Frequent path redundancy errors are reported for datastores in iSCSI Environment

Products

• VMware vSphere ESXi

Issue/Introduction

• An ESXi host frequently reports path redundancy errors for iSCSI storage devices. The issue is observed in the vSphere UI under Hosts >  Monitor > Events.

  Symptoms:

  • Events log shows the warning: Path redundancy to storage device <naa.ID> degraded. Path <vmhba#:C#:T#:L# > is down.Affected datastores: <Datastore_Name>
  • iSCSI storage connectivity is unstable or degraded.

Environment

VMware vSphere ESXi

iSCSI Storage arrays

Cause

There is a Maximum Transmission Unit (MTU) configuration mismatch between the virtual switch (vSwitch) and the VMkernel network interface (vmk) used for iSCSI traffic.

Specifically, the issue occurs when the VMkernel interface is configured for a larger MTU (e.g., Jumbo Frames or 8800) while the underlying vSwitch (or the further components in the path) is configured for a smaller default MTU (1500). This causes large packets sent by the VMkernel adapter to be dropped at the vSwitch level, leading to path instability.

Resolution

The MTU on the virtual switch must be increased to match or exceed the MTU configured on the VMkernel interface. 

To configure the MTU, you can follow the KB Article iSCSI and Jumbo Frames configuration on VMware ESXi host

Additional Information

 

• For iSCSI traffic, it is best practice to ensure end-to-end Jumbo Frame configuration ( Storage Array -> Physical Switch -> vSwitch -> VMkernel Interface).

• Inconsistent MTU sizes are a common cause of "flapping" paths where paths go up and down frequently.