Start by determining whether iSCSI has lost a session or path, whether multipathing has stopped providing redundancy, or whether storage is still reachable but responding slowly. Check the host, network, multipath layer, and storage array together before changing settings: a single event, ping, or latency reading rarely identifies the cause. The right fix depends on the hypervisor, network design, and array’s supported host configuration.
First, establish what has failed
Record the affected host, hypervisor version and build, datastore or LUN, affected virtual machines, incident start time, and any recent network, storage, driver, or firmware changes. Note whether the problem is continuous or intermittent and whether other hosts or LUNs are affected. Then classify what the host sees:
As an Amazon Associate I earn from qualifying purchases.
- One path is down: storage may still be accessible, but redundancy may be reduced or gone.
- All paths are down: investigate host connectivity, shared network or target failure, and array availability promptly.
- Paths remain online but I/O is slow: investigate latency, queues, network errors, and array load separately from hard disconnection.
- Paths or sessions repeatedly change state: treat the issue as intermittent and correlate transitions with host logs, network counters, and array events.
This classification helps distinguish a fault isolated to one host or path from congestion or failure affecting a shared network, controller, or array. Microsoft’s iSCSI Storage Connectivity Troubleshooting Guidance – Windows Server lists possible disconnection causes including network instability, MPIO errors, adapter readiness, VLAN or MTU mismatch, older drivers or firmware, and SAN/NAS resource exhaustion; none of those symptoms alone establishes a root cause.
Free tools Windows power users keep installed
One-click scans. No signup required.
Check sessions and multipath state on the host
Windows Server and Hyper-V
Gather state before restarting services or rebuilding connections. These commands expose useful host-side information:
#1 Best Overall
- Full-Scale Professional Network-Attached Storage – Business storage solution with hard drives included and optimized to store, share, and back up data for environments of any size.
- Advanced Hardware and Firmware – Product designed for stability and security, capable of handling heavy data loads without dropping performance.
- Purpose-Built for Data Protection – Secure NAS on closed system with 256-bit drive encryption, two-factor authentication, and flexible backup features to keep your data safe.
- Snapshots for Instant Data Backup and Recovery – Snapshots can be created and used to recover data near instantaneously, with little or no system disruptions, and mitigate ransomware.
- Fast Data Transfers – Native 10GbE port for high-speed file transfers with no cable upgrade needed.
Get-IscsiConnectionandGet-IscsiSessionshow iSCSI connections and sessions.mpclaim -s -dreports MPIO disk and path information.Get-NetAdapterandGet-NetAdapterStatisticsshow adapter state and counters.Get-VMSwitchhelps document the Hyper-V virtual-switch configuration.
Compare the output with the intended target portals, adapters, and paths. Confirm that every expected path is online and maps to the correct target and LUN. Check MPIO configuration on each relevant cluster node rather than assuming all nodes are configured identically. Microsoft recommends different network adapters for iSCSI connections intended to provide redundancy; if only one path is detected, verify the adapter and target configuration and restore the missing connection using the storage vendor’s supported procedure.
Windows disconnect symptoms can be accompanied by Event 157 or iSCSI events 9, 20, 27, 39, or 153. In a failover cluster, CSV pause symptoms can include events 5120, 5142, or 153. Treat these as diagnostic leads, not unique identifiers of a particular failed component. Review the events around the incident alongside host state, adapter and switch counters, and storage-array logs.
VMware ESXi
Inspect vmkernel.log around the incident for iSCSI receive failures, SCSI errors, and path-state transitions. Check device and path state with the supported tools for the ESXi release installed on the host, then compare those records with physical-switch and array evidence. For software-iSCSI connections that flap between offline and online, Broadcom lists possible causes including a target closing TCP sessions, network errors, inconsistent MTU, duplicate target IP addresses, and SAN or array saturation. A successful ping does not establish that the storage path is healthy.
Quick wins for a faster PC:
Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Clear out junk files and repair common Windows errorsFree Scan →Verify the network path end to end
Trace each intended iSCSI path from the host interface to its target portal. Check the configuration and physical health at each point:
- Confirm the intended host adapters, VMkernel or other host interfaces, target portals, VLANs, and routes.
- Ensure the MTU setting is consistent end to end; check Jumbo Frame settings as well as ordinary MTU where applicable.
- Inspect host-adapter and switch-port counters for errors, drops, packet loss, or link flaps; check cabling and switch logs.
- Check for duplicate target IP addresses and verify that the host is connecting to the expected targets.
- Keep storage traffic isolated from management, client, or production traffic where the design calls for separation.
Microsoft’s Windows Server checklist asks, “Are iSCSI, management, and client networks segregated and correctly routed?” For the Windows cluster configuration described in its guidance, Microsoft says not to use the same adapters for iSCSI and production or cluster traffic. Apply that advice in the context of the documented design for your deployment, rather than treating it as a universal network architecture rule. For ESXi software iSCSI, verify the VMkernel-to-physical-NIC mapping and consistent MTU.
If basic checks do not resolve a network question, use a network trace and correlate it with switch and host evidence. For ESXi software-iSCSI flapping that remains unexplained, Broadcom advises capturing TCP traffic for the storage OEM to analyze.
Check MPIO, compatibility, and array health
A visible disk is not proof that redundancy is working. Confirm that each intended path is online, reaches the expected target and LUN, and is recognized by the host’s multipathing software. Check that Windows MPIO/DSM or ESXi multipathing settings match the storage vendor’s host configuration guide. Do not choose a path-selection policy or timeout solely because it appears to improve a different array or topology.
Do these 3 things before closing this tab:
1Clear out junk files and repair common Windows errors2Fix the driver behind crashes, sound loss and screen glitches3Repair Windows errors before they cause bigger problemsVerify that the network and storage adapters, drivers, firmware, hypervisor release, and array software are supported together. Review array controller health, target logs, resource use, and any reported exhaustion. Also confirm SAN zoning and LUN masking where applicable. If the host sees fewer paths than designed, investigate discovery, adapter readiness, zoning, target mapping, and MPIO configuration before assuming the array needs replacement.
Rank #2
- Full-Scale Professional Network-Attached Storage – Business storage solution with hard drives included and optimized to store, share, and back up data for environments of any size.
- Advanced Hardware and Firmware – Product designed for stability and security, capable of handling heavy data loads without dropping performance.
- Purpose-Built for Data Protection – Secure NAS on closed system with 256-bit drive encryption, two-factor authentication, and flexible backup features to keep your data safe.
- Snapshots for Instant Data Backup and Recovery – Snapshots can be created and used to recover data near instantaneously, with little or no system disruptions, and mitigate ransomware.
- Fast Data Transfers – Native 10GbE port for high-speed file transfers with no cable upgrade needed.
For Windows Server 2022 Hyper-V deployments, Microsoft marks LBFO NIC teaming as deprecated and points to Switch Embedded Teaming (SET). Confirm the exact Windows Server version and the vendor-supported network design before changing a teaming configuration.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Diagnose slow I/O separately from disconnects
Collect measurements over the affected time window and determine the scope: one LUN, one path or controller, one host, or the whole array. Compare host-visible device and path latency with network evidence and array-side measurements. Queue pressure, a busy controller, or a problematic path can make guest I/O slow even while sessions and paths appear online.
On Windows Server
Use Performance Monitor to examine disk latency and queue behavior. Check adapter and switch counters, storage distribution across controllers, current drivers and firmware, and whether antivirus or other filter drivers are affecting I/O. In clusters, check CSV and volume access, capacity, physical hardware, and filter drivers. Review cluster logs when applicable. Microsoft also identifies updates and controller distribution as areas to investigate in high-latency cases.
Do not apply filesystem allocation-unit guidance indiscriminately to existing volumes or different workloads; a format or layout change is not a substitute for identifying the measured bottleneck.
On ESXi
Separate device or path latency and queue pressure from network round-trip time. Broadcom describes an ESXi storage performance case covering ESXi 7.x, 8.x, and later in which Round Robin behavior and an IOPS limit are relevant to unbalanced controller utilization. In that article’s context, the default Round Robin IOPS limit is 1000. That is a configuration value, not a performance result or a universal recommendation. Check the array’s supported configuration and workload before changing the limit or path-selection policy.
A separate Broadcom article specific to ESXi 8.x reports a default iSCSI adapter failover duration of 25 to 35 seconds in the case it discusses, and explains that pending I/O can fill a device queue during path recovery and stall guest I/O. This is a scoped vendor case, not a guaranteed failover duration across ESXi releases, adapters, or storage arrays. Do not infer a universal timeout from it.
Apply a targeted fix, then verify recovery
Correct the fault supported by the evidence, one change at a time. Examples include repairing a failed link or adapter, correcting a demonstrated MTU, VLAN, route, or target-mapping mismatch, restoring MPIO discovery, installing a supported driver or firmware version, or addressing array-side resource exhaustion. Change path-selection or timeout parameters only when the host and storage vendors’ guidance supports the setting for the specific array and topology.
PC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11Crashes, No Sound, or Screen Glitches?
Random freezes, missing sound and display glitches usually trace back to one bad driver. Find and replace yours safely.Free scan · under a minute- Preserve the baseline: retain relevant host logs, session and path output, adapter and switch counters, array evidence, and the configuration before the change.
- Make the supported correction: avoid stacking unrelated setting changes, which makes it harder to identify the cause and can introduce another failure.
- Confirm redundancy: verify that all intended sessions and paths return and map to the expected targets and LUNs.
- Recheck service: repeat the original health observation or workload and confirm that path errors, latency, and guest symptoms have normalized.
- Escalate with evidence: if the issue persists, provide the host, network, multipath, and array records together to the relevant platform or storage support team.
For a Windows cluster, perform path and MPIO checks on every node and confirm shared-volume access and capacity. For either platform, keep the before-and-after configuration with the incident record so that any remaining problem can be compared against the original state.
Quick Recap
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.
