nwsp2 lite - stock Seagate AF HD 2TB not booting correctly

[fvdw]

what we are going to do is talking directly to u-boot

For this we need a program called "clunc". I compiled it in ubuntu 11.0 for x86 PC's (all intels and amd's) so I guess it should also run on your laptop on your ubuntu version

file moved to download forum

Assuming your laptop with ubuntu and nwsp2 is connected to the LAN
nwsp2 switched off

Attached two files put them on your laptop in ubuntu in a folder for example with name "clunc" and be sure that they are set executable.
Open a terminal window in ubuntu and go to this directory where you put the files
now run the command

Code: Select all

./clunc -vi ip-nas

replace ip-nas by the actual ip address of your nwsp2, or in case you don't know this use an ip address in the dhcp range of your dhcp server (typcally 192.168.x.y (the number x must match with the number on that place of the ip addres of your laptop) then choosen ip number must not be in use by another pc.

If the program runs ok you should get a response like this

Code: Select all

root@fvdwsl-nwsp2:/clunc # ./clunc -vi 192.168.1.147
waiting U-Boot...

now start the nwsp-2
It will power up and you should get a short flash of the red led meaning it detected that clunc is running and wants to connect to the netconsole
on you laptop you should now see in the terminal window this

Code: Select all

U-Boot NetConsole
-----------------

interactive shell

Marvell>>

Now we can talk to u-boot :mrgreen: and try to figure out what is wrong
Be very very careful with commands you give here it is possible to give commands that makes you u-boot corrupt or you can even delete it and then your nwsp2 will be a nice black brick.

Now lets check the harddisk
give the command "ide reset"
On my system it gives this reponse

Code: Select all

Marvell>> ide reset
ide reset

Reset IDE:
Marvell Serial ATA Adapter
Integrated Sata device found
[0 0 0]: Enable DMA mode
  Device 0 @ 0 0:
Model: ST2000DL001-9VT156                       Firm: CC41     Ser#:             5YD2JCJX
            Type: Hard Disk
            Supports 48-bit addressing
            Capacity: 1907729.0 MB = 1863.0 GB (3907029168 x 512)

Marvell>>

Now give command "ide part"
On my system this gives

Code: Select all

Marvell>> ide part
ide part
## Testing for valid DOS partition ##

Partition Map for IDE device 0  --   Partition Type: DOS

Partition     Start Sector     Num Sectors     Type
    1               128520         1028160      83
    2              1156680         1028160      83
    3              2184840         1028160      83
    4              3213000      -391172296       5 Extd
    5              3341520         1670760      83
    6              5140800          128520      83
    7              5397840         1028160      83
    8              6554520      -394513816      83
Marvell>>

The number Num sectors for partition 4 and 8 are wrong is an error in u-boot code of lacie on my system.
Important here is that u-boot is able to find partition 6 and 7.
if this is the case then it is able to load the kernel from sda6 and load root file system form sda7 as necesarry to start the nwsp-2

If this is case give command "printenv bootcmd", this should give this output

Code: Select all

Marvell>> printenv bootcmd
printenv bootcmd
bootcmd=run disk_disk
Marvell>>

Now try to boot it by using the command "run bootcmd", this should give this output

Code: Select all

Marvell>> run bootcmd
run bootcmd

** Device 1 not available
## Valid DOS partition found ##

Loading from IDE device 0, partition 6: Name: hda6
  Type: U-Boot
   Image Name:   Linux-2.6.39.4
   Created:      2012-12-23  21:59:32 UTC
   Image Type:   ARM Linux Kernel Image (uncompressed)
   Data Size:    3123456 Bytes =  3 MB
   Load Address: 00008000
   Entry Point:  00008000
## Booting image at 00800000 ...
   Image Name:   Linux-2.6.39.4
   Created:      2012-12-23  21:59:32 UTC
   Image Type:   ARM Linux Kernel Image (uncompressed)
   Data Size:    3123456 Bytes =  3 MB
   Load Address: 00008000
   Entry Point:  00008000
   Verifying Checksum ... OK
OK

Starting kernel ...




if you get this then it means the kernel is found , loaded and started
If the nwsp2 then doesn't boot the problem must be in sda7

But for the moment enough to start with

[frenbu]

ok i've done all the steps, the output is exactly the same of yours, so the kernel loading is ok...
http://pastebin.com/hXNqrr1C

i'll check later on sda7, i'll try to write again the firmware on it and i'll update you!

thank you fvdw!

[frenbu]

I've tried to write again the .img file on partition 7, checked if it's correctly mountable and despite all seems ok the boot sequence is still hanging at the same point... sdb7 doesn't even contain the boot.log file.
What can i do for further debug?? Is there a way to get a serial console to monitor the boot process?

[fvdw]

mmm... :scratch
lets try if we can load an image from a disk partition 7
In clunc at the Marvell prompt
give the command

Code: Select all

Marvell>>ide reset
.
.
Marvell>> ext2load disk 0:7 0x800000 /boot/UIMAGE-26394-NWSP2CL-9

(this assumes that you have installed fvdw-sl-11.0, if not then kernel name must be adapted)
If it loads give command "bootm" to boot it
If it doesn't load then something is still wrong with format of sda7

[frenbu]

Ok i've done it, ext2load gives me an error:

Code: Select all

Marvell>> ide reset
ide reset

Reset IDE:
Marvell Serial ATA Adapter
Integrated Sata device found
[0 0 0]: Enable DMA mode
  Device 0 @ 0 0:
Model: ST2000DM001-9YN164                       Firm: CC4H     Ser#:             S240JEVJ
            Type: Hard Disk
dev_desc->blksz = 512
            Supports 48-bit addressing
            Capacity: 1907729.0 MB = 1863.0 GB (3907029168 x 512)

Marvell>> ext2load disk 0:7 0x800000 /boot/UIMAGE-26394-NWSP2CL-9
ext2load disk 0:7 0x800000 /boot/UIMAGE-26394-NWSP2CL-9

** Block device disk 0 not supported
Marvell>>


:doh
however partitions and device are recognized:

Code: Select all

Marvell>> ide part
ide part
## Testing for valid DOS partition ##

Partition Map for IDE device 0  --   Partition Type: DOS

Partition     Start Sector     Num Sectors     Type
    1               128520         1028160      83
    2              1156680         1028160      83
    3              2184840         1028160      82
    4              3213000      -391172296       5 Extd
    5              3341520         1670760      83
    6              5140800          128520      83
    7              5397840         1028160      83
    8              6554520      -394513816      83


What next? :)

[fvdw]

oops my error

the command should be

Code: Select all

 ext2load ide 0:7 0x800000 /boot/UIMAGE-26394-NWSP2CL-9

:whistle

if partition 7 won't load then you can try 1 and 2 (those also contain the firmware)

[frenbu]

Now the command is ok but after bootm the kernel loads and nothing happens (same as run bootcmd), tried this with partition 1,2 and 7 but after bootm there's some disk activity and then nothing.

Code: Select all

Marvell>> ext2load ide 0:7 0x800000 /boot/UIMAGE-26394-NWSP2CL-9
ext2load ide 0:7 0x800000 /boot/UIMAGE-26394-NWSP2CL-9
## Valid DOS partition found ##
## Valid DOS partition found ##

2996012 bytes read
Marvell>> bootm
bootm
## Booting image at 00800000 ...
   Image Name:   Linux-2.6.39.4
   Created:      2011-11-11  21:39:10 UTC
   Image Type:   ARM Linux Kernel Image (uncompressed)
   Data Size:    2995948 Bytes =  2.9 MB
   Load Address: 00008000
   Entry Point:  00008000
   Verifying Checksum ... OK
OK
bootargs from environment variables : console=ttyS0,115200 root=/dev/sda7 ro reset=0 productType=ASTON_KW cap=gpt,lba64

Starting kernel ...


What can be wrong with partition 7? :hammerhead I've already checked and wrote it on disk several times and it's mountable!
I also tried to change the root in bootargs environment variable in combination with loading the kernel from the first partition:

Code: Select all

setenv bootargs console=ttyS0,115200 root=/dev/sda1 ro reset=0 productType=ASTON_KW

but the result is the same.....

[frenbu]

However, it doesn't boot even with the old Wested Digital HD that ran for several weeks before my tests with the Seagate HD...

I'm feeling so fooled by my nwsp2 :laughat

[fvdw]

:hairpull

there must be something preventing it from booting. After the kernel is loaded it performs /sbin/init in the root file system and that reads /etc/inittab in which it reads it has to execute /etc/init.d/rcS
sda7 is readable as proven by loading the kernel from it.
Maybe we should check the md5 sum of the image you use to check if it is not corrupt.
It should be this

Code: Select all

md5sum fvdw-sl-11-0-nwsp2-6jan2012.img
e1cb6d22509b97b4166e5a5774534bb8  fvdw-sl-11-0-nwsp2-6jan2012.img


(the md5 sum of the kernel is supplied with the first time install package)

[fvdw]

to see if the file the kernel need is there you could run

Code: Select all

Marvell>> ide reset
...
Marvell>> ext2ls ide 0:7 /etc/init.d

it should list this

Code: Select all

ext2ls ide 0:7 /etc/init.d

<DIR>       4096 .
<DIR>       4096 ..
            3779 rc.banner.copy
           10378 rcS
            5402 rc.bootup1
           10378 rcS.copy
             284 insert-modules
            5402 rc.bootup1.copy
            3779 rc.banner
            3013 rc.bootup2.copy
            3013 rc.bootup2
Marvell>>