iSCSI on Windows Server with MPIO
A lot of ONTAP LUNs end up on Windows: SQL Server databases, file servers, backup media servers. Windows makes iSCSI look easy because it’s all in one small control panel, and that’s exactly why people skip the one step that matters.
That step is MPIO. In this lesson you’ll install it first, connect Windows Server with multipath turned on, and see what goes wrong when you don’t. The practice shell plays both sides: the ONTAP cluster in one terminal and a Windows Server 2025 PowerShell prompt in another.
Where this lesson fits
Lesson 26 of 56 in NetApp ONTAP Storage Administration · Phase 2: Protocols and day-to-day administration · Before this: Lesson 25, iSCSI on Linux · Next: Lesson 27, FC target ports and zoning
| Time | You need | By the end you can |
|---|---|---|
| 75 minutes | The practice shell (version 1.2 or later) in two terminals | Install MPIO, connect persistent multipath iSCSI sessions, bring a LUN online as an NTFS volume, and spot a missing MPIO setup |
What it is, in one line
On Windows Server, the built-in iSCSI Initiator logs in to the ONTAP SVM, and the MPIO feature with the Microsoft DSM joins the paths so one LUN shows up as one disk.
In Simple Words
One friend, two phone numbers
Your friend has two phone numbers, office and mobile. If your phone book saves them as two separate people, you might tell “both of them” different things and create a mess.
Save them as one contact with two numbers and the phone simply tries the other number when one is busy. MPIO is that contact card: two paths, one disk.
Engineer's View
How an engineer describes it
The Microsoft iSCSI Initiator service (MSiSCSI) owns the IQN, usually iqn.1991-05.com.microsoft:hostname.domain. You add target portals, then connect to the target once per portal. Tick Enable multi-path and Add this connection to the list of Favorite Targets (persistent) each time.
The Multipath I/O feature, plus Enable-MSDSMAutomaticClaim -BusType iSCSI, lets the Microsoft DSM claim the LUN and present one disk. The DSM reads ONTAP’s ALUA states and, for ALUA disks, uses the Round Robin with Subset policy, which sends I/O only to optimized paths. NetApp’s Windows Host Utilities set the recommended timeouts.
On the ONTAP side, use igroup ostype windows and LUN ostype windows_2008 for Windows Server 2008 and later, so the partition alignment is right.
How it works, step by step
1Install MPIO before any LUN is connected
Add the Multipath I/O feature, claim iSCSI devices for the Microsoft DSM, and restart. Doing this after the disk is in use means downtime later.
2Start the initiator and send the IQN
Set MSiSCSI to start automatically and give the IQN to the storage team for the igroup.
3Add portals and connect with multipath
Add one LIF per node as a target portal, then connect to the target once through each portal, with multipath and persistence on.
4Bring the disk online
The LUN shows as an offline RAW disk. Initialize it as GPT, create a partition and format NTFS. Check mpclaim -s -d shows one MPIO disk.
| Job | GUI | PowerShell |
|---|---|---|
| Install MPIO | Server Manager › Add Roles and Features › Multipath I/O | Install-WindowsFeature Multipath-IO |
| Claim iSCSI disks | MPIO control panel › Discover Multi-Paths › Add support for iSCSI devices | Enable-MSDSMAutomaticClaim -BusType iSCSI |
| Add a portal | iSCSI Initiator › Discovery › Discover Portal | New-IscsiTargetPortal |
| Connect | iSCSI Initiator › Targets › Connect › Enable multi-path | Connect-IscsiTarget -IsMultipathEnabled $true -IsPersistent $true |
| New volume | Disk Management › Online, Initialize, New Simple Volume | Initialize-Disk, New-Partition, Format-Volume |
On the ONTAP side, System Manager’s Storage › LUNs › + Add with host operating system Windows creates the volume, LUN, igroup and mapping in one form.
Hands-on lab: connect win1 with MPIO and multipath
Before you run any command, double-check it
Read every command twice before you press Enter. Check the target: the server name, the user, the disk or device name, the path and the sizes. These commands are for learning and were tested in our lab, not on your system. At work, run the read-only commands first, follow your change process, make sure a backup exists, and ask a second person to check anything that changes or deletes data. You run them at your own risk.
This lab uses the SaatVedha ONTAP practice shell, our own learning tool. It is not NetApp software: real ONTAP output looks like this but is not identical.
You’ll use two terminals. The first runs the cluster shell (python ontapsim.py).
The second runs python ontapsim.py --host win1, which opens an administrator PowerShell prompt on win1, a Windows Server 2025 machine at 192.168.56.31 joined to saatvedha.lab. Type the same cmdlets you would on a real server; restarts take a second instead of a minute.
1Build the storage side
Save these lines as setup-26.txt: two aggregates, an SVM with one iSCSI LIF per node, and a 50 GB LUN mapped to win1’s IQN. Note the windows_2008 OS type for the LUN and windows for the igroup, the values ONTAP uses for every current Windows Server release.
storage aggregate create -aggregate aggr1_01 -node saat-cluster-01 -diskcount 8 -disktype SAS
storage aggregate create -aggregate aggr1_02 -node saat-cluster-02 -diskcount 8 -disktype SAS
vserver create -vserver svm1 -aggregate aggr1_01 -rootvolume-security-style unix
network interface create -vserver svm1 -lif svm1_iscsi1 -service-policy default-data-iscsi -home-node saat-cluster-01 -home-port e0d -address 192.168.56.131 -netmask 255.255.255.0
network interface create -vserver svm1 -lif svm1_iscsi2 -service-policy default-data-iscsi -home-node saat-cluster-02 -home-port e0d -address 192.168.56.132 -netmask 255.255.255.0
vserver iscsi create -vserver svm1
volume create -vserver svm1 -volume vol_sql1 -aggregate aggr1_01 -size 60g -space-guarantee none -snapshot-policy none
lun create -vserver svm1 -path /vol/vol_sql1/lun_sql1 -size 50g -ostype windows_2008 -space-reserve disabled
lun igroup create -vserver svm1 -igroup ig_win1 -protocol iscsi -ostype windows -initiator iqn.1991-05.com.microsoft:win1.saatvedha.lab
lun mapping create -vserver svm1 -path /vol/vol_sql1/lun_sql1 -igroup ig_win1Reset the lab and run it:
python ontapsim.py -c "simctl reset"
python ontapsim.py --run setup-26.txt✅ Checkpoint: No line in the output starts with Error.
2Install MPIO first, then read the IQN
Open the second terminal with python ontapsim.py --host win1 and run:
Install-WindowsFeature -Name Multipath-IO
Set-Service MSiSCSI -StartupType Automatic
Start-Service MSiSCSI
Get-InitiatorPort
Restart-Computer
Enable-MSDSMAutomaticClaim -BusType iSCSI
Restart-ComputerYou should see this:
PS C:\Users\Administrator> Install-WindowsFeature -Name Multipath-IO
Success Restart Needed Exit Code Feature Result
------- -------------- --------- --------------
True Yes SuccessRest... {Multipath I/O}
WARNING: You must restart this server to finish the installation process.
PS C:\Users\Administrator> Set-Service MSiSCSI -StartupType Automatic
PS C:\Users\Administrator> Start-Service MSiSCSI
PS C:\Users\Administrator> Get-InitiatorPort
InstanceName NodeAddress PortAddress ConnectionType
------------ ----------- ----------- --------------
ROOT\ISCSIPRT\0000_0 iqn.1991-05.com.microsoft:win1.saatvedha.lab ISCSI ANY PORT iSCSI
PS C:\Users\Administrator> Restart-Computer
(win1 restarts. You are logged on again after about a minute.)
PS C:\Users\Administrator> Enable-MSDSMAutomaticClaim -BusType iSCSI
PS C:\Users\Administrator> Restart-Computer
(win1 restarts. You are logged on again after about a minute.)Order matters: MPIO and the automatic claim for iSCSI go in before the first login, with a restart after each. The NodeAddress column is win1’s IQN, the value the storage team puts in the igroup.
✅ Checkpoint: Install-WindowsFeature shows Success True and the IQN ends in win1.saatvedha.lab.
3Connect to both LIFs with multipath
Discovery from one LIF finds the target; then connect once through each LIF, with multipath and persistence turned on:
Test-NetConnection 192.168.56.131 -Port 3260
New-IscsiTargetPortal -TargetPortalAddress 192.168.56.131
Get-IscsiTarget
Connect-IscsiTarget -NodeAddress iqn.1992-08.com.netapp:sn.c84b2d40ac3f32ca940fcb7e13e6d933:vs.3 -TargetPortalAddress 192.168.56.131 -IsMultipathEnabled $true -IsPersistent $true
Connect-IscsiTarget -NodeAddress iqn.1992-08.com.netapp:sn.c84b2d40ac3f32ca940fcb7e13e6d933:vs.3 -TargetPortalAddress 192.168.56.132 -IsMultipathEnabled $true -IsPersistent $true
Get-Disk
mpclaim -s -d
mpclaim -s -d 0You should see this:
PS C:\Users\Administrator> Test-NetConnection 192.168.56.131 -Port 3260
ComputerName : 192.168.56.131
RemoteAddress : 192.168.56.131
RemotePort : 3260
InterfaceAlias : Ethernet 2
SourceAddress : 192.168.56.31
TcpTestSucceeded : True
PS C:\Users\Administrator> New-IscsiTargetPortal -TargetPortalAddress 192.168.56.131
InitiatorInstanceName :
InitiatorPortalAddress :
IsDataDigest : False
IsHeaderDigest : False
TargetPortalAddress : 192.168.56.131
TargetPortalPortNumber : 3260
PSComputerName :
PS C:\Users\Administrator> Get-IscsiTarget
IsConnected NodeAddress PSComputerName
----------- ----------- --------------
False iqn.1992-08.com.netapp:sn.c84b2d40ac3f32ca940fcb7e13e6d933:vs.3
PS C:\Users\Administrator> Connect-IscsiTarget -NodeAddress iqn.1992-08.com.netapp:sn.c84b2d40ac3f32ca940fcb7e13e6d933:vs.3 -TargetPortalAddress 192.168.56.131 -IsMultipathEnabled $true -IsPersistent $true
AuthenticationType : NONE
InitiatorInstanceName : ROOT\ISCSIPRT\0000_0
InitiatorNodeAddress : iqn.1991-05.com.microsoft:win1.saatvedha.lab
InitiatorPortalAddress : 0.0.0.0
InitiatorSideIdentifier : 400001370000
IsConnected : True
IsDataDigest : False
IsDiscovered : True
IsHeaderDigest : False
IsPersistent : True
NumberOfConnections : 1
SessionIdentifier : ffffb38d1b2b2010-4000013700000002
TargetNodeAddress : iqn.1992-08.com.netapp:sn.c84b2d40ac3f32ca940fcb7e13e6d933:vs.3
TargetSideIdentifier : 1e00
PSComputerName :
PS C:\Users\Administrator> Connect-IscsiTarget -NodeAddress iqn.1992-08.com.netapp:sn.c84b2d40ac3f32ca940fcb7e13e6d933:vs.3 -TargetPortalAddress 192.168.56.132 -IsMultipathEnabled $true -IsPersistent $true
AuthenticationType : NONE
InitiatorInstanceName : ROOT\ISCSIPRT\0000_0
InitiatorNodeAddress : iqn.1991-05.com.microsoft:win1.saatvedha.lab
InitiatorPortalAddress : 0.0.0.0
InitiatorSideIdentifier : 400001370001
IsConnected : True
IsDataDigest : False
IsDiscovered : True
IsHeaderDigest : False
IsPersistent : True
NumberOfConnections : 1
SessionIdentifier : ffffb38d1b2b2010-4000013700000003
TargetNodeAddress : iqn.1992-08.com.netapp:sn.c84b2d40ac3f32ca940fcb7e13e6d933:vs.3
TargetSideIdentifier : 1f00
PSComputerName :
PS C:\Users\Administrator> Get-Disk
Number Friendly Name Serial Number HealthStatus OperationalStatus Total Size Partition
Style
------ ------------- ------------- ------------ ----------------- ---------- ----------
0 VBOX HARDDISK VB4c3e0b2a-9d7f61c5 Healthy Online 50 GB GPT
1 NETAPP LUN C-Mode Dmi1wGdgb5T2 Healthy Offline 50 GB RAW
PS C:\Users\Administrator> mpclaim -s -d
For more information about a particular disk, use 'mpclaim -s -d #' where # is the MPIO disk number.
MPIO Disk System Disk LB Policy DSM Name
-------------------------------------------------------------------------------
MPIO Disk0 Disk 1 RRWS Microsoft DSM
PS C:\Users\Administrator> mpclaim -s -d 0
MPIO Disk0: 02 Paths, Round Robin with Subset, Implicit Only
Controlling DSM: Microsoft DSM
SN: 600A0980446D69317747646762355432
Supported Load Balance Policies: FOO RR RRWS LQD WP LB
Path ID State SCSI Address Weight
---------------------------------------------------------------------------
0000000077030000 Active/Optimized 003|000|000|000 0
TPG_State : Active/Optimized , TPG_Id: 1007, : 2
0000000077030001 Active/Unoptimized 003|000|001|000 0
TPG_State : Active/Unoptimized, TPG_Id: 1008, : 3Two sessions, yet Get-Disk lists one NETAPP disk: MPIO joined the paths. mpclaim -s -d 0 shows the ALUA states from Lesson 24, one Active/Optimized path through the LUN’s node and one Active/Unoptimized through the partner. Round Robin with Subset uses only the optimized path until it fails.
The GUI way is the iSCSI Initiator (iscsicpl): Discovery › Discover Portal, then Targets › Connect with Enable multi-path ticked, once per portal under Advanced.
✅ Checkpoint: One NETAPP disk, and mpclaim shows MPIO Disk0 with 02 Paths.
4Bring the disk online as an NTFS volume
Set-Disk -Number 1 -IsOffline $false
Initialize-Disk -Number 1 -PartitionStyle GPT
New-Partition -DiskNumber 1 -UseMaximumSize -DriveLetter E
Format-Volume -DriveLetter E -FileSystem NTFS -NewFileSystemLabel SQLDATA -Confirm:$false
Get-Volume -DriveLetter EYou should see this:
PS C:\Users\Administrator> Set-Disk -Number 1 -IsOffline $false
PS C:\Users\Administrator> Initialize-Disk -Number 1 -PartitionStyle GPT
PS C:\Users\Administrator> New-Partition -DiskNumber 1 -UseMaximumSize -DriveLetter E
DiskPath: \\?\mpio#disk&ven_netapp&prod_lun_c-mode&rev_9181#1&7f6ac24&0&446d693177476467#{53f56307-b6bf-11d0-94f2-00a0c91efb8b}
PartitionNumber DriveLetter Offset Size Type
--------------- ----------- ------ ---- ----
2 E 16777216 49.98 GB Basic
PS C:\Users\Administrator> Format-Volume -DriveLetter E -FileSystem NTFS -NewFileSystemLabel SQLDATA -Confirm:$false
DriveLetter FriendlyName FileSystemType DriveType HealthStatus OperationalStatus SizeRemaining Size
----------- ------------ -------------- --------- ------------ ----------------- ------------- ----
E SQLDATA NTFS Fixed Healthy OK 49.63 GB 49.98 GB
PS C:\Users\Administrator> Get-Volume -DriveLetter E
DriveLetter FriendlyName FileSystemType DriveType HealthStatus OperationalStatus SizeRemaining Size
----------- ------------ -------------- --------- ------------ ----------------- ------------- ----
E SQLDATA NTFS Fixed Healthy OK 49.63 GB 49.98 GBUse the disk number Get-Disk gave you; on a real server, double-check it’s the NETAPP disk before you initialize anything. The DiskPath starts with mpio#disk, more proof that Windows is using the multipath device.
✅ Checkpoint: Drive E: labelled SQLDATA is Healthy, about 50 GB.
5See the sessions from ONTAP
In the cluster shell:
vserver iscsi session show -vserver svm1
lun igroup show -vserver svm1 -igroup ig_win1 -instanceYou should see this:
saat-cluster::> vserver iscsi session show -vserver svm1
Tpgroup Initiator Initiator
Vserver Name TSIH Name Alias
---------- ------------- ---- ---------------------------------------------- ----------
svm1 svm1_iscsi1 1 iqn.1991-05.com.microsoft:win1.saatvedha.lab win1
svm1 svm1_iscsi2 2 iqn.1991-05.com.microsoft:win1.saatvedha.lab win1
2 entries were displayed.
saat-cluster::> lun igroup show -vserver svm1 -igroup ig_win1 -instance
Vserver Name: svm1
Igroup Name: ig_win1
Protocol: iscsi
OS Type: windows
Portset Binding Igroup: -
Igroup UUID: 7a673985-526e-90db-4859-10580a4ef67b
Initiators: iqn.1991-05.com.microsoft:win1.saatvedha.lab (logged in)✅ Checkpoint: Two sessions from win1 and the initiator shows (logged in).
Try it yourself: Build the ONTAP side for a Windows LUN in the Command Builder (choose ostype windows).
Break It, Then Fix It
The mistake that causes real data loss on Windows: a LUN reached through two paths without MPIO. Windows then sees two disks that are really one.
1Lose MPIO
In win1’s PowerShell, let simctl remove the MPIO feature, the way a rebuilt server or a skipped build step would, then restart:
simctl break no-mpio
Restart-Computer
Get-Disk
mpclaim -s -dYou’ll get this:
PS C:\Users\Administrator> simctl break no-mpio
simctl: fault "no-mpio" is active on win1. Now find it from the host and the cluster.
PS C:\Users\Administrator> Restart-Computer
(win1 restarts. You are logged on again after about a minute.)
PS C:\Users\Administrator> Get-Disk
Number Friendly Name Serial Number HealthStatus OperationalStatus Total Size Partition
Style
------ ------------- ------------- ------------ ----------------- ---------- ----------
0 VBOX HARDDISK VB4c3e0b2a-9d7f61c5 Healthy Online 50 GB GPT
2 NETAPP LUN C-Mode Dmi1wGdgb5T2 Healthy Offline 50 GB RAW
3 NETAPP LUN C-Mode Dmi1wGdgb5T2 Healthy Offline 50 GB RAW
PS C:\Users\Administrator> mpclaim -s -d
mpclaim : The term 'mpclaim' is not recognized as the name of a cmdlet, function, script file, or operable program. Check
the spelling of the name, or if a path was included, verify that the path is correct and try again.
At line:1 char:1
+ mpclaim -s -d
+ ~~~~~~~
+ CategoryInfo : ObjectNotFound: (mpclaim:String) [], CommandNotFoundException
+ FullyQualifiedErrorId : CommandNotFoundExceptionThe same serial number twice: Disk 2 and Disk 3 are one LUN seen through each session. If someone brings one online and “formats the new disk”, they wipe the data under the other. mpclaim doesn’t even exist, because it ships with the MPIO feature.
✅ Checkpoint: Two NETAPP disks with the same serial number, and mpclaim is not recognized.
2Fix it: MPIO, claim, restart
Install-WindowsFeature -Name Multipath-IO
Restart-Computer
Enable-MSDSMAutomaticClaim -BusType iSCSI
Restart-Computer
Get-Disk
mpclaim -s -d
Get-Volume -DriveLetter E
simctl fix no-mpioYou should see this:
PS C:\Users\Administrator> Install-WindowsFeature -Name Multipath-IO
Success Restart Needed Exit Code Feature Result
------- -------------- --------- --------------
True Yes SuccessRest... {Multipath I/O}
WARNING: You must restart this server to finish the installation process.
PS C:\Users\Administrator> Restart-Computer
(win1 restarts. You are logged on again after about a minute.)
PS C:\Users\Administrator> Enable-MSDSMAutomaticClaim -BusType iSCSI
PS C:\Users\Administrator> Restart-Computer
(win1 restarts. You are logged on again after about a minute.)
PS C:\Users\Administrator> Get-Disk
Number Friendly Name Serial Number HealthStatus OperationalStatus Total Size Partition
Style
------ ------------- ------------- ------------ ----------------- ---------- ----------
0 VBOX HARDDISK VB4c3e0b2a-9d7f61c5 Healthy Online 50 GB GPT
1 NETAPP LUN C-Mode Dmi1wGdgb5T2 Healthy Online 50 GB GPT
PS C:\Users\Administrator> mpclaim -s -d
For more information about a particular disk, use 'mpclaim -s -d #' where # is the MPIO disk number.
MPIO Disk System Disk LB Policy DSM Name
-------------------------------------------------------------------------------
MPIO Disk0 Disk 1 RRWS Microsoft DSM
PS C:\Users\Administrator> Get-Volume -DriveLetter E
DriveLetter FriendlyName FileSystemType DriveType HealthStatus OperationalStatus SizeRemaining Size
----------- ------------ -------------- --------- ------------ ----------------- ------------- ----
E SQLDATA NTFS Fixed Healthy OK 49.63 GB 49.98 GB
PS C:\Users\Administrator> simctl fix no-mpio
simctl: fixed no-mpio.One disk again, back online with its GPT partition, and E: is still there with its data. The last line just tidies the lab.
✅ Checkpoint: Get-Disk shows one NETAPP disk and mpclaim lists MPIO Disk0.
If you ever see two identical NETAPP disks on a Windows host, stop and check MPIO before anyone touches Disk Management.
This happened at work
The “new” disk in Madurai
A textile exporter in Madurai moved its ERP database server to a LUN on ONTAP. The server team connected the iSCSI target through both LIFs, as the storage team had asked, but nobody had installed MPIO on that server.
A month later an admin needed space for a report export. He opened Disk Management, found an offline 500 GB disk that looked unused, brought it online and formatted it. It was the ERP LUN, seen through its second path, and the database stopped within seconds.
They restored the volume from the previous night’s ONTAP snapshot, which took about three hours with the checks, and lost half a day of entries that staff typed in again. Afterwards MPIO became part of the Windows build image, and the storage team began checking mpclaim -s -d output before closing any Windows SAN request.
Common mistakes and troubleshooting
| Mistake | What happens | Do this instead |
|---|---|---|
| Connecting before MPIO is installed | The LUN shows up twice; someone may format the copy | Install MPIO and claim iSCSI disks first, then connect |
| Not ticking Enable multi-path | Same problem: two disks instead of one | Use -IsMultipathEnabled $true on every connection |
| Sessions not persistent | The disk is gone after a restart | Use -IsPersistent $true (Favorite Targets) |
| LUN ostype windows on Windows 2008 and later | Partition alignment is off, small performance loss | Use LUN ostype windows_2008; igroup ostype windows |
| Connecting through only one LIF | A node failover cuts off the disk | Add a portal on each node and connect through each |
| Sharing one LUN between two standalone servers | File system corruption | Only clustered hosts (WSFC with CSV) may share a LUN |
Interview questions
Click a question to open it. Each one is written the way an interviewer would ask it. Under it you will see why they ask it and one way you could answer, in your own words.
Fresher level
1. What is MPIO on Windows Server and why do you need it for ONTAP LUNs?
Why they ask this: They want to see that you know the one setting that prevents duplicate disks.
How you might answer: MPIO is the Windows feature that joins several paths to the same LUN into one disk. ONTAP gives every LUN at least two paths, one through each node, so without MPIO Windows shows the LUN twice and fails over badly.
2. How do you find a Windows server’s iSCSI initiator name?
Why they ask this: They’re checking you can get the value the igroup needs.
How you might answer: In the iSCSI Initiator, on the Configuration tab, or with (Get-InitiatorPort).NodeAddress in PowerShell. It usually looks like iqn.1991-05.com.microsoft followed by the host name.
2–5 years experience
3. A Windows admin sees two identical NetApp disks in Disk Management. What do you tell them?
Why they ask this: They want to see you recognise a dangerous situation and act calmly.
How you might answer: I’d tell them not to bring the second disk online or format anything. It’s almost certainly one LUN seen through two paths without MPIO. Then I’d check that MPIO is installed and claimed for iSCSI, reconnect the sessions with multipath on, and confirm mpclaim -s -d shows one disk.
4. Which ostype values do you use for a Windows LUN and igroup, and why does it matter?
Why they ask this: They’re testing attention to detail that affects alignment and support.
How you might answer: I use windows_2008 for the LUN on Windows Server 2008 and later, and windows for the igroup. The LUN ostype sets the geometry so NTFS partitions line up with WAFL blocks; the wrong one can cost performance.
Senior level
5. You’re asked to standardise Windows SAN hosts across 60 servers. What do you include?
Why they ask this: They want to see you think in builds and checks, not one-off fixes.
How you might answer: MPIO and the iSCSI claim go into the base image, along with NetApp’s Windows Host Utilities for timeouts. The connection script adds a portal per node and connects with multipath and persistence on.
Every build is signed off with mpclaim -s -d, the session count and a restart test, and the output goes in the ticket.
6. A SQL Server failover cluster needs shared LUNs from ONTAP. What do you watch out for?
Why they ask this: They’re checking that you understand shared disks need a cluster and careful igroup design.
How you might answer: All cluster nodes go in one igroup so they all see the same LUNs, and Windows Server Failover Clustering decides who writes. I’d confirm MPIO on every node first, keep the quorum disk small, and never map those LUNs to a server outside the cluster.
Quick quiz
Q1. What must you install on Windows Server before connecting a multipath iSCSI LUN?
- Hyper-V
- The Multipath I/O feature
- IIS
- Storage Spaces
Show answer
B — MPIO joins the paths into one disk.
Q2. Which PowerShell switch makes an iSCSI connection come back after a restart?
- -IsMultipathEnabled $true
- -IsPersistent $true
- -AuthenticationType ONEWAYCHAP
- -ReportToPnP
Show answer
B — persistent connections are Favorite Targets that reconnect at boot.
Q3. Without MPIO, a LUN with two paths shows up as…
- One disk
- Two disks
- No disk
- A network drive
Show answer
B — each path becomes its own disk.
Q4. Which LUN ostype suits Windows Server 2008 and later?
- windows
- windows_2008
- linux
- vmware
Show answer
B — windows_2008 gives the right alignment for newer Windows.
Q5. Which command lists MPIO disks and their load-balance policy?
- diskpart list
- mpclaim -s -d
- Get-Volume
- iscsicli ListTargets
Show answer
B — mpclaim -s -d shows MPIO disks and policies.
Quick recap
- Install MPIO and claim iSCSI disks before you connect anything.
- Add a portal on each node and connect through each, with multipath and persistence on.
- One LUN must show as one disk; check with mpclaim -s -d.
- Use LUN ostype windows_2008 and igroup ostype windows.
- Two identical disks means stop: fix MPIO before anyone formats.
Next lesson
Lesson 27: FC target ports, UTA2 and zoning → You’ll set ports to FC target mode, create FC LIFs and zone a host to their virtual WWPNs.
Further reading: ONTAP iSCSI Windows
Want to go deeper into ONTAP iSCSI Windows? Read NetApp’s ONTAP concepts guide for the SAN side, and Microsoft’s MPIO and iSCSI Initiator documentation for host details. You can also see the full plan for this course on the NetApp ONTAP Storage Administration course page.