BIOS can't see cloned HDD

I’ve 42.2 in an old box I’m using as a local backup server with SyncThing (works like a charm, great software).

Motherboard is an Asus M2N-MX SE PLUS with AMI BIOS version 1804 from August 2010 (last version available). 1st edition MOBO manual is dated September 2007. CPU is AMD Sempron LE-1200, 2GB RAM installed.

This box has a working 250 GB Samsung HDD connected to a SATA2 (SATA 3G) internal port:


Gparted device info
Model: ATA SAMSUNG HD250HJ
Size: 232.89 GiB
Path: /dev/sda
Partition table: msdos
Heads: 255
Sectors/Track: 63
Cylinders: 30401
Total sectors: 488397168
Sector size: 512

# fdisk -l /dev/sda
Disco /dev/sda: 232,9 GiB, 250059350016 bytes, 488397168 setores
Unidades: setor de 1 * 512 = 512 bytes
Tamanho de setor (lógico/físico): 512 bytes / 512 bytes
Tamanho E/S (mínimo/ótimo): 512 bytes / 512 bytes
Tipo de rótulo do disco: dos
Identificador do disco: 0x000dd089

Dispositivo Inicializar   Início       Fim   Setores Tamanho Id Tipo
/dev/sda1                   2048   4208639   4206592      2G 82 Linux swap / Solaris
/dev/sda2   *            4208640  35663909  31455270     15G 83 Linux
/dev/sda3               35665920 488396799 452730880  215,9G 83 Linux

I’ve an available 500 GB HDD where I cloned the 250 GB disk using an USB3 dual disk docking station.

I’ve previously cloned the same 500 GB HDD to a larger one, and it worked OK in the 500 GB HDD original machine (a core i3).

500 GB 7200.12 Seagate Barracuda connected to USB docking station, after increase of sda3 partition.
HDD date code is 11285 (build date Jan 11 2011), firmware version CC46


Gparted device info
Model: ST350041 8AS (connected to USB docking station)
Size: 465.76 GiB
Path: /dev/sdc
Partition table: msdos
Heads: 255
Sectors/Track: 63
Cylinders: 60801
Total sectors: 976773168
Sector size: 512

# fdisk -l /dev/sdc
Disco /dev/sdc: 465,8 GiB, 500107862016 bytes, 976773168 setores
Unidades: setor de 1 * 512 = 512 bytes
Tamanho de setor (lógico/físico): 512 bytes / 512 bytes
Tamanho E/S (mínimo/ótimo): 512 bytes / 512 bytes
Tipo de rótulo do disco: dos
Identificador do disco: 0x000dd089

Dispositivo Inicializar   Início       Fim   Setores Tamanho Id Tipo
/dev/sdc1                   2048   4208639   4206592      2G 82 Linux swap / Solaris
/dev/sdc2   *            4208640  35663909  31455270     15G 83 Linux
/dev/sdc3               35665920 976773119 941107200  448,8G 83 Linux

Now, after switching the 250 GB HDD for the cloned 500 GB HDD connected to the same SATA port, it is not detected by the BIOS, and I can’t see why… The same old BIOS had no problem detecting a newer 1TB Caviar Green HDD on the same SATA port.

Samsung site says this 500 GB HDD may require a jumper in the back to be recognized by older BIOS, I tried that but no change.

With the HDD docking station connected to an USB2 port on the old computer, when plugging the 500 GB drive in the docking station and mounting the third partition, dmesg gives:


[51093.502400] usb 4-1: new SuperSpeed USB device number 5 using xhci_hcd
[51093.514498] usb 4-1: New USB device found, idVendor=152d, idProduct=0551
[51093.514500] usb 4-1: New USB device strings: Mfr=1, Product=2, SerialNumber=5
[51093.514501] usb 4-1: Product: USB to ATA/ATAPI Bridge
[51093.514502] usb 4-1: Manufacturer: JMicron
[51093.514503] usb 4-1: SerialNumber: 580FFFFFFFFF
[51093.514861] usb-storage 4-1:1.0: USB Mass Storage device detected
[51093.514992] scsi13 : usb-storage 4-1:1.0
[51094.516593] scsi 13:0:0:0: Direct-Access     ST350041 8AS                   PQ: 0 ANSI: 2 CCS
[51094.516752] sd 13:0:0:0: Attached scsi generic sg7 type 0
[51094.517165] sd 13:0:0:0: [sdf] 976773168 512-byte logical blocks: (500 GB/465 GiB)
[51094.517863] sd 13:0:0:0: [sdf] Write Protect is off
[51094.517865] sd 13:0:0:0: [sdf] Mode Sense: 28 00 00 00
[51094.518154] sd 13:0:0:0: [sdf] No Caching mode page found
[51094.518155] sd 13:0:0:0: [sdf] Assuming drive cache: write through
[51094.547135]  sdf: sdf1 sdf2 sdf3
[51094.548210] sd 13:0:0:0: [sdf] Attached SCSI disk
[51105.516364] EXT4-fs (sdf3): mounted filesystem with ordered data mode. Opts: (null)

I can read and write to the mounted partition, no dmesg or GUI (dolphin) errors/warnings.

The only thing that occurs to me is that the 500 GB HDD may have been originally formated as gtp - something the docking station can translate to whatever protocol USB uses. But when connected directly it should’nt have been detected by the BIOS anyway? I don’t know the docking station cloning level - it’s a made-in-China thing that works quite well but has documentation issues… Anyway, from gparted and fdisk infos the partitioning is msdos, boot flag is in the same partition, etc.

What am I not seeing?

TIA

Do you really mean detected by the bios - I suspect you really mean not detected by the kernel. You should be able to find mounts for each partition on a disk via dmesg but if one is mounted then the disk was detected.

In order to get some help I suspect you will need to post 3 things. First open the console and enter lsblk and copy past the output on here between code tags - the # icon gives those. Then reboot and the same with the other disk in place.

Then edit /ect/fstab and copy past that the same way.

You can do both as a normal user. Changing drives on a Linux system can be tricky.

;)Not that I can help any further but I’m pretty sure anyone that can will want more info. Those should cover it.

John

If your machine fails to boot so you can’t use lsblk with the new drive in as it looks like you have a usb dock so do one with the new drive in it.

You can also check that the disk is formatted correctly while it’s in the dock - use yast - partitioner. This comes up with a warning but can be aborted so no harm done.

Before you remove the disk from the dock open each partition with file manager. That then allows the icons to be used to eject them. If you click these before opening them with file manager some rather strange things can happen. I had the system suddenly start using one of them suggesting it over mounted the one it should have been using.

I should have added does the machine actually boot up with the new drive in it to my first post.

John

Hi John,

No, I mean the BIOS. With the HDD on the same port as the older one, BIOS only see the DVD drive (so no HDD to load the kernel from). With the older HDD, BIOS see both HDD and DVD drives.

Thank you for you suggestions. I’m aware this is not a common issue, very few google hits and none quite like it.

I’ve yet to find some time to put the backup machine offline and make some tests, like mounting the “new” HDD back in the newer CPU and see if it’s still recognized (most probably); reformat as MBR and see if the older CPU recognizes the MBR-formatted empty HDD (possible); play around with AHCI BIOS settings (improbable), etc.

AFAICS, since the HDD works normally through the USB dock it should be a configuration rather than a hardware issue.

Further testing this weekend: HDD shows erratic behavior in two different MOBOs, one legacy BIOS and another UEFI. Sometimes it was detected, sometimes not.

I could not discern a reason - smart full test (~ 84 min) ran without issues, reformatted partition to msdos (it was gtp), then it was seen in the older CPU, then not anymore after reboots.

Thus this HDD was re-purposed as a door stop…