Forum: >>> Magnum BBS <<<

Kernel only seeing 2TB of an 8TB disk

From Alex King@21:1/5 to All on Tue Mar 11 09:00:01 2025

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?

Thanks,
Alex

The disk is an 8TB model:

root@fj2:/home/installer# smartctl -i /dev/sdb
smartctl 7.3 2022-02-28 r5338 [x86_64-linux-6.1.0-29-amd64] (local build) Copyright (C) 2002-22, Bruce Allen, Christian Franke, www.smartmontools.org

=== START OF INFORMATION SECTION ===
Model Family:     Seagate BarraCuda 3.5 (SMR)
Device Model:     ST8000DM004-2U9188
Serial Number:    WSC2L45L
LU WWN Device Id: 5 000c50 0f79d1878
Firmware Version: 0001
User Capacity:    8,001,563,222,016 bytes [8.00 TB]
Sector Sizes:     512 bytes logical, 4096 bytes physical
Rotation Rate:    5400 rpm
Form Factor:      3.5 inches
Device is:        In smartctl database 7.3/5319
ATA Version is:   ACS-3 T13/2161-D revision 5
SATA Version is: SATA 3.1, 6.0 Gb/s (current: 3.0 Gb/s)
Local Time is:    Tue Mar 11 21:06:13 2025 UTC
SMART support is: Available - device has SMART capability.
SMART support is: Enabled

However it's showing in the kernel as 2TB:

root@fj2:/home/installer# cat /proc/partitions
major minor #blocks name

11        0    1048575 sr0
   8        0    7913472 sda
   8       16 2147483647 sdb
   8       17     524288 sdb1
   8       18     499712 sdb2
   8       19   18505728 sdb3
   9        0   18488320 md0
   7        0    3887968 loop0

root@fj2:/home/installer# lsblk
NAME    MAJ:MIN RM SIZE RO TYPE MOUNTPOINTS
loop0     7:0    0 3.7G 1 loop /debiancd
sda       8:0    0 7.5G 0 disk
sdb       8:16   0    2T 0 disk
|-sdb1    8:17   0 512M 0 part /boot/efi
|-sdb2    8:18   0 488M 0 part /boot
`-sdb3    8:19   0 17.6G 0 part
`-md0   9:0    0 17.6G 0 raid1 /
sr0      11:0    1 1024M 0 rom

root@fj2:/home/installer# dmesg|grep sdb
[   12.977869] sd 7:0:0:0: [sdb] 4294967294 512-byte logical blocks:
(2.20 TB/2.00 TiB)
...

root@fj2:/home/installer# gdisk -l /dev/sdb
GPT fdisk (gdisk) version 1.0.9

Partition table scan:
MBR: protective
BSD: not present
APM: not present
GPT: present

Found valid GPT with protective MBR; using GPT.
Disk /dev/sdb: 4294967294 sectors, 2.0 TiB
Model: ST8000DM004-2U91
Sector size (logical/physical): 512/512 bytes
Disk identifier (GUID): DA432E93-F76C-40A0-88CC-22BDF590DF11
Partition table holds up to 128 entries
Main partition table begins at sector 2 and ends at sector 33
First usable sector is 34, last usable sector is 4294967260
Partitions will be aligned on 2048-sector boundaries
Total free space is 4255907771 sectors (2.0 TiB)

Number Start (sector)    End (sector) Size       Code Name
   1            2048         1050623   512.0 MiB   EF00 1
   2         1050624         2050047   488.0 MiB   8300 2
   3         2050048        39061503   17.6 GiB    FD00 3

root@fj2:/home/installer# lsb_release -a
No LSB modules are available.
Distributor ID: Debian
Description:    Debian GNU/Linux 12 (bookworm)
Release:        12
Codename:       bookworm

root@fj2:/home/installer# cat /proc/version
Linux version 6.1.0-29-amd64 ([email protected]) (gcc-12
(Debian 12.2.0-14) 12.2.0, GNU ld (GNU Binutils for Debian) 2.40) #1 SMP PREEMPT_DYNAMIC Debian 6.1.123-1 (2025-01-02)

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

From Yassine Chaouche@21:1/5 to All on Tue Mar 11 09:50:02 2025

This is a multi-part message in MIME format.
Le 3/11/25 à 09:17, Alexander V. Makartsev a écrit :

Can you tell us more information about hardware setup?

Could RAID setup play a role?

Best,

--
yassine -- sysadm
http://about.me/ychaouche
Looking for side gigs.

<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
</head>
<body>
<div class="moz-cite-prefix">Le 3/11/25 à 09:17, Alexander V.
Makartsev a écrit : 
</div>
<blockquote type="cite"
cite="mid:[email protected]">Can you
tell us more information about hardware setup? 
</blockquote>
Could RAID setup play a role? 
 
Best, 
<pre class="moz-signature" cols="0">--
yassine -- sysadm
<a class="moz-txt-link-freetext" href="http://about.me/ychaouche">http://about.me/ychaouche</a>
Looking for side gigs.</pre>
</body>
</html>

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

From Thomas Schmitt@21:1/5 to Alex King on Tue Mar 11 11:00:01 2025

Hi,

Alex King wrote:

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.

root@fj2:/home/installer# smartctl -i /dev/sdb
[...]
User Capacity:    8,001,563,222,016 bytes [8.00 TB]
Sector Sizes:     512 bytes logical, 4096 bytes physical

root@fj2:/home/installer# cat /proc/partitions
major minor #blocks name
[...]
   8       16 2147483647 sdb

Observation:

2147483647 = 2 exp 31 - 1

This could mean that the kernel bonks at a 31-bit limit.

root@fj2:/home/installer# cat /proc/version
Linux version 6.1.0-29-amd64 [...]

But where would an amd64 kernel have such an ancient limitation ?

root@fj2:/home/installer# gdisk -l /dev/sdb
[...]
Disk /dev/sdb: 4294967294 sectors, 2.0 TiB
[...]
Sector size (logical/physical): 512/512 bytes

This agrees to /proc/partitions. man proc_partitions says that it
counts blocks of 1024 bytes.
4294967294 / 2 = 2147483647

Could there be confusion about the physical block size ?
Probably not:
The byte count resulting from block count multiplied by the physical
block size 4096 as reported by smartctl does not match the reported
"User Capacity" of smartctl:
2147483647 * 4096 = 8,796,093,018,112
That's higher than the smartctl value by nearly 10 percent, which makes
it rather improbable as miscalculation suspect.

So there remains only a simple 31-bit ceiling as suspect.
It's not a 32-bit rollover where the result remainder of this division
would have to be near to 0 , 1/2 or 1:
8001563222016 / 512 / 4294967296 = 7.277379356324673

Have a nice day :)

Thomas

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

From [email protected]@21:1/5 to Alex King on Tue Mar 11 11:30:01 2025

On Tue, Mar 11, 2025 at 08:13:16PM +1300, Alex King wrote:

Hi,

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?

It seems that the combination of MBR partition table and 512 byte blocks
limits you to partition sizes (and offsets) of roughly 2T, so it might
be this what's biting you:

https://superuser.com/questions/1393198/what-is-the-maximum-size-of-hard-drive-used-mbr-partitioning

Cheers
--
t

-----BEGIN PGP SIGNATURE-----

iF0EABECAB0WIQRp53liolZD6iXhAoIFyCz1etHaRgUCZ9APtgAKCRAFyCz1etHa RntlAJ45RLx3IWAWJjk3FJJANMw/xYxy4ACfflwTtRWzyE6lduphKCzyhbXBudY=
=Cr+u
-----END PGP SIGNATURE-----

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

From [email protected]@21:1/5 to Nicolas George on Tue Mar 11 11:50:01 2025

On Tue, Mar 11, 2025 at 11:40:29AM +0100, Nicolas George wrote:

[email protected] (HE12025-03-11):

It seems that the combination of MBR partition table and 512 byte blocks limits you to partition sizes (and offsets) of roughly 2T, so it might
be this what's biting you:

Highly doubtful considering these informations:

8 16 2147483647 sdb

Found valid GPT with protective MBR; using GPT.
Disk /dev/sdb: 4294967294 sectors, 2.0 TiB

Right.

Cheers
--
t

-----BEGIN PGP SIGNATURE-----

iF0EABECAB0WIQRp53liolZD6iXhAoIFyCz1etHaRgUCZ9AUTwAKCRAFyCz1etHa Rm1RAJ9X7HW47mA+PsT+Jj3lIKK+WXGbkQCcD18XEnFIGdwrm3JwyWJKlnY8Wig=
=8Sqq
-----END PGP SIGNATURE-----

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

From Thomas Schmitt@21:1/5 to All on Tue Mar 11 11:20:01 2025

Hi,

correction of the usual copy+paste error:

I wrote:

8001563222016 / 512 / 4294967296 = 7.277379356324673

The result stems from a different calculation with 2 exp 31.
With 2 exp 32 it is 3.638689678162337 .

Have a nice day :)

Thomas

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

From Nicolas George@21:1/5 to All on Tue Mar 11 11:50:01 2025

[email protected] (HE12025-03-11):

It seems that the combination of MBR partition table and 512 byte blocks limits you to partition sizes (and offsets) of roughly 2T, so it might
be this what's biting you:

Highly doubtful considering these informations:

� 8�� 16 2147483647 sdb

Found valid GPT with protective MBR; using GPT.
Disk /dev/sdb: 4294967294 sectors, 2.0 TiB

Regards,

--
Nicolas George

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

From Nicolas George@21:1/5 to All on Tue Mar 11 12:20:01 2025

Timothy M Butterworth (HE12025-03-11):

I have a 16TB drive that is working properly. I reformatted the drive to
the ext4 file system with default settings and it works great. Try to

Good for you.

reformat the drive especially if the current format is FAT.

We have both the kernel and gdisk telling us the block device has the
wrong size: the issue is absolutely not the filesystem on it.

Also, this is not Windows, please do not use “format” as a verb for a
disk.

Regards,

--
Nicolas George

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

From Thomas Schmitt@21:1/5 to Timothy M Butterworth on Tue Mar 11 12:30:01 2025

Hi,

Timothy M Butterworth wrote:

I have a 16TB drive that is working properly. I reformatted the drive to the
ext4 file system with default settings and it works great. Try to reformat the drive especially if the current format is FAT.

Several of the shown inquiry methods did not refer to filesystems and
also not to partitions alone:
cat /proc/partitions
lsblk
gdisk -l /dev/sdb
plus the line grepped from dmesg:
[ 12.977869] sd 7:0:0:0: [sdb] 4294967294 512-byte logical blocks: (2.20 TB/2.00 TiB)

They all consistently say that the disk has 2 exp 32 - 2 blocks
of 512 bytes = 2 TiB - 1024 bytes.
Only smartctl, which does not depend on the kernel's size assessment,
reports a capacity of 8 TB + 1.563222016 GB.

So 2 TiB is obviously a misperception of the overall disk size by the
kernel. Linux hands this size to user space by ioctl(BLKGETSIZE) or ioctl(BLKGETSIZE64).

gdisk uses them:
https://codesearch.debian.net/search?q=package%3Agdisk+BLKGETSIZE
(on amd64, "long sz" and "long long b" have the same size of 8 bytes.)

util-linux, the mother of lsblk uses them:
https://codesearch.debian.net/search?literal=0&q=package%3Autil-linux+ioctl.*BLKGETSIZE&page=3
(One has to go to the end of the last page to find the code which
does the work. The preceding matches are i18n noise.)

Have a nice day :)

Thomas

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

From Pierre Tomon@21:1/5 to All on Tue Mar 11 13:50:01 2025

Le Tue, 11 Mar 2025 20:13:16 +1300,
Alex King <[email protected]> a écrit :

Partition table scan:
MBR: protective
BSD: not present
APM: not present
GPT: present

Found valid GPT with protective MBR; using GPT.

Try to remove protective MBR.

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

From gene heskett@21:1/5 to [email protected] on Tue Mar 11 13:50:01 2025

On 3/11/25 06:26, [email protected] wrote:

On Tue, Mar 11, 2025 at 08:13:16PM +1300, Alex King wrote:

Hi,

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?

It seems that the combination of MBR partition table and 512 byte blocks limits you to partition sizes (and offsets) of roughly 2T, so it might
be this what's biting you:

https://superuser.com/questions/1393198/what-is-the-maximum-size-of-hard-drive-used-mbr-partitioning

Cheers

A problem that does not exist in gpt partition tables. Unforch, changing
it now will require a 100% backup/restore OR a reinstall to fix. fdisk
can do this table change by entering a g at the main screen, creating an
empty gpt table, and then an n will allow the whole disk as one
partition if the defaults are accepted. Then w the new table, and format
to the fs of choice. Then install bookworm fresh. I'd save your /home
to another drive big enough to hold it, remove that drive for safe
keeping, do the changes, remount that drive and copy your work back to
the new installs /home as needed.

Cheers, Gene Heskett, CET.
--
"There are four boxes to be used in defense of liberty:
soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author, 1940)
If we desire respect for the law, we must first make the law respectable.
- Louis D. Brandeis

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

From Christopher David Howie@21:1/5 to gene heskett on Tue Mar 11 19:20:02 2025

On 3/11/25 8:48 AM, gene heskett wrote:

A problem that does not exist in gpt partition tables. Unforch, changing
it now will require a 100% backup/restore OR a reinstall to fix. fdisk
can do this table change by entering a g at the main screen, creating an empty gpt table, and then an n will allow the whole disk as one
partition if the defaults are accepted. Then w the new table, and format
to the fs of choice. Then install bookworm fresh. I'd save your /home
to another drive big enough to hold it, remove that drive for safe
keeping, do the changes, remount that drive and copy your work back to
the new installs /home as needed.

The comment you are replying to is incorrect. The partition table shown
to us by OP is GPT, not MBR.

Found valid GPT with protective MBR; using GPT.

The disk's corresponding block device itself reports a size of 2TB as
shown in the lsblk output.

root@fj2:/home/installer# lsblk
NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINTS
sdb 8:16 0 2T 0 disk

The problem is below the partition table; the kernel does not know the
actual size of the disk.

This could be for any number of reasons, including controller and drive firmware bugs. We need more info from OP, which Alexander requested and
OP has not yet replied to.

--
Chris Howie
http://www.chrishowie.com
http://en.wikipedia.org/wiki/User:Crazycomputers

If you correspond with me on a regular basis, please read this document: http://www.chrishowie.com/email-preferences/

PGP fingerprint: 2B7A B280 8B12 21CC 260A DF65 6FCE 505A CF83 38F5

------------------------------------------------------------------------
IMPORTANT INFORMATION/DISCLAIMER

This document should be read only by those persons to whom it is
addressed. If you have received this message it was obviously addressed
to you and therefore you can read it.

Additionally, by sending an email to ANY of my addresses or to ANY
mailing lists to which I am subscribed, whether intentionally or
accidentally, you are agreeing that I am "the intended recipient," and
that I may do whatever I wish with the contents of any message received
from you, unless a pre-existing agreement prohibits me from so doing.

This overrides any disclaimer or statement of confidentiality that may
be included on your message.

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

From Alex King@21:1/5 to Alexander V. Makartsev on Wed Mar 12 00:20:01 2025

This is a multi-part message in MIME format.
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.

--
With kindest regards, Alexander.
Debian - The universal operating system
https://www.debian.org

<!DOCTYPE html>
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
</head>
<body>
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: <a class="moz-txt-link-abbreviated" href="mailto:pci@0000:01:00.0">pci@0000:01:00.0</a> 
 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: <a class="moz-txt-link-abbreviated" href="mailto:scsi@7:0.0.0">scsi@7:0.0.0</a> 
 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 

 

<div class="moz-cite-prefix">On 11/03/25 21:17, Alexander V.
Makartsev wrote: 
</div>
<blockquote type="cite"
cite="mid:[email protected]">
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
<div class="moz-cite-prefix">On 11.03.2025 12:13, Alex King wrote: 
</div>
<blockquote type="cite"
cite="mid:[email protected]">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? 
 
</blockquote>
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. 
 
<div class="moz-signature">-- 
<pre style="font-size: .8rem;"> With kindest regards, Alexander.</pre>
<img src="mailbox:///home/alex/temp/temp/ForwardedMessage.eml?number=0&header=quotebody&part=1.2.2&filename=DiO5GXyu0o2yFuSu.gif"
alt="" style="float:left; display:inline;"
moz-do-not-send="true">
<pre style="display:inline; font-size: .6rem;"> Debian - The universal operating system
<a href="https://www.debian.org" class="moz-txt-link-freetext"
moz-do-not-send="true">https://www.debian.org</a></pre>
</div>
</blockquote>
</body>
</html>

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

From Titus Newswanger@21:1/5 to Alex King on Wed Mar 12 02:00:01 2025

On 3/11/25 18:12, Alex King wrote:

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.)

01:00.0 SCSI storage controller: Broadcom / LSI SAS1064ET PCI-Express Fusion-MPT SAS (rev 08)

Quite possibly reflashing will fix it. IBM and Broadcom appear to have
firmware available when I google 'LSI SAS1064ET flashing'

I noticed an old forum referring to this raid card at https://forums.unraid.net/topic/24497-sas1064et-and-3tb-hard-drives/
however, that doesn't appear to have a solution.

On a Dell R620 I needed to flash the Broadcom LSI SAS (PERC H710) so the
OS could access the HDDs directly, I used:
https://fohdeesha.com/docs/perc.html It worked very well, but I don't
see anything about support for the IBM there.

--
Titus Newswanger
Curtiss WI

--- SoupGate-Win32 v1.05
* Origin: fsxNet Usenet Gateway (21:1/5)

Who's Online
Recent Visitors
- Krenn
  Sun Jun 7 03:07:26 2026
  from Sydney, Nsw via Telnet
- Krenn
  Sun Jun 7 01:30:12 2026
  from Sydney, Nsw via Telnet
- Centurion
  Sat Jun 6 23:27:30 2026
  from Berea, Ohio via Telnet
- Ab Cadd
  Sat Jun 6 15:42:53 2026
  from Sheboygan, Wi via Telnet
- Centurion
  Sat Jun 6 15:32:28 2026
  from Berea, Ohio via Telnet
- Krenn
  Sat Jun 6 11:38:56 2026
  from Sydney, Nsw via Telnet
- Furryboy
  Sat Jun 6 10:56:29 2026
  from Romania, Galati via SSH
- Centurion
  Fri Jun 5 22:28:01 2026
  from Berea, Ohio via Telnet

System Info

Sysop:	Keyop
Location:	Huddersfield, West Yorkshire, UK
Users:	715
Nodes:	16 (0 / 16)
Uptime:	161:23:54
Calls:	12,094
Calls today:	2
Files:	15,000
Messages:	6,517,775

Kernel only seeing 2TB of an 8TB disk

Who's Online

Recent Visitors

System Info