You can replace the capacity drive in a host when you see an Operation status is down for storage device alert. The alert description says HDD_DOWN_ALERT.

Procedure

  1. Expand the alert and note the rack number, host name, and disk type displayed in the Description field.
  2. On the SDDC Manager Dashboard, click View Details in the Physical Resources section.
  3. Click the affected rack and then click the host name.

    The Host Details page displays host details.

  4. If the host does not belong to a workload domain (the Workload Domain field is blank), pull the disk out of the host and replace it with a new disk. For details, refer to the vendor documentation.
  5. If the host is part of a workload domain (the Workload Domain field displays the domain name), follow the steps below.
    1. Note the ESX Cluster name.

    2. In vCenter Server, click the vCenter link.

    3. Navigate to the ESX Cluster name you had noted earlier.

    4. Select the VM Storage Policy with Number of Failures to Tolerate greater than zero.

    5. In the Manage tab, select Virtual SAN > General.

    6. In the Virtual SAN is Turned On field, click Edit.

    7. In Add disks to Storage, select Manual and click OK.

    8. Select the host with the failed disk, click the Manage tab and select Disk Management in the Virtual SAN section.

    9. Select the disk group with the failed capacity drive.

    10. Select the failed capacity drive and click Remove selected disk(s) from disk group.

    11. Wait for the disk to be deleted and then remove the disk from the host.

    12. Add the new disk to the host and wait for vCenter Server to detect it.

      If vCenter Server is unable to detect the drive, confirm that the disk is seated properly in the slot and perform a device re-scan.

    13. Select the host with the newly replaced disk, click the Manage tab and select Disk management in the Virtual SAN section.

    14. In the Virtual SAN is Turned On field, click Edit.

    15. In Add disks to Storage, select Automatic and click OK.

    16. Log out of all open systems.