Permanent Device Loss (PDL) state occurs for storage paths following a reboot of an HPE Elektra array after a hard panic where the LUNs go through a forced consistency check

Products

• VMware vSphere ESXi
• VMware vSphere ESX 8.x
• VMware vSphere ESX 7.x
• VMware vSphere ESXi 8.0

Issue/Introduction

A VCF Administrator observed some storage paths report a PDL state following the reboot of an HPE Elektra array while the array is still performing a consistency check on its own LUNs, which HPE Elektra calls Virtual Volumes (VV):

2025-09-28T01:56:55.584Z Wa(180) vmkwarning: cpu15:2097578)WARNING: VMW_SATP_ALUA: satp_alua_issueInquiry:101: Target reported LUN_NOT_CONNECTED / NO_DEVICE with PQ: 0x1, PDT: 0x1f, path: vmhba64:C0:T1:L1
2025-09-28T01:56:55.584Z Wa(180) vmkwarning: cpu15:2097578)WARNING: VMW_SATP_ALUA: satp_alua_getTargetPortInfo:190: Could not get page 83 INQUIRY data for path "vmhba64:C0:T1:L1" - Device is permanently unavailable (195887410)
2025-09-28T01:56:55.584Z Wa(180) vmkwarning: cpu15:2097578)WARNING: VMW_SATP_ALUA: satp_alua_issueInquiry:101: Target reported LUN_NOT_CONNECTED / NO_DEVICE with PQ: 0x1, PDT: 0x1f, path: vmhba65:C0:T0:L1
2025-09-28T01:56:55.584Z Wa(180) vmkwarning: cpu15:2097578)WARNING: VMW_SATP_ALUA: satp_alua_getTargetPortInfo:190: Could not get page 83 INQUIRY data for path "vmhba65:C0:T0:L1" - Device is permanently unavailable (195887410)
2025-09-28T01:56:55.584Z Wa(180) vmkwarning: cpu15:2097578)WARNING: VMW_SATP_ALUA: satp_alua_issueInquiry:101: Target reported LUN_NOT_CONNECTED / NO_DEVICE with PQ: 0x1, PDT: 0x1f, path: vmhba65:C0:T1:L1
2025-09-28T01:56:55.584Z Wa(180) vmkwarning: cpu15:2097578)WARNING: VMW_SATP_ALUA: satp_alua_getTargetPortInfo:190: Could not get page 83 INQUIRY data for path "vmhba65:C0:T1:L1" - Device is permanently unavailable (195887410)
2025-09-28T01:56:55.584Z Wa(180) vmkwarning: cpu15:2097578)WARNING: VMW_SATP_ALUA: satp_alua_issueInquiry:101: Target reported LUN_NOT_CONNECTED / NO_DEVICE with PQ: 0x1, PDT: 0x1f, path: vmhba64:C0:T0:L1
2025-09-28T01:56:55.584Z Wa(180) vmkwarning: cpu15:2097578)WARNING: VMW_SATP_ALUA: satp_alua_getTargetPortInfo:190: Could not get page 83 INQUIRY data for path "vmhba64:C0:T0:L1" - Device is permanently unavailable (195887410)

Environment

ESXi (All versions)
HPE Elektra array

Cause

When the VVs go through consistency checks on the HPE Elektra array after the array panic, they will not be available for access as they are not exported by the array OS, so the LUNs would not be available and the array will constantly return the PDL condition to the ESXi host:

025-09-28T01:58:09.611Z cpu72:4892562)VMW_SATP_ALUA: satp_alua_issueCommandOnPath:887: Path "vmhba3:C0:T2:L0" (PERM LOSS) command 0xa3 failed with status Device is permanently unavailable. H:0x0 D:0x2 P:0x0 Valid sense data: 0x5 0x25 0x0.

If the VVs are unable to reach a "predefined" state in the initial VV scan where they are created in the VV layer on the Elektra array and can provide basic information to the SCSI layer, they are not exported prior to the host ports opening up. This can result in the array constantly returning the PDL state for those storage paths.

Resolution

Even when the consistency check completes on the array, the LUNs are not properly exported and available to the ESXi hosts until the initiators in the host perform a fabric logout/login to the Elektra array ports. While an ESXi reboot would certainly result in a fabric logout/login for the HBAs, the fabric connection can be simply be reset from the ESXi command line:

# esxcli storage san fc reset -A vmhbaX

For more information on how to use this command, please refer to KB 324547

HPE Elekta engineering in reviewing ways to change this behavior from the array side so a future code release from HPE Elektrais expected to change this behavior. For now, resetting the fabric connection for the ESXi HBAs after the Elektra array has finished its consistency check is the only workaround.