vSAN detected and fixed a medium or checksum error for component ######## on disk group ########
search cancel

vSAN detected and fixed a medium or checksum error for component ######## on disk group ########

book

Article ID: 405199

calendar_today

Updated On:

Products

VMware vSAN VMware vSAN 7.x VMware vSAN 8.x

Issue/Introduction

Symptoms:

  • The ESXi host logs an explicit "Checksum error detected" event.

  • From ESXi, var/run/log/hostd.log, there will be an event with "Checksum error detected"

    • YYYY-MM-DDTHH:MM:SS.ZZ cpu17:2099249)WARNING: LSOM: LSOMScrubReadComplete:2838: Throttled: Checksum error detected on component ########-####-####-####-########, comp offset 179160940544 (computed CRC ######## != saved CRC ########
    • YYYY-MM-DDTHH:MM:SS.ZZ info hostd[2105242] [Originator@6876 sub=Vimsvc.ha-eventmgr] Event 588 : vSAN detected and fixed a medium or checksum error for component ########-####-####-####-######## on disk group ########-####-####-####-########.

  • Urgent vSAN traces capture status changes indicating disk or object health transitions (e.g., DOMTraceLeafSSDHealthChangeCb and DOMTraceLeafDiskDecomStateCb), occasionally flagging instances where the owner health is false ('ownerHealthy': False) or returning specific error codes like 'VMK_NOT_FOUND'.

    • From ESXi, var/run/log/vsantraces, there will be events as below:

      YYYY-MM-DDThh:mm:ss.msZ vsantraceUrgent[2098686]: 2026-05-06T00:08:57.201086 [35359646] [cpu34] [LEAF] DOMTraceLeafSSDHealthChangeCb:2594: {'obj': 0x45da29489540, 'ssdUUID': '#######-####-####-####-###########', 'errorCode': 'VMK_OK', 'ownerHealthy': True, 'healthFlags': 0x0}
      YYYY-MM-DDThh:mm:ss.msZ vsantraceUrgent[2098686]: 2026-05-06T00:04:14.471535 [35355688] [cpu7] [LEAF] DOMTraceLeafDiskDecomStateCb:3651: {'obj': 0x45ba426b0a40, 'subscrUUID': '#######-####-####-####-###########', 'connState': 'CONNECTED', 'ownerHealthy': False, 'decomState': 0x0, 'majorVersion': 0, 'errorCode': 'VMK_NOT_FOUND'}

Environment

VMware vSAN (All Versions)

Cause

The alerts are triggered when vSAN detects a data checksum mismatch, which is primarily caused by physical block degradation (SCSI Medium Errors) on a drive, or by outdated storage controller or drive firmware misreporting data.

From ESXi, var/run/log/vmkernel.log there is an error event for Device with "Medium error"

YYYY-MM-DDThh:mm:ss.msZ In(182) vmkernel: cpu18:2097994)ScsiDeviceIO: 4697: Cmd(0x45bbe55c5080) 0x28, CmdSN 0xce37bfa from world 0 to dev "naa.###################" failed H:0x0 D:0x2 P:0x0 Valid sense data: 0x3 0x11 0x1 Medium Error, LBA: ######
YYYY-MM-DDThh:mm:ss.msZ In(182) vmkernel: cpu2:2097992)ScsiDeviceIO: 4697: Cmd(0x45bbe930c940) 0x28, CmdSN 0xce37c24 from world 0 to dev "naa.###################" failed H:0x0 D:0x2 P:0x0 Valid sense data: 0x3 0x11 0x1 Medium Error, LBA: #######
YYYY-MM-DDThh:mm:ss.msZ In(182) vmkernel: cpu0:2097994)ScsiDeviceIO: 4697: Cmd(0x45bc06571180) 0x28, CmdSN 0xce37c2e from world 0 to dev "naa.###################" failed H:0x0 D:0x2 P:0x0 Valid sense data: 0x3 0x11 0x1 Medium Error, LBA: #######

Resolution

If logs confirm a physical Medium Error (Sense Key 0x03), the affected drive must be replaced.

  1. Identify the failed physical disk NAA ID and its physical slot location using the command: esxcli storage core device physical get -d <NAA_ID>.
  2. Place the affected ESXi host into Maintenance Mode using the Ensure Accessibility option.
  3. Engage the hardware vendor to replace the defective physical disk in the identified slot.
  4. Once replaced, add the new disk back to the vSAN disk group.
  5. Exit Maintenance Mode and monitor the vSAN Resyncing Objects view to ensure all components return to an Active/Healthy state.
  6. For detailed replacement steps, refer to Replace a Capacity Device in vSAN Cluster.

If no disks with medium errors are found from cluster logs, open a case with Broadcom Support. and upload the logs for further investigation.

Additional Information

Powered by Wolken Software