Re: Kernel only seeing 2TB of an 8TB disk
Yes, I can tell you more. The hardware as you guessed is not
actually new, it is recycled equipment that is new to me and newly
installed with Debian.
The machine is a Cisco MCS server, possibly a Cisco MCS7800
series. (I'm not where the server is physically now to check.) (Cisco
MCS7800 servers apparently went end of support on 2019-12-31.)
This is a 2U server which is a re-badged IBM machine, the
firmware screens at startup are all IBM branded.
The disks are 2 bays in the front of the server rackmount case.
No external controller.
By the way, I don't recommend these servers, it has an uncommon
display chipset (MGA G200EV) which the kernel doesn't handle
correctly at boot. Hence also the installer doesn't display. I
installed this machine by booting off a Debian 12 live image, then
using parted to partition the disk and debootstrap to install.
I also have another machine (fj, installed a while ago) which I
believe is identical hardware, and which is successfully using a
4TB drive:
Model Family: Western Digital Red
Device Model: WDC WD40EFRX-68N32N0
# dmesg|grep sda
[ 2.484956] sd 1:0:0:0: [sda] 7814037168 512-byte logical
blocks: (4.00 TB/3.64 TiB)
[ 2.484961] sd 1:0:0:0: [sda] 4096-byte physical blocks
[ 2.484974] sd 1:0:0:0: [sda] Write Protect is off
[ 2.484977] sd 1:0:0:0: [sda] Mode Sense: 00 3a 00 00
[ 2.484995] sd 1:0:0:0: [sda] Write cache: disabled, read
cache: enabled, doesn't support DPO or FUA
[ 2.485050] sd 1:0:0:0: [sda] Preferred minimum I/O size 4096
bytes
[ 2.521157] sda: sda1 sda2 sda3 sda4
[ 2.521853] sd 1:0:0:0: [sda] Attached SCSI disk
c.f. the problematic 8G one on fj2:
Model Family: Seagate BarraCuda 3.5 (SMR)
Device Model: ST8000DM004-2U9188
# dmesg|grep sdb
[ 12.977869] sd 7:0:0:0: [sdb] 4294967294 512-byte logical
blocks: (2.20 TB/2.00 TiB)
[ 13.045178] sd 7:0:0:0: [sdb] Write Protect is off
[ 13.045186] sd 7:0:0:0: [sdb] Mode Sense: 73 00 00 08
[ 13.050816] sd 7:0:0:0: [sdb] Write cache: enabled, read cache:
enabled, doesn't support DPO or FUA
[ 13.145348] sdb: sdb1 sdb2 sdb3
[ 13.146139] sd 7:0:0:0: [sdb] Attached SCSI disk
The differences between these two systems are different drives of
course, and the problematic disk is the only disk and therefore
the boot disk, whereas the working one is not the boot disk (there
is a 1T disk sdb which is the boot disk, and I never got the
system booting off the 4T disks, there was previously a pair in
that machine.)
Hardware info:
oot@fj2:/home/installer# lscpu|head
Architecture: x86_64
CPU op-mode(s): 32-bit, 64-bit
Address sizes: 36 bits physical, 48 bits
virtual
Byte Order: Little Endian
CPU(s): 4
On-line CPU(s) list: 0-3
Vendor ID: GenuineIntel
BIOS Vendor ID: Intel(R) Corporation
Model name: Intel(R) Xeon(R)
CPU X3430 @ 2.40GHz
BIOS Model name: Intel(R) Xeon(R)
CPU X3430 @ 2.40GHz CPU @ 2.4GHz
root@fj2:/home/installer# lspci|grep storage
01:00.0 SCSI storage controller: Broadcom / LSI SAS1064ET
PCI-Express Fusion-MPT SAS (rev 08)
Selected lshw info:
fj2
description: Rack Mount Chassis
product: System x3250 M3 -[4251PAB]- (XxXxXxX)
vendor: IBM
version: 05
serial: KQ7F85W
width: 64 bits
capabilities: smbios-2.5 dmi-2.5 smp vsyscall32
configuration: boot=normal chassis=rackmount family=System X
sku=XxXxXxX uuid=48cd79b1-0f24-ef3c-ae22-0901f4b70880
*-core
description: Motherboard
product: 00D9060
vendor: IBM
physical id: 0
version: SIT
serial: 26F0EV
slot: Rear Bottom
...
*-firmware
description: BIOS
vendor: IBM Corp.
physical id: 13
version: -[GYE148AUS-1.11]-
date: 02/09/2011
size: 1MiB
capacity: 4MiB
capabilities: pci pnp upgrade shadowing cdboot
bootselect edd int13floppynec int13floppytoshiba int13floppy360
int13floppy1200 int13floppy720 int13floppy2880 int5printscreen
int9keyboard int14serial int17printer int10video acpi usb
ls120boot zipboot biosbootspecification netboot
...
*-scsi
description: SCSI storage controller
product: SAS1064ET PCI-Express Fusion-MPT SAS
vendor: Broadcom / LSI
physical id: 0
bus info: pci@0000:01:00.0
logical name: scsi7
version: 08
width: 64 bits
clock: 33MHz
capabilities: scsi pm pciexpress msi msix
bus_master cap_list rom
configuration: driver=mptsas latency=0
resources: irq:16 ioport:3000(size=256)
memory:91b10000-91b13fff memory:91b00000-91b0ffff
*-disk
description: ATA Disk
product: ST8000DM004-2U91
physical id: 0.0.0
bus info: scsi@7:0.0.0
logical name: /dev/sdb
version: 0001
serial: WSC2L45L
size: 2047GiB (2199GB)
capacity: 2TiB (2199GB)
capabilities: 15000rpm gpt-1.00 partitioned
partitioned:gpt
configuration: ansiversion=5
guid=da432e93-f76c-40a0-88cc-22bdf590df11 logicalsectorsize=512
sectorsize=512
I suspect it's a firmware issue. I've been through the setup
screens and didn't see anything that looked relevant. I can try
to find if Cisco could give me updated firmware for the server,
but I'm not confident.
I feel like there could be something basic I'm missing given the
other server can address 4TB. Maybe someone with IBM firmware
experience knows which direction I should look.
Thanks,
Alex
On 11/03/25 21:17, Alexander V.
Makartsev wrote:
On 11.03.2025 12:13, Alex King wrote:
Hi,
I've installed a large disk in a new machine and loaded Debian
(bookworm) on it, but it's showing as limited to 2TB when the
disk should be larger.
How can I get Debian to use the full 8TB on this disk?
Can you tell us more information about hardware setup?
How old is hardware of this "new machine"?
What's motherboard make and model?
Is HDD attached directly to the motherboard or using some kind of
eternal SATA\SAS controller?
It could be due to hardware\firmware limitations\incompatibility
of a SATA controller, or a fimware bug in drive itself.
Reply to: