Missing drives after bios update

TNG · September 28, 2019, 12:58am

Hi,

I came across this forum after searching for solutions to an issue I have after updating the bios of my Asus Z10PE D16 WS which reset my raid setting to AHCI and when I went into the bios to put it back to RAID 2 of my 4 HGST 4TB were marked as missing and the volume failed to appear in Windows 10 file explorer. I opened up the IRSTE programme and could see 6 drives where there should be 4 with 2 of them having identical serial numbers to 2 of the others with :0 appended and marked as missing with scsi id of 255. Please see the following report:

System Information
OS name: Microsoft Windows 10 Pro
OS version: 10.0.18362
System name: CONFUCIUS
System manufacturer: ASUSTeK COMPUTER INC.
System model: Z10PE-D16 WS
Processor: GenuineIntel Intel64 Family 6 Model 63 Stepping 2 2.395 GHz
Processor: GenuineIntel Intel64 Family 6 Model 63 Stepping 2 2.395 GHz
BIOS: American Megatrends Inc., 4101

Intel® Rapid Storage Technology enterprise Information
User interface version: 5.5.0.2012
Language: English (United Kingdom)
Intel controller: NVMe*
Number of NVMe* ports: 0
Intel controller: SATA (AHCI)
Number of SATA ports: 1
RAID option ROM version: 0.0.0.0
Intel controller: SATA (AHCI)
Number of SATA ports: 4
SATA driver version: 5.5.0.1334
NVMe* driver version: 5.5.0.1360
ISDI version: 5.5.0.2012

Storage System Information
RAID Configuration

Array Name: sSATA_Array_0000
Size: 14,904.09 GB
Available space: 0.02 GB
Number of volumes: 1
Volume member: RAIDSTORAGE
Number of array disks: 4
Array disk: PK1334PCK2XMMS
Array disk: PK2338P4HGXZKC
Array disk: PK2338P4H75RUC:0
Array disk: PK2338P4HJJB5C:0
Disk data cache: Enabled

Volume name: RAIDSTORAGE
Status: Failed
Type: RAID 5
Size: 11,178.06 GB
System volume: No
Data stripe size: 128 KB
Write-back cache: Enabled
Initialized: No
Close RAID Write Hole: Off
Parity errors: 0
Blocks with media errors: 0
Physical sector size: 4,096 Bytes
Logical sector size: 512 Bytes

Hardware Information

Controller name: NVMe* Devices
Type: NVMe*
Mode: RAID
Number of volumes: 0
Number of spares: 0
Number of available disks: 0
Rebuild on Hot Insert: Disabled
Manufacturer: 8086
Model number: 2F9C
Product revision: 2

Controller name: Intel(R) C600+/C220+ series chipset SATA AHCI Controller
Type: SATA
Mode: AHCI
Number of volumes: 0
Number of spares: 0
Number of available disks: 0
Rebuild on Hot Insert: Disabled
Manufacturer: 8086
Model number: 8D02
Product revision: 5

Controller name: Intel(R) C600+/C220+ series chipset sSATA RAID Controller
Type: SATA
Mode: RAID
Number of volumes: 1
Volume: RAIDSTORAGE
Number of spares: 0
Number of available disks: 2
Rebuild on Hot Insert: Disabled
Manufacturer: 8086
Model number: 2827
Product revision: 5
Direct attached disk: PK1334PCK2XMMS
Direct attached disk: PK2338P4H75RUC
Direct attached disk: PK2338P4HJJB5C
Direct attached disk: PK2338P4HGXZKC
Direct attached disk: PK2338P4H75RUC:0
Direct attached disk: PK2338P4HJJB5C:0

Disk on Controller 2, Port 0
Status: Normal
Type: SATA disk
Location type: Internal
Usage: Array disk
Size: 3,726 GB
System disk: No
Disk data cache: Enabled
Command queuing: NCQ
Model: HGST HDN724040ALE640
Serial number: PK1334PCK2XMMS
SCSI device ID: 0
Firmware: MJAOA5E0
Physical sector size: 4,096 Bytes
Logical sector size: 512 Bytes

Disk on Controller 2, Port 1
Status: Normal
Type: SATA disk
Location type: Internal
Usage: Available
Size: 3,726 GB
System disk: No
Disk data cache: Enabled
Command queuing: NCQ
Model: HGST HDN724040ALE640
Serial number: PK2338P4H75RUC
SCSI device ID: 1
Firmware: MJAOA5E0
Physical sector size: 4,096 Bytes
Logical sector size: 512 Bytes

Disk on Controller 2, Port 2
Status: Normal
Type: SATA disk
Location type: Internal
Usage: Available
Size: 3,726 GB
System disk: No
Disk data cache: Enabled
Command queuing: NCQ
Model: HGST HDN724040ALE640
Serial number: PK2338P4HJJB5C
SCSI device ID: 2
Firmware: MJAOA5E0
Physical sector size: 4,096 Bytes
Logical sector size: 512 Bytes

Disk on Controller 2, Port 3
Status: Normal
Type: SATA disk
Location type: Internal
Usage: Array disk
Size: 3,726 GB
System disk: No
Disk data cache: Enabled
Command queuing: NCQ
Model: HGST HDN724040ALE640
Serial number: PK2338P4HGXZKC
SCSI device ID: 3
Firmware: MJAOA5E0
Physical sector size: 4,096 Bytes
Logical sector size: 512 Bytes

Disk on Controller 2, Port Unknown
Status: Missing
Type: Unknown disk
Location type: Unknown
Usage: Array disk
Size: 0 GB
System disk: No
Disk data cache: Enabled
Command queuing: None
Model:
Serial number: PK2338P4H75RUC:0
SCSI device ID: 255
Firmware:
Physical sector size: 00 Bytes
Logical sector size: 00 Bytes

Disk on Controller 2, Port Unknown
Status: Missing
Type: Unknown disk
Location type: Unknown
Usage: Array disk
Size: 0 GB
System disk: No
Disk data cache: Enabled
Command queuing: None
Model:
Serial number: PK2338P4HJJB5C:0
SCSI device ID: 255
Firmware:
Physical sector size: 00 Bytes
Logical sector size: 00 Bytes

Empty port
Port: 0
Controller SATA (AHCI)
Port location: Internal

-------------------------------------------------
System Report

System Information
OS name: Microsoft Windows 10 Pro
OS version: 10.0.18362
System name: CONFUCIUS
System manufacturer: ASUSTeK COMPUTER INC.
System model: Z10PE-D16 WS
Processor: GenuineIntel Intel64 Family 6 Model 63 Stepping 2 2.395 GHz
Processor: GenuineIntel Intel64 Family 6 Model 63 Stepping 2 2.395 GHz
BIOS: American Megatrends Inc., 4101

Intel® Rapid Storage Technology enterprise Information
User interface version: 5.5.0.2012
Language: English (United States)
Intel controller: NVMe*
Number of NVMe* ports: 0
Intel controller: SATA (AHCI)
Number of SATA ports: 1
RAID option ROM version: 0.0.0.0
Intel controller: SATA (AHCI)
Number of SATA ports: 4
SATA driver version: 5.5.0.1334
NVMe* driver version: 5.5.0.1360
ISDI version: 5.5.0.2012

Storage System Information
RAID Configuration

Array Name: sSATA_Array_0000
Size: 14,904.09 GB
Available space: 0.02 GB
Number of volumes: 1
Volume member: RAIDSTORAGE
Number of array disks: 4
Array disk: PK1334PCK2XMMS
Array disk: PK2338P4HGXZKC
Array disk: PK2338P4H75RUC:0
Array disk: PK2338P4HJJB5C:0
Disk data cache: Enabled

Volume name: RAIDSTORAGE
Status: Failed
Type: RAID 5
Size: 11,178.06 GB
System volume: No
Data stripe size: 128 KB
Write-back cache: Enabled
Initialized: No
Close RAID Write Hole: Off
Parity errors: 0
Blocks with media errors: 0
Physical sector size: 4,096 Bytes
Logical sector size: 512 Bytes

Hardware Information

Controller name: NVMe* Devices
Type: NVMe*
Mode: RAID
Number of volumes: 0
Number of spares: 0
Number of available disks: 0
Rebuild on Hot Insert: Disabled
Manufacturer: 8086
Model number: 2F9C
Product revision: 2

Controller name: Intel(R) C600+/C220+ series chipset SATA AHCI Controller
Type: SATA
Mode: AHCI
Number of volumes: 0
Number of spares: 0
Number of available disks: 0
Rebuild on Hot Insert: Disabled
Manufacturer: 8086
Model number: 8D02
Product revision: 5

Controller name: Intel(R) C600+/C220+ series chipset sSATA RAID Controller
Type: SATA
Mode: RAID
Number of volumes: 1
Volume: RAIDSTORAGE
Number of spares: 0
Number of available disks: 2
Rebuild on Hot Insert: Disabled
Manufacturer: 8086
Model number: 2827
Product revision: 5
Direct attached disk: PK1334PCK2XMMS
Direct attached disk: PK2338P4H75RUC
Direct attached disk: PK2338P4HJJB5C
Direct attached disk: PK2338P4HGXZKC
Direct attached disk: PK2338P4H75RUC:0
Direct attached disk: PK2338P4HJJB5C:0

Disk on Controller 2, Port 0
Status: Normal
Type: SATA disk
Location type: Internal
Usage: Array disk
Size: 3,726 GB
System disk: No
Disk data cache: Enabled
Command queuing: NCQ
Model: HGST HDN724040ALE640
Serial number: PK1334PCK2XMMS
SCSI device ID: 0
Firmware: MJAOA5E0
Physical sector size: 4,096 Bytes
Logical sector size: 512 Bytes

Disk on Controller 2, Port 1
Status: Normal
Type: SATA disk
Location type: Internal
Usage: Available
Size: 3,726 GB
System disk: No
Disk data cache: Enabled
Command queuing: NCQ
Model: HGST HDN724040ALE640
Serial number: PK2338P4H75RUC
SCSI device ID: 1
Firmware: MJAOA5E0
Physical sector size: 4,096 Bytes
Logical sector size: 512 Bytes

Disk on Controller 2, Port 2
Status: Normal
Type: SATA disk
Location type: Internal
Usage: Available
Size: 3,726 GB
System disk: No
Disk data cache: Enabled
Command queuing: NCQ
Model: HGST HDN724040ALE640
Serial number: PK2338P4HJJB5C
SCSI device ID: 2
Firmware: MJAOA5E0
Physical sector size: 4,096 Bytes
Logical sector size: 512 Bytes

Disk on Controller 2, Port 3
Status: Normal
Type: SATA disk
Location type: Internal
Usage: Array disk
Size: 3,726 GB
System disk: No
Disk data cache: Enabled
Command queuing: NCQ
Model: HGST HDN724040ALE640
Serial number: PK2338P4HGXZKC
SCSI device ID: 3
Firmware: MJAOA5E0
Physical sector size: 4,096 Bytes
Logical sector size: 512 Bytes

Disk on Controller 2, Port Unknown
Status: Missing
Type: Unknown disk
Location type: Unknown
Usage: Array disk
Size: 0 GB
System disk: No
Disk data cache: Enabled
Command queuing: None
Model:
Serial number: PK2338P4H75RUC:0
SCSI device ID: 255
Firmware:
Physical sector size: 00 Bytes
Logical sector size: 00 Bytes

Disk on Controller 2, Port Unknown
Status: Missing
Type: Unknown disk
Location type: Unknown
Usage: Array disk
Size: 0 GB
System disk: No
Disk data cache: Enabled
Command queuing: None
Model:
Serial number: PK2338P4HJJB5C:0
SCSI device ID: 255
Firmware:
Physical sector size: 00 Bytes
Logical sector size: 00 Bytes

Empty port
Port: 0
Controller SATA (AHCI)
Port location: Internal

Can anyone advise me on how to fix this please? How do I remove the :0 from the serial numbers of the missing disk so that the array recognises the 2 drives that are there?

Thank you in advance for your help

Edit by Fernando: Put the detailed system information into a “spoiler”(to save space and for better readability)

Bert2018 · October 6, 2021, 9:03am

Hi TNG,

This is an old post but I thought I would reply anyway as I’m sure there are many others with similar or exactly the same issues. I saw in many of the threads regarding raid 5, that they were not successful trying to recover it. If it was raid 1, many were recovered just by deleting the raid in BIOS and then restarting and recreating the array again (after changing the BIOS default setting of AHCI back to RAID.

This happens because the computer BIOS settings revert to default after a clear CMOS, BIOS update etc. If you do not get into the BIOS to change the default back to Raid this is exactly what happens. Fixing a mirror is easy because both drives contain the same information, without any parity. Raid 5 is much more difficult to deal with due to parity, and also the order of the drives is critical. Be careful not to change the ports that they are plugged into. You will see in the Windows GUI for IRST it shows the disks are missing, it has the correct serial number shown except for a :0 or :1 after it. You should also notice that the raid 5 array has 2 disks shown with a yellow triangle. Below the array you will see both your missing drives listed there. They will display the ports they are connected to as well as any other drives your system has that are on that controller.

Before you do anything else, you should be making an image of your disks before any software writes to them. The BIOS/software only thinks they are missing at this point and they are still perfectly working raid 5 members, so be careful.

I don’t know of any any software to manage them in Windows except the Intel IRST software you have installed, and it has few options that are useful in this case. If you look at Intel’s site, you will find that they support Linux. Originally, many years ago with dmraid. They support mdadm currently (for many years now), which gives you full control over them outside of the Windows GUI.

You can just create a bootable USB with Linuxmint on it, as you can configure mdadm without having to install the Linux OS. Mdadm does detect the IRST disks after a reboot, so you may need a minimal install in order to test after. You could install it to a portable USB drive that you can boot from after installation. You will need drives of the same size or larger to make copies of your existing 6 disks including copies of the ones still showing as good. If you have a NAS or other large storage you can copy the files to that. Each 4TB disk will make a 4TB file if uncompressed, regardless of how much data the disk actually contains.

You may wish to look at raid 6 as a minimum, or ZFS for example, to give better odds for recovery.

In Linux you can use dd:

sudo dd if=/dev/sdg of=/mnt/imsmRaid/sdc- PK2338P4H75RUC.img ← use this to copy disk image to file
sudo dd if=/dev/sdg of=/dev/sdm ← use this to clone disk to disk

Making a mistake while trying to recover a raid array can affect all disks, so have a backup for all. At 4TB each, you will be busy for many hours. Your system report shows only 4 GB, not TB, but the model number says 4TB.

Once you have Linuxmint booted, and internet connection, you can run from terminal – sudo apt remove dmraid then sudo apt install mdadm.
Reboot after doing this and mdadm will find all your raid arrays connected to the IRST(e) ports
At terminal prompt run:
sudo mdadm --detail-platform

This will show you the IRST(e) and connected port information
also run this:
sudo fdisk -l

This will show you all the disks. You will need to make note of the /dev/sdX letters for your raid 5.

cd /dev/md/
ls -lia
you will see a list of all the raid disks (not including the raid 5). The two you are interested are called imsm0 and imsm1
type
sudo mdadm --detail /dev/md/imsm0

then type
sudo mdadm --detail /dev/md/imsm1

normally imsm0 is the raid 5 array and imsm1 is your raid 0 or raid 1 array, but verify this.

Your working arrays will display a member array name, the raid 5 will be blank. The raid 5 array should show 4 disks but be missing the 2 disks that had :0 after their serial number.

you can also use “sudo blkid” to show the UUID of every connected disk to verify you will know the correct /dev/sdX device

once all that is done, stop the raid 5 array with:
sudo mdadm --stop /dev/md/imsm0

then run:
sudo mdadm -C --force /dev/md/imsm0 /dev/sdc /dev/sdd /dev/sde /dev/sdf /dev/sdg /dev/sdh -n 6 -e imsm

You will need to use the correct 6 /dev/sdX device identifiers for the 6 HGST drives in order for your raid 5 array.
you may have to enter y to continue

It should respond with:
mdadm: container /dev/md/imsm0 prepared.

then run:
sudo mdadm -C /dev/md/RAIDSTORAGE --force --level=5 --raid-devices=6 /dev/md/imsm0

then run:
cat /proc/mdstat

it should show you that it is starting the resync with an estimated finish time. When it is done you can see if all your files were restored successfully!

[>…] resync = 0.1% (2410076/1953509376) finish=175.3min speed=185390K/sec