Free tools Windows power users keep installed
One-click scans. No signup required.
iSCSI multipathing gives a hypervisor more than one route to storage so I/O can continue if an individual path fails. Reliable failover depends on more than adding connections: the paths must be independent, the hypervisor must use the right multipathing controls for its platform, and a controlled test must confirm that traffic actually moves to a healthy alternate path. VMware ESXi and Windows Server/Hyper-V configure those controls differently, and the correct policy and recovery behavior depend on the array and its supported configuration.
Plan independent paths before configuring the host
Start with the storage vendor’s supported topology for the exact hypervisor release, array model and firmware, and multipathing module or DSM. The guidance here covers VMware ESXi and Microsoft Windows Server/Hyper-V, but it does not establish a release-by-release compatibility matrix or a detailed Red Hat Virtualization procedure.
As an Amazon Associate I earn from qualifying purchases.
Record the following for the change plan:
- Hypervisor release, array model and firmware, and the supported NMP configuration or Windows DSM.
- Host initiator IQN, target portal addresses, intended LUN/device identifier, and expected number of paths.
- Which host adapter, switch route, and storage target or controller each path uses.
- The array-supported path selection policy and the expected behavior when a path or controller is unavailable.
A path count alone does not establish redundancy. Two sessions can still share a host NIC, switch, or storage-side failure domain. Build independent routes to the extent the array’s supported topology allows. Broadcom’s ESXi example uses separate iSCSI VMkernel ports mapped to dedicated physical NICs; Microsoft likewise recommends different network adapters for redundant iSCSI connections.
Choose the platform-specific multipathing mechanism
| Platform | Host-side control | Key design check | Evidence to inspect |
|---|---|---|---|
| VMware ESXi | Software iSCSI networking, applicable iSCSI port binding, and NMP SATP/PSP policy | Whether the supported network design requires port binding, and whether the selected policy is supported for this array | Device path state and NMP failover messages in host logs |
| Windows Server / Hyper-V | MPIO feature plus Microsoft DSM or the array vendor’s DSM | Whether each intended iSCSI connection uses a different network adapter and the device is claimed by the appropriate DSM | Available MPIO hardware, settings, path state, and Windows event logs |
These controls are not interchangeable. Do not carry an ESXi policy recommendation over to Windows, or apply Windows DSM settings to a different array. The array vendor’s supported configuration should determine the policy and any device-specific settings.
#1 Best Overall
- [I210AT CHIPSET] Engineered with the industrial-grade I210AT controller for unmatched stability and native OS support including Server, , and VMware ESXi without additional drivers.
- [TRUE GIGABIT PERFORMANCE] Delivers full 1000Mbps bandwidth with auto-negotiation for seamless integration into existing networks while supporting jumbo frames and advanced features like PXE boot and WOL.
- [M.2 A+E KEY DESIGN] Space-saving form factor ideal for compact systems including mini-ITX motherboards, industrial PCs, and embedded applications where PCIe slots are limited.
- [ENTERPRISE-GRADE FEATURES] Supports server functions including iSCSI, FCoE, DPDK, and VLAN tagging - perfect for virtualization hosts, NAS builds, and network appliances.
- [BROAD COMPATIBILITY] Verified operation across 7/8/10, Server 2008-2016, FreeBSD, distributions, and VMware ESXi for flexible deployment scenarios.
Configure iSCSI multipathing on VMware ESXi
Set up the software iSCSI network
- Confirm that the host’s software iSCSI adapter and the intended VMkernel interfaces are configured for the array’s supported design.
- Where the supported design requires software iSCSI port binding, bind each relevant VMkernel port to the software iSCSI adapter. This matters in applicable multi-homing designs: Broadcom warns that same-subnet routing can prevent traffic from moving to another interface as expected when the necessary binding is absent.
- Check that the intended VMkernel ports use the planned independent physical NICs and network routes. Confirm target reachability over each intended route rather than assuming that multiple configured interfaces create independent paths.
Rescan and check the device policy
- Rescan the host’s storage adapters using the procedure supported for your ESXi release and confirm that the intended device appears.
- Check that the device shows the expected paths and that each path is in the expected state. If paths are missing or degraded, investigate the adapter-to-VMkernel mapping, physical network route, target portal, and array presentation before changing policy.
- Verify the array-recommended NMP SATP and path selection policy (PSP). Broadcom documents Most Recently Used (MRU) as the default for LUNs from active/passive arrays, but that is not a universal recommendation: confirm the policy supported for the specific array before changing it.
For path troubleshooting, use esxcfg-mpath -b -d <device-ID> to inspect path state for the device. The NMP failover sequence can also be investigated in /var/log/vmkernel.log or /var/log/messages. Interpret the output against the device identifier and the intended topology.
Configure iSCSI multipathing on Windows Server and Hyper-V
Enable MPIO and claim the storage device
- Enable the MPIO feature on each applicable host. In a VMM-managed fabric, Microsoft says MPIO must be enabled on each host that uses iSCSI or Fibre Channel storage.
- Ensure the discovered iSCSI device is supported and claimed by the correct DSM. Use Microsoft DSM where appropriate, or install the array vendor’s DSM and follow its configuration guidance.
- If MPIO was enabled only after the host was added to VMM, Microsoft notes that device hardware IDs need manual configuration or a vendor DSM should be installed.
Check adapters and MPIO state
Configure redundant iSCSI connections over different network adapters. Microsoft’s MPIO troubleshooting guidance states: “For redundancy, make sure that each iSCSI connection uses a different network adapter.” Then use the applicable host tools to inspect configuration:
Rank #2
- [M.2 A+E COMPATIBILITY] Built with an M.2 A+E Key interface this network adapter is designed for compatible industrial computers embedded systems and server devices needing a single port RJ45 wired connection.
- [1000MBPS GIGABIT SPEED] Supports 1000 100 and 10 Mbps auto negotiation to match existing Ethernet networks smoothly delivering stable wired performance for data transfer office networking and device expansion.
- [I210AT STABLE CHIPSET] Equipped with the I210AT solution this adapter offers high performance strong stability and broad compatibility making it a dependable choice for professional networking and server use.
- [BROAD OS SUPPORT] Compatible with 7 8 8.1 10 Server 2008 Server 2012 Server 2016 FreeBSD and VMware ESXi to support varied deployment requirements.
- [ADVANCED NETWORK FEATURES] Supports PXE DPDK WOL iSCSI FCoE Jumbo Frame VLAN IEEE 1588 and Ethernet suitable for industrial control embedded computing digital multimedia and network equipment.
Get-MPIOAvailableHWlists MPIO-capable hardware information available to the host.Get-MPIOSettingdisplays MPIO settings.mpclaimcan be used to inspect or manage MPIO claiming, as appropriate for the task and Windows version.
Confirm that the intended device is claimed by the expected DSM and that the physical and logical paths are present and healthy. Consult the hardware vendor for path verification settings; do not substitute settings from an unrelated array or service.
Verify failover with a controlled test
A healthy path display is necessary but does not prove that I/O will continue when a route fails. Test in a maintenance window after obtaining workload, cluster, and storage-owner approval. Follow the platform, array, and change-management procedures for taking a path out of service; do not improvise by disconnecting production storage.
- Capture the baseline. Record the device/LUN identifier, path count and state, active or preferred path where the platform reports it, and current application I/O behavior.
- Identify the precise failure being tested. Record the NIC, switch link, target portal, or controller-side route to be removed and the intended alternate path. Test one failure domain at a time so the result is interpretable.
- Take the selected path out of service using the supported maintenance procedure. Record the exact time and observe both hypervisor path state and workload I/O. On ESXi, Broadcom notes that disabling the active path forces failover; use this only within a supported controlled test.
- Confirm that I/O continues over an alternate path. Inspect the device path state and relevant logs, and check the application or workload for errors, retries, pauses, or interruption. A path transition in a log alone does not establish that the application experienced no impact.
- Restore the path and verify recovery. Confirm that the recovered route returns to a healthy state and that the final path count and policy match the intended design.
Do not promise a universal failover time. Microsoft describes troubleshooting scenarios in which delay can exceed 30 seconds; this is an operational scenario, not a benchmark or expected duration for every host and array. Set acceptance criteria from the supported design and measure behavior on the actual workload.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Capture evidence and troubleshoot what the test shows
Keep a before-and-after record for each failure test. Include:
- Device/LUN identifier and path count and state before, during, and after the test.
- The specific component or route removed from service and the timestamp of removal and restoration.
- ESXi kernel or system log entries, or Windows event log entries, that show path changes and errors.
- Observed time until alternate-path I/O and until the failed path returned to healthy state.
- Application-level I/O effects, including errors, retries, pauses, or recovery behavior.
- Confirmation that the final path policy and device claim remain as designed.
If a test fails, use the symptom to narrow the investigation:
Quick wins for a faster PC:
Clear out junk files and repair common Windows errorsFree Scan →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Quick Recap
Best Value
- ✅Friendly reminder: Please make sure that there is an M.2 slot on the motherboard to use it, and some PC motherboards do not support PCIE and M.2 slots to work at the same time, please confirm before placing an order to avoid unnecessary trouble✅
- Controller：Original Mellanox ConnectX-4 Lx controller,which provide true hardware-based I/O isolation with unmatched scalability and efficiency, achieving the most cost-effective and flexible solution for Web 2.0, cloud, data analytics, database, and storage platforms.
- PCI Express v3.0(8.0GT/s) x8, comes with M.2SFF8087 connector and 35cm 8087 cable.
- iPXE, DPDK, iSCSI, UEFI, TCP/IP, UDP/IP, Jumbo Frames, RDMA（RoCE v1, RoCE V2）,ASAP², VMDq, SR-IOV, RSS, IPsec supported.
- Operating Systems Supported: Windows; Windows Server; Linux Stable Kernel version; Ubuntu; Vmware ESXi; Citrix XenServer; Deepin; RHEL/CENTOS; Freebsd; OFED AND WINOF-2; Mikrotik; Debian; BCLINUX; ALIOS; Euler; KYLIN; etc.
- Fewer paths than expected: Check target/LUN presentation, initiator configuration, portal reachability over each intended network route, adapter mapping, and DSM or NMP device claiming.
- Paths exist but share a failure domain: Compare the host NIC, switch route, and target/controller route for every path. Multiple sessions do not provide meaningful redundancy if they converge on the component being tested.
- ESXi does not use the expected alternate interface: Recheck whether the network topology requires software iSCSI port binding, especially in the same-subnet multi-homing case, then verify VMkernel-to-physical-NIC mapping.
- Windows paths are missing or unhealthy: Confirm MPIO is enabled on the host, the intended adapter separation is present, and the device is claimed by the supported DSM. Review Windows events and ask the vendor for recommended path verification settings.
- Unexpected policy or recovery behavior: Compare the configured NMP SATP/PSP or Windows DSM settings with the exact array’s support guidance. Do not transplant settings from Broadcom’s vSAN iSCSI Target Service examples or another array; those examples are scoped to that service and its supported configuration.
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
