Print

[Robert]

Hi, I own a Drobo NAT system, that is formatted with HFS+ file-system because I mostly use Mac systems. I want to attach the Drobo to the Plug and use it as a file-server via Samba. So, the first thing I have done is attach the Drobo to the USB port. But not working. Here is the message log:

usb 1-1: new high speed USB device using orion-ehci and address 3
usb 1-1: configuration #1 chosen from 1 choice
scsi1 : SCSI emulation for USB Mass Storage devices
usb-storage: device found at 3
usb-storage: waiting for device to settle before scanning
scsi 1:0:0:0: Direct-Access TRUSTED Mass Storage 2.00 PQ: 0 ANSI: 5
sd 1:0:0:0: Attached scsi generic sg0 type 0
usb-storage: device scan complete
sd 1:0:0:0: [sda] Too big for this kernel. Use a kernel compiled with support for large block devices.
sd 1:0:0:0: [sda] Unsupported sector size -75.
sd 1:0:0:0: [sda] 0 512-byte hardware sectors: (0 B/0 B)
sd 1:0:0:0: [sda] Write Protect is off
sd 1:0:0:0: [sda] Mode Sense: 03 00 00 00
sd 1:0:0:0: [sda] Assuming drive cache: write through
sd 1:0:0:0: [sda] Attached SCSI disk

This sounds like I need to compile a different kernel. Will this solve the problem? And if, how to do it? What do I need to configure? I have never compile a kernel.

[NewIT_Marcus]

You didn't say what kernel you have now, nor what kind of system you have (eSATA, non-eSATA, pre-configured ... ).

Suffice to say, the kernels at http://sheeva.with-linux.com/sheeva/ are easy to install and might have the feature that you require.

[Robert]

I think you mean with "system" what kind of Plug I use. Well it's the standard plug either with Ubuntu (internal) or Debian (SD card) pre-configured.

The NAT is attached via USB, as I wrote.

[NewIT_Marcus]

I think you mean with "system" what kind of Plug I use. Well it's the standard plug either with Ubuntu (internal) or Debian (SD card) pre-configured.

The NAT is attached via USB, as I wrote.

OK, so you have 2.6.30.2 kernel. The procedure from updating the kernel from http://sheeva.with-linux.com/sheeva/ is simple. Download and chmod a+x the ReadMe. Then execute it with --nandkernel. The SD card has to be done manually. Here's the code I use for 2.6.32.7:

Code: [Select]

mkdir /mnt/mmc-kernel-partition/
mkdir /mnt/mmc-rootfs-partition/
mount /dev/mmcblk0p1 /mnt/mmc-kernel-partition/
mount /dev/mmcblk0p2 /mnt/mmc-rootfs-partition/
cp /mnt/USB-stick/2.6.32.7/sheeva-2.6.32.7-uImage /mnt/mmc-kernel-partition/uImage
tar x -C /mnt/mmc-rootfs-partition/ --overwrite -zf /mnt/USB-configuration-stick/2.6.32.7/sheeva-2.6.32.7-Modules.tar.gz


[Robert]

Thanks will try.

What options do I have to change in the Kernel to support large block devices?

[NewIT_Marcus]

Thanks will try.

What options do I have to change in the Kernel to support large block devices?

I googled a bit and had about as much luck as you, I'm afraid.

[Robert]

Regarding the Sheeva Kernel update. the README states:

# This is a mainline Linux Kernel and you must set
# the mainlineLinux and arcNumber env variables in U-Boot
# and change the bootargs for a successful boot.
#
# setenv mainlineLinux yes
# setenv arcNumber 2097
####### change bootargs, replace nand_mtd with orion_nand and add rootfstype=jffs2
# setenv bootargs rootfstype=jffs2 console=ttyS0,115200 mtdparts=orion_nand:[email protected](uImage),[email protected](rootfs) rw root=/dev/mtdblock1 rw ip=192.168.1.9:192.168.1.4:192.168.1.4:255.255.255.0:DB88FXX81:eth0:none
# saveenv


Do I have to do this manually or why is it commented in the script? Or is this setup already correct for the Plug?

[Robert]

And, I can replace the Ubuntu Kernel with the Sheeva one? Not sure if they are really the same of if Ubuntu did some patches for the ARM platform.

[NewIT_Marcus]

You change the environment variables in U-Boot, ie you interrupt the boot process and type (or copy & paste) the commands at the Marvell>> prompt in U-Boot.

ie they are executed in a different context from the downloaded script.

These are the right kernels to be using.

[Robert]

Ok. Never touched u-boot.

I didn't set the variables, the update went through and the plug boots without any problems.

Now back to see the block device problem.

Is there any instruction how to compile and build an own kernel for the plug?

[Robert]

Ok, the new kernel looks promising:

usb 1-1: new high speed USB device using orion-ehci and address 2
scsi2 : usb-storage 1-1:1.0
scsi 2:0:0:0: Direct-Access     TRUSTED  Mass Storage     2.00 PQ: 0 ANSI: 5
sd 2:0:0:0: Attached scsi generic sg0 type 0
sd 2:0:0:0: [sda] Very big device. Trying to use READ CAPACITY(16).
sd 2:0:0:0: [sda] 34359738368 512-byte logical blocks: (17.5 TB/16.0 TiB)
sd 2:0:0:0: [sda] Write Protect is off
sd 2:0:0:0: [sda] Mode Sense: 03 00 00 00
sd 2:0:0:0: [sda] Assuming drive cache: write through
sd 2:0:0:0: [sda] Very big device. Trying to use READ CAPACITY(16).
sd 2:0:0:0: [sda] Assuming drive cache: write through
 sda:

Now trying to mount and see if I can get the HFS+ to work.

[Robert]

Ok, HFS+ makes problems:

sd 2:0:0:0: [sda] Very big device. Trying to use READ CAPACITY(16).
sd 2:0:0:0: [sda] Assuming drive cache: write through
sd 2:0:0:0: [sda] Attached SCSI disk
hfs: volumes larger than 2TB are not supported yet
hfs: unable to find HFS+ superblock


So, going to get rid of HFS+ and use standard Unix EXT3 filesystem. IMO keeps trouble lowest.

[NewIT_Marcus]

Ok. Never touched u-boot.

I didn't set the variables, the update went through and the plug boots without any problems.

Now back to see the block device problem.

Is there any instruction how to compile and build an own kernel for the plug?

Yes. These instructions are very helpful.

[nickk]

Hi Robert,

I was looking to enable large block device support as well so I thought while I was going through the process, it would be a good idea to look for HFS support as well since someone else was looking for it.

So, it looks like you need:

Symbol: HFSPLUS_FS [=n]                                                                     ¦
  ¦ Prompt: Apple Extended HFS file system support                                              ¦
  ¦   Defined at fs/hfsplus/Kconfig:1                                                           ¦
  ¦   Depends on: MISC_FILESYSTEMS [=y] && BLOCK [=y]                                           ¦
  ¦   Location:                                                                                 ¦
  ¦     -> File systems                                                                         ¦
  ¦       -> Miscellaneous filesystems (MISC_FILESYSTEMS [=y])                                  ¦
  ¦   Selects: NLS [=y] && NLS_UTF8 [=y]
 
  Symbol: HFS_FS [=n]                                                                         ¦
  ¦ Prompt: Apple Macintosh file system support (EXPERIMENTAL)                                  ¦
  ¦   Defined at fs/hfs/Kconfig:1                                                               ¦
  ¦   Depends on: MISC_FILESYSTEMS [=y] && BLOCK [=y] && EXPERIMENTAL [=y]                      ¦
  ¦   Location:                                                                                 ¦
  ¦     -> File systems                                                                         ¦
  ¦       -> Miscellaneous filesystems (MISC_FILESYSTEMS [=y])                                  ¦
  ¦   Selects: NLS [=y]     

You can find those in:
File systems->Miscellaneous Filesystems->Apple Extended HFS file system support
and
File systems->Miscellaneous Filesystems->Apple Macintosh file system support (EXPERIMENTAL) 

For anyone else who is looking for LBD (large block device support), it can be found in:

Symbol: LBDAF [=y]                                                                          ?
  ? Prompt: Support for large (2TB+) block devices and files                                    ?
  ?   Defined at block/Kconfig:26                                                               ?
  ?   Depends on: BLOCK [=y] && !64BIT [=64BIT]                                                 ?
  ?   Location:                                                                                 ?
  ?     -> Enable the block layer

So this is in:
Enable the block layer->Support for large (2TB+) block devices and files

[Robert]

Hi, so there are changes in the source code which than requires a new kernel compilation?

Have you tried it out with a 6TB device for example? I'm always a bit cautious when tweaking filesystem code. At least it should be very stable :-)

I have switched over to EXT3 filesystem now, which works pretty good. Only drawback is that I have a 2TB size limit on my drobo with EXT3. But that's OK, I can live with it.