VMs restart due to vSphere HA following network switch maintenance.

Products

• VMware vSphere ESXi

Issue/Introduction

Multiple virtual machines (VMs) experience unplanned restarts triggered by vSphere HA isolation responses. This behavior typically occurs during or immediately after physical network switch maintenance or migration involving iSCSI storage traffic. Users may observe that hosts lose connectivity to storage despite physical links remaining up.

Symptoms:

• Multiple VMs restart across different hosts in a cluster.
  
  
• vSphere HA reports an isolation event or "Host failed" status.
  
  
• ESXi hosts report All Paths Down (APD) or Permanent Device Loss (PDL) for iSCSI datastores.
  
  
• In /var/run/log/vmkernel.log, you see entries similar to: 
  
  [vob.iscsi.connection.stopped] iSCSI connection stopped for vmhba## 

[vob.scsi.scsipath.por] Path vmhba##:C#:T#:L# has been restored

WARNING: NMP: nmp_DeviceRequestFastDeviceProbe:###: NMP device "####" state in doubt; requested fast solaris probe

Exact error messages observed in logs:

• • 39496:2026-07-17T10:05:26.312Z In(14) vobd[2097960]:  [iscsiCorrelator] 3148336535856us: [vob.iscsi.connection.stopped] iScsi connection 0 stopped for vmhba64:C#:T#
39497:2026-07-17T10:05:26.318Z In(14) vobd[2097960]:  [iscsiCorrelator] 3148336541835us: [vob.iscsi.connection.stopped] iScsi connection 0 stopped for vmhba64:C#:T#

    
    
  • Wa(180) vmkwarning: cpu61:2098245)WARNING: NMP: nmp_DeviceRequestFastDeviceProbe:235: NMP device "naa.600#############################" state in doubt; requested fast path state update...
Wa(180) vmkwarning: cpu61:2098245)WARNING: NMP: nmp_DeviceRequestFastDeviceProbe:235: NMP device "naa.600#############################" state in doubt; requested fast path state update...
    
    
    
  • 2026-07-17T09:56:05.928Z In(182) vmkernel: cpu1:2097863)NMP: nmp_ThrottleLogForDevice:3898: H:0x5 D:0x0 P:0x0 . Act:EVAL. cmdId.initiator=0x430a5e795c80 CmdSN 0x46f04b
    2026-07-17T09:56:17.541Z In(182) vmkernel: cpu48:2098232)NMP: nmp_ThrottleLogForDevice:3898: H:0x2 D:0x0 P:0x0 . Act:EVAL. cmdId.initiator=0x430f2697fac0 CmdSN 0xffff958bd7ac5730
    
  • iSCSI connection is being marked "OFFLINE" (Event:4)

Environment

• VMware ESXi 8.x  
• VMware ESX 9.x

Cause

The issue is caused by a temporary loss of iSCSI storage connectivity during switch maintenance. Specifically, if "MAC Learning" is disabled on the storage VLANs of the physical switches, the switch may fail to forward iSCSI traffic correctly following a port failover or migration, leading to a storage timeout and triggering vSphere HA.

Resolution

To resolve this issue, the external network infrastructure must be corrected to ensure consistent storage pathing. Involving the inhouse network team for further investigation is required. 

Additional Information

For further assistance, see Contact Broadcom Support.

To speak with a customer representative or a Support Engineer, see Contact Support. Scroll to the bottom of the page and click on your respective region.