ESXi slow responding due to storage path become standby

Products

• VMware vSphere ESXi

Issue/Introduction

• SAS storage from Dell SCv3020 is connected to ESXi host. 
• ESXi host become slow responding and displays extremely high CPU usage (e.g 500%)
• Virtual machine become not responding.
• In the ESXi /var/run/log/vmkernel.log:
  
  2025-05-18T01:45:39.947Z Wa(180) vmkwarning: cpu8:2097808)WARNING: NMP: nmpDeviceAttemptFailover:608: Retry world restore device "<lun-naa-id>" no more commands to retry
2025-05-18T01:45:39.954Z Wa (180) vmkwarning: cpu37:2097807)WARNING: NMP: nmp_SelectPathAndIssueCommand: 5632: PSP selected path "vmhba3:C0:T1:L6" in a bad state (standby) on device "<lun-naa-id>"

Environment

VMware vSphere ESXi 7.0.x

Cause

The Dell storage SCv3020 supports PSP is Round Robin. If the PSP is other, the path to the SAS storage become standby. 

Resolution

To resolve the issue, change the PSP for the storage to VMW_PSP_RR

1. SSH to ESXi host
2. Check the PSP for the LUN:
   
   esxcli storage nmp device list -d <lun-naa-id>
   
   For example:
   
   # esxcli storage nmp device list -d naa.####cfc000000####c2007e03f93####
   Device Display Name: FreeNAS iSCSI Disk naa.####cfc000000####c2007e03f93####
   Storage Array Type: VMW_SATP_ALUA
   Storage Array Type Device Config: {implicit_support=on; explicit_support=off; explicit_allow=on; alua_followover=on; action_OnRetryErrors=off; {TPG_id=1,TPG_state=AO}}
   Path Selection Policy: VMW_PSP_MRU
   Path Selection Policy Device Config: Current Path=vmhba65:C0:T0:L0
   Path Selection Policy Device Custom Config:
   Working Paths: vmhba65:C0:T0:L0
   
   
3. Set the PSP to VMW_PSP_RR:
   
   esxcli storage nmp device set --device <lun-naa-id> --psp VMW_PSP_RR
   
   For example:
   
   esxcli storage nmp device set --device naa.####cfc000000####c2007e03f93#### --psp VMW_PSP_RR

Additional Information

Dell EMC SC Series: Best Practices with VMware vSphere