"Clear partition table" fails with "Failed to update partition table of {disk}" using esx-ui (VMware Host Client).

Products

• VMware vSphere ESXi

Issue/Introduction

• For creating a new VMFS volume on an existing LUN(holds a valid partition table), clearing this partition table is needed but "Clear partition table" operation in VMware Host Client fails with "Failed to update partition table of {disk}".
  
  
• Some storage feature such as volume replication (LUN replication) or disaster recovery solution is involved.
  
  
• In hostd.log we can see following warnings.
  

  YYYY-MM-DDTHH:MM:SS.mmmZ Wa(164) Hostd[2099330]: [Originator@6876 sub=Partitionsvc opID=esxui-<<opid>> sid=<<sid>>ad8f user=root] Unable to setptbl on /vmfs/devices/disks/naa.<<naa.id>>
  YYYY-MM-DDTHH:MM:SS.mmmZ Wa(164) Hostd[2099291]: --> Status: 255
  YYYY-MM-DDTHH:MM:SS.mmmZ Wa(164) Hostd[2099291]: -->  Output: gpt
  YYYY-MM-DDTHH:MM:SS.mmmZ Wa(164) Hostd[2099291]: --> 0 0 0 0
  YYYY-MM-DDTHH:MM:SS.mmmZ Wa(164) Hostd[2099291]: -->
  YYYY-MM-DDTHH:MM:SS.mmmZ Wa(164) Hostd[2099291]: -->  Error: Error: Input/output error during write on /dev/disks/naa.<<naa.id>>
  YYYY-MM-DDTHH:MM:SS.mmmZ Wa(164) Hostd[2099291]: --> SetPtableGpt: Unable to commit to disk
  YYYY-MM-DDTHH:MM:SS.mmmZ Wa(164) Hostd[2099291]: -->
  YYYY-MM-DDTHH:MM:SS.mmmZ Wa(164) Hostd[2099291]: -->

• In vmkernel.log we can see following scsi messages.
  

  YYYY-MM-DDTHH:MM:SS.mmmZ In(182) vmkernel: cpu24:2098051)NMP: nmp_ThrottleLogForDevice:3893: Cmd 0x2a (0x45e98ac42200, 28263597) to dev "naa.<<naa.id>>" on path "vmhbaX:C0:T0:LZZ" Failed:
  YYYY-MM-DDTHH:MM:SS.mmmZ In(182) vmkernel: cpu24:2098051)NMP: nmp_ThrottleLogForDevice:3898: H:0x0 D:0x2 P:0x0 Valid sense data: 0x7 0x22 0x0. Act:NONE. cmdId.initiator=0x430931552380 CmdSN 0x2
  YYYY-MM-DDTHH:MM:SS.mmmZ In(182) vmkernel: cpu24:2098051)ScsiDeviceIO: 4686: Cmd(0x45e98ac42200) 0x2a, CmdSN 0x2 from world 28263597 to dev "naa.<<naa.id>>" failed H:0x0 D:0x2 P:0x0 Valid sense data: 0x7 0x22 0x0

Environment

VMware vSphere ESXi

Cause

vmkernel logs show that SCSI write command ( 0x2a; write(10) in this case ) failed with 'D:0x2' (CHECK CONDITION) and sense keys '0x7' (DATA PROTECT).  Therefore writing a new partition table to the storage device failed and UI showed "Failed to update partition table of {{disk}}". And this suggests that this storage LUN is write protected for some reason such as "LUN Replication" or "DR (Disaster Recovery) Solution". 

Resolution

In most case this kind of protection may be applied by storage's device replication or DR solution. So need to involve the storage vendor or solution vendor for further investigation.