This article explains how to identify and address predictive failure drives
What is Predicted Drive Failure and how to solve it?
Periodically, a Patrol Read is launched on a Virtual Disk. This is a consistency check on the RAID and the hard drives composing the Virtual Drive.
If the consistency check finds bad blocks, it marks them as bad and reallocates the information to another section of this disk.
A disk is always a bit bigger than specified, so it has room to relocate some bad blocks. This can also be a firmware or a reporting issue so it is recommended to update the firmware before any drive replacement.
Perform a backup before any steps are taken.
Perform firmware updates to the
Hard Drive,
RAID controller, BIOS, and
iDRAC to ensure that they are running the latest firmware. This is to ensure that all reporting information is up-to-date and correct.
More information about firmware update is located in the Dell article
Updating Firmware and Drivers on Dell PowerEdge Servers.
Note: Reboot is required to complete firmware install for a RAID card.
If you are running Dell
OpenManage Server Administrator, ensure that you are running the latest version of OMSA. See Dell article
Support for Dell OpenManage Server Administrator (OMSA)
Solution:
If the hard drive is still showing predictive failure after updates,
a hard drive replacement is required.
Export the RAID controller log using Dell articles:
Send the log to
Dell Technical Support for further review.
Offline the disk and replace the hard drive.
Replace the drive with a new physical disk drive of equal or higher capacity and perform the Replaced Member operation. The Replaced Member operation copies data from the source physical disk drive of a virtual disk drive to a target physical disk drive that is not a part of the virtual disk drive.
Predictive Failures can occur due to hard drive errors detected by Self-Monitoring, Analysis, and Reporting Technology (SMART) or due to RAID Array errors such as a Double Fault or a Puncture.
More troubleshooting or maintenance may be required to ensure the health of the array.
SupportAssist Enterprise - Setting up Automated Dispatch: SupportAssist Enterprise: Setting up Automated Dispatch - YouTube