and then run the following command:
 
    ```console
    diskpart san policy=OnlineAll
    ```
 
 1. To update the path configuration, follow these guidelines:
-   1. For redundancy, make sure each iSCSI connection uses a different network adapter.
-   1. If the system only detects one path, rebuild the iSCSI connections.
+   - For redundancy, make sure that each iSCSI connection uses a different network adapter.
+   - If the system detects only one path, rebuild the iSCSI connections.
 
 ### Update drivers and firmware
 
@@ -82,7 +82,7 @@ Use this checklist for systematic troubleshooting.
 ### Verify the health of the file system and disks
 
 - To check and repair file system errors, run `chkdsk <drive>: /f /r` at a Windows command prompt.
-- If chkdsk fails while scanning a RAW volume, attempt to recover the data or restore it from a backup.
+- If chkdsk fails while scanning a RAW volume, try to recover the data or restore it from a backup.
 - If the resilient file system (ReFS) is unstable, consider reformatting the file system to NTFS.
 
 ### Check for cluster and CSV disk issues
@@ -97,24 +97,25 @@ Use this checklist for systematic troubleshooting.
   > [!NOTE]  
   >
   > - In these commands, \<Resource_Name> is the name of the disk resource.
-  > - The value of the PendingTimeout property is measured in milliseconds. The value shown here's higher than the default value.
+  > - The value of the `PendingTimeout` property is measured in milliseconds. The value that's shown here is greater than the default value.
 
 - For stale ClusterStorage folders, stop the cluster service, use `takeown` to take ownership and `icacls` to reset permissions, and then use `rmdir`to delete.
 
 ### Check the state of the MPIO paths
 
 - If only some paths are visible, verify zoning, cabling, and HBA status on both server and the SAN.
 - If you observe Event IDs 153, 129, or 140, you might have storage or network bottlenecks. Review your underlying storage or network traffic.
-- If MPIO paths are missing or incorrectly configured following a restart, see [After maintenance or a restart, administrative tools don't show disks or paths](#after-maintenance-or-a-restart-administrative-tools-dont-show-disks-or-paths) for instructions.
+- If MPIO paths are missing or configured incorrectly after a restart, see [After maintenance or a restart, administrative tools don't show disks or paths](#after-maintenance-or-a-restart-administrative-tools-dont-show-disks-or-paths) for instructions.
 
 ### Tune Registry and policy settings
 
 [!INCLUDE [Registry important alert](../../../includes/registry-important-alert.md)]
 
-- If you observe excessive logging (such as unusual growth of the lsass.log file), set the following registry value.
+- If you observe excessive logging (such as unusual growth of the Lsass.log file), set the following registry value:
 
   - Subkey: `HKLM\SYSTEM\CurrentControlSet\Control\Lsa`
   - Value, data, and type: `LogToFile` = `0` (DWORD)
+
   After you change a registry value, restart the computer.
 
 - To configure MPIO verification settings, run the following cmdlets at a PowerShell command prompt window:
@@ -124,40 +125,41 @@ Use this checklist for systematic troubleshooting.
   Set-MPIOSetting -CustomPathRecovery Enabled
   ```
 
-- To configure intervals and timeouts for path verification, confirm appropriate settings together with your hardware vendor. Set the following values in the `HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\mpio\Parameters` subkey:
+- To configure intervals and timeouts for path verification, verify the appropriate settings by working with your hardware vendor. Set the following values in the `HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\mpio\Parameters` subkey:
 
   - `PathVerifyEnabled` (DWORD): 1 to turn on
   - `PathRecoveryInterval` (DWORD): Time in seconds (default: 30)
   - `PDORemovePeriod` (DWORD): Time in seconds (default: 20)
+  
   After you change a registry value, restart the computer.
 
 ### Fix file system corruption or RAW disks
 
-- If your disk is RAW and can't be fixed, delete and recreate the partition, then restore data from backup.
+- If your disk is RAW and can't be fixed, delete and re-create the partition. Then, restore data from backup.
 - If your clustered disks have stale reservations, use `Clear-ClusterDiskReservation` at a PowerShell command prompt to clear them.
 
 ### Switch DSM or remove third-party multipath drivers
 
-1. If you're using third-party DSMs, use the Programs and Features control panel to uninstall them.
+1. If you're using third-party DSMs, use the Programs and Features item in Control Panel to uninstall them.
 1. Restart the computer.
 1. In the MPIO tool, select **Devices**, and then remove the hardware IDs.
-1. Restart the computer again, then in the MPIO tool, re-add the hardware IDs for MSDSM.
+1. Restart the computer again. Then, in the MPIO tool, restore the hardware IDs for MSDSM.
 1. To make sure that Microsoft DSM claims your disks, run `mpclaim -s -d` at the Windows command prompt.
 
 ## Common issues and solutions
 
 The following sections describe the most common issues, and how to fix them.
 
-### Windows Server 2019 or 2022: VM IO performance degrades when Hyper-V uses resilient change tracking (RCT) is in use
+### Windows Server 2022 or 2019: VM IO performance degrades when Hyper-V uses resilient change tracking (RCT) is in use
 
-If you experience this issue, make sure that your computers are up to date. [October 23, 2025—KB5070884 (OS Build 20348.4297)](https://support.microsoft.com/en-us/topic/october-23-2025-kb5070884-os-build-20348-4297-out-of-band-9c001fdc-f0d2-4636-87bb-494a59da55d0) and subsequent updates contain a fix for this issue.
+If you experience this issue, make sure that your computers are up to date. The [October 23, 2025—KB5070884 (OS Build 20348.4297)](https://support.microsoft.com/en-us/topic/october-23-2025-kb5070884-os-build-20348-4297-out-of-band-9c001fdc-f0d2-4636-87bb-494a59da55d0) update and subsequent updates contain a fix for this issue.
 
 > [!NOTE]  
-> After you install this update, VMs that were previously backed up by using a host-level backup application might not be able to start. To fix this issue, delete any .rct and .mrt files that are associated with the affected virtual hard disks. Then try again to start the VMs. If the issue persists, contact Microsoft Support.
+> After you install this update, VMs that were previously backed up by using a host-level backup application might not be able to start. To fix this issue, delete any .rct and .mrt files that are associated with the affected virtual hard disks. Then, try again to start the VMs. If the issue persists, contact Microsoft Support.
 
 ### Some cluster disk resources remain in "Online Pending" state
 
-The default `PendingTimeout` value is too low for cluster disks that have quota or File Server Resource Manager (FSRM) changes. As a result, the cluster disks don't come online. To increase this `PendingTimout` value, run the following cmdlets at the PowerShell command prompt:
+The default `PendingTimeout` value is too low for cluster disks that have quota or File Server Resource Manager (FSRM) changes. Therefore, the cluster disks don't come online. To increase this `PendingTimout` value, run the following cmdlets at the PowerShell command prompt:
 
 ```powershell
 Get-ClusterResource "<Resource_Name>" | Set-ClusterParameter PendingTimeout 300000
@@ -170,7 +172,7 @@ Get-ClusterResource "<Resource_Name>" | Set-ClusterParameter PendingTimeout 3000
 
 ### Cluster disk resources are slow to come online
 
-You might observe cluster disk resources taking three to five minutes to come online. This behavior also generates Event ID 1069 and Event ID 4874.
+You might observe that cluster disk resources take three to five minutes to come online. This behavior also generates Event ID 1069 and Event ID 4874.
 
 To fix this issue, follow these steps:
 
@@ -186,9 +188,9 @@ The Disk Management or Failover Cluster Manager tools don't show all of the disk
 To fix this issue, follow these steps:
 
 1. Check the physical disk connections, switch zoning, and storage connectivity.
-1. In Device Manager (enable "show hidden"), remove ghost or hidden devices (or run `devnode clean` at a Windows command prompt).
-1. If you're using an unsupported DSM or duplicate DSMs, use the Programs and Features control panel to uninstall them.
-1. Restart, reinstall, or re-enable the Multipath-IO feature (to do this, you can run `Install-WindowsFeature Multipath-IO` at a PowerShell command prompt).
+1. In Device Manager (enable "show hidden"), remove ghost or hidden devices, or run `devnode clean` at a Windows command prompt.
+1. If you're using an unsupported DSM or duplicate DSMs, use the Programs and Features item in Control Panel to uninstall them.
+1. Restart, reinstall, or re-enable the Multipath-IO feature. To perform this step, run `Install-WindowsFeature Multipath-IO` at a PowerShell command prompt).
 1. At a Windows command prompt, run the following commands, in sequence:
 
     ```console
@@ -199,19 +201,19 @@ To fix this issue, follow these steps:
 
 1. In the MPIO tool, select **Discover Multi-Paths**, add the missing hardware IDs.
 
-1. To review and update the storage area network (SAN) policy, open a Windows command prompt window and then run the following command:
+1. To review and update the storage area network (SAN) policy, open a Windows Command Prompt window, and then run the following command:
 
    ```console
    diskpart san policy=OnlineAll
    ```
 
 1. Use Disk Management or diskpart to bring the missing disks online.
 
-1. To review the "Favorite Targets" list, use the iSCSI Initiator tool (In the search bar, type *iscsiocpl*, and then in the search results, select **iSCI Initiator**, and then select **Favorite Targets**.)
-
+1. To review the "Favorite Targets" list, use the iSCSI Initiator tool (In the search bar, type *iscsiocpl*, select **iSCI Initiator** in the in the search results, and then select **Favorite Targets**.
+   
 ### MPIO is in a degraded state, slow or unresponsive during failover
 
-During a failover, you observe delays that might exceed 30 seconds. You also observe IO error messages, and messages such as "MPIO is in a degraded state." You might observe the following events:
+During a failover, you might observe delays that exceed 30 seconds. You also observe IO error messages, and messages such as "MPIO is in a degraded state." You might also observe the following events:
 
 - Event ID 46
 - Event ID 129
@@ -256,7 +258,7 @@ To fix this issue, follow these steps:
     Set-VHD -Path \<VHD-Path> -ResetDiskIdentifier
     ```
 
-1. If the repeating events refer to third-party DSMs, consult storage vendor or migrate to a supported native DSM.
+1. If the repeating events refer to third-party DSMs, consult the storage vendor or migrate to a supported native DSM.
 
 ### Duplicate disks or changed disk numbering
 
@@ -283,7 +285,7 @@ To diagnose and fix these performance issues, follow these steps:
 
 ### Cluster disk resource or LUN disappears after you expand or resize it
 
-After you increase disk or LUN capacity, Disk Management or Failover Cluster Manager don't display the related volume. The volume reappears if you take the cluster role offline and then online again.
+After you increase disk or LUN capacity, Disk Management or Failover Cluster Manager doesn't display the related volume. The volume reappears if you take the cluster role offline and then online again.
 
 To fix this issue, follow these steps:
 
@@ -322,12 +324,12 @@ To fix this issue, follow these steps:
 | Symptom | Cause | Resolution |
 | --- | --- | --- |
 | Disk missing in MPIO properties and Disk Mgmt | Zoning or DSM conflict | Check Device Manager, remove ghost devices, reinstall MPIO, restart |
-| Event IDs 153/129/11 | Path loss, driver/firm | Update firmware/drivers, set MPIO policy, check physical connectivity |
-| Failover delay, slow cluster resource online | Timeout, configuration  error | Raise resource timeout, fix quotas/snapshots, update drivers |
-| Duplicate disks/wrong disk order | Bad DSM/configuration | Remove extra DSM, use persistent identifiers (GUID/WWN) |
-| High IO latency, app slow, Event ID 833 | Driver/antivirus | Exclude storage from AV, update drivers, Perfmon, use 64K clusters |
-| Disk GUID duplicate, Event ID 158 | No MPIO/clone VHDs | Enable MPIO, Set-VHD -ResetDiskIdentifier |
-| Cluster disk disappears after expansion | Driver/event miss | Cycle role offline and online, rescan, extend with DiskPart |
+| Event IDs 153/129/11 | Path loss, driver/firm | Update firmware or drivers, set MPIO policy, check physical connectivity |
+| Failover delay, slow cluster resource online | Timeout, configuration  error | Raise resource timeout, fix quotas and snapshots, update drivers |
+| Duplicate disks/wrong disk order | Bad DSM or configuration | Remove extra DSM, use persistent identifiers (GUID/WWN) |
+| High IO latency, app slow, Event ID 833 | Driver and antivirus | Exclude storage from AV, update drivers, Perfmon, use 64K clusters |
+| Disk GUID duplicate, Event ID 158 | No MPIO or clone VHDs | Enable MPIO, Set-VHD -ResetDiskIdentifier |
+| Cluster disk disappears after expansion | Driver or event miss | Cycle role offline and online, rescan, extend with DiskPart |
 | Paths not detected until restart | Storport configuration issue | `Update-HostStorageCache`, restart |
 | "Requested resource in use," can't bring disk online | Metadata or disk issue | Use `chkdsk`, check logs, engage storage team |
 
@@ -345,7 +347,7 @@ If these procedures don't resolve your issue, contact Microsoft Support. Use the
   ```
 
   > [!NOTE]
-  > Before you collect traces, for Windows Server 2016 and later, make sure you install the latest cumulative updates.
+  > Before you collect traces for Windows Server 2016 or later versions, make sure that you install the latest cumulative updates.
 
 - **Performance Monitor (Perfmon):** Use the GUI or command line interface. The following example collects data at five-second intervals:
 
@@ -356,16 +358,16 @@ If these procedures don't resolve your issue, contact Microsoft Support. Use the
 
 - **PowerShell:** Use the following cmdlets to gather information:
   - `Get-VMNetworkAdapter -VMName \<YourVMName>`
-  - `Import-VM` for import verification and error export.
+  - `Import-VM` (for import verification and error export)
 - **System and cluster logs:**
-  - Windows Event Viewer: Collect logs from Hyper-V VMMS, cluster services, and storage
+  - Windows Event Viewer: Collect logs from Hyper-V VMMS, cluster services, and storage.
   - Cluster validation reports
-- **Registry Editor:** Audit the permissions under `HKLM\SOFTWARE\Microsoft\Windows NT\CurrentVersion\Virtualization\Worker`
-- **BIOS/UEFI:** Export settings, or create screenshots of them
-- **Minidump files** If you observe a stop or bug check error (also known as a bluescreen error), retrieve these files
-- **Network trace logs** If you observe connectivity issues, collect network traces
-- **Exported VM configuration files:** If you're troubleshooting import or export issues, export configuration files for the affected VMs
-- **Driver versions** Note the current versions that your storage components, network components, and security agents use
+- **Registry Editor:** Audit the permissions under `HKLM\SOFTWARE\Microsoft\Windows NT\CurrentVersion\Virtualization\Worker`.
+- **BIOS/UEFI:** Export settings, or create screenshots of them.
+- **Minidump files** If you observe a stop or bug check error (also known as a bluescreen error), retrieve these files.
+- **Network trace logs** If you observe connectivity issues, collect network traces.
+- **Exported VM configuration files:** If you're troubleshooting import or export issues, export configuration files for the affected VMs.
+- **Driver versions:** Note the current versions that your storage components, network components, and security agents use.
 - **Storport traces:** Run commands that resemble the following examples at the Windows command prompt:
 
   ```console