General NAS-Central Forums

Welcome to the NAS community
It is currently Sat Sep 23, 2017 8:08 pm

All times are UTC




Post new topic Reply to topic  [ 73 posts ]  Go to page Previous  1, 2, 3, 4, 5  Next
Author Message
PostPosted: Fri Sep 27, 2013 12:59 pm 
Offline

Joined: Mon Jun 16, 2008 10:45 am
Posts: 6039
heavyarms wrote:
FYI:
The kernel log itself is more interesting.
Quote:
So do I have to put in any image for sdb7?
Yes. For instance you could use the arch-linux arm rootfs. It won't fit on your sda7, but you can delete sda8 and 9, and enlarge sda7. Put an ext3 filesystem on it, and extract (as root) the tgz file on it.


Top
 Profile  
 
PostPosted: Fri Sep 27, 2013 2:17 pm 
Offline

Joined: Wed Aug 21, 2013 2:20 pm
Posts: 38
Quote:
The kernel log itself is more interesting.


sorry but where can I get this?

I Tried to load arch-linux but failed. Getting the same, stops at verifying and OK

Code:
Loading from IDE device 1, partition 6: Name: hdb6
  Type: U-Boot
   Image Name:   Linux-3.10.10
   Created:      2013-09-27   7:34:05 UTC
   Image Type:   ARM Linux Kernel Image (uncompressed)
   Data Size:    2202432 Bytes =  2.1 MB
   Load Address: 00008000
   Entry Point:  00008000
## Booting image at 00400000 ...
   Image Name:   Linux-3.10.10
   Created:      2013-09-27   7:34:05 UTC
   Image Type:   ARM Linux Kernel Image (uncompressed)
   Data Size:    2202432 Bytes =  2.1 MB
   Load Address: 00008000
   Entry Point:  00008000
   Verifying Checksum ... OK
OK


Top
 Profile  
 
PostPosted: Fri Sep 27, 2013 3:58 pm 
Offline

Joined: Wed Aug 21, 2013 2:20 pm
Posts: 38
I did try to load the D2 Network UImage and images Kernel loads successfully, I'm now getting IP Address bu stops at this with LED blinking blue:

Code:
 unicorn daemon:
  missing or empty sysfs LED directory: /sys/class/leds
 
      ipcommd   [DAEMON_FAIL_START_RCODE]


Here's the full log, I've used both with 1 and 2 hdd:

https://docs.google.com/file/d/0B3RJUVbE6jxWbTdrZ2NnQ1dyWWM/edit?usp=sharing

And I can turn it off using the switch :D


Top
 Profile  
 
PostPosted: Fri Sep 27, 2013 5:28 pm 
Offline

Joined: Mon Jun 16, 2008 10:45 am
Posts: 6039
heavyarms wrote:
Quote:
The kernel log itself is more interesting.


sorry but where can I get this?
I guess the kernel did *not* boot ok. The log you posted was the bootloader, which reads the kernel from disk, and checks it. It's last message is 'Verifying checksum...OK'.
The next line you *should* see is 'Uncompressing the kernel .... Done, booting the kernel'. This is the decompressing piggyback inside the uImage, and after that the actual kernel log starts, which ends at 'Freeing up init memory'.
So apparently this ancient bootloader is not able to boot the latest kernel, or I made a mistake, somewhere.

heavyarms wrote:
I did try to load the D2 Network UImage and images Kernel loads successfully, I'm now getting IP Address bu stops at this with LED blinking blue:
Have you tried if the webinterface was running?
If not, it might help to apply the rest of the update, and not only the kernel. I'll try to extract it from the files I've got.


Top
 Profile  
 
PostPosted: Fri Sep 27, 2013 10:39 pm 
Offline

Joined: Wed Aug 21, 2013 2:20 pm
Posts: 38
Mijzelf wrote:
I guess the kernel did *not* boot ok. The log you posted was the bootloader, which reads the kernel from disk, and checks it. It's last message is 'Verifying checksum...OK'.
The next line you *should* see is 'Uncompressing the kernel .... Done, booting the kernel'. This is the decompressing piggyback inside the uImage, and after that the actual kernel log starts, which ends at 'Freeing up init memory'.
So apparently this ancient bootloader is not able to boot the latest kernel, or I made a mistake, somewhere.


Sorry, yeah I saw that it did not uncompressed the kernel. Anyways, hope the d2 kernel/image might work

Mijzelf wrote:
Have you tried if the webinterface was running?
If not, it might help to apply the rest of the update, and not only the kernel. I'll try to extract it from the files I've got.


I can go to http://ipaddress but it will not allow me, it just says that the system is still booting


Top
 Profile  
 
PostPosted: Sat Sep 28, 2013 8:38 am 
Offline

Joined: Wed Mar 18, 2009 11:13 pm
Posts: 773
reading the above I think your problem could be a result of the strange logic lacie uses for connecting dks to the bus and identifying them disks. You can see it already in the boot loader, the boot loader sees it as hdb. (and not hda). The kernel tries to load the root file system from sda. If the kernel is not setup well it might mix sda and sdb, or if the dev nodes are not set correctly in the root file system you will be unable to mount a disk. The kernel message about not able to mount sda7 points in this direction. To get it resolved the kernel needs to map the the sata ports properly and correct the device nodes in your root file system.
I recently made a kernel for a 5big1 based on linux 2.6.39.4. It suffered from similar problems.
The big network seems to have the Soc as big network but a different main board, I am not sure if it uses (like the 5big1) a pci interface to access the disk. That is is crucial to know in order how to make a setup file and compile kernel

The fact the this arm-arch kernel won't work is simple it doesn't support this board.


Top
 Profile  
 
PostPosted: Sat Sep 28, 2013 8:59 am 
Offline

Joined: Mon Jun 16, 2008 10:45 am
Posts: 6039
Maybe you (fvdw) should know an EBD is not 'symmetric', like other 2 disk Lacie devices. On a 2Big or D2 network both disks are equal, regarding partition table and so on. The EBD is not. It has one 'classical' disk, and a 2nd, unpartitioned one. The sda2 of the bootdisk is expanded with the 2nd disk in a linear raid volume.

As you could see in the thread, the bootloader only loads it's kernel from device 1, and not from device 0. So the bootscript is designed to enforce this behaviour. Further, in the EBD update kernel, you can see first sdb is found, and then sda. So I guess indeed the devicenames are swapped.

A last note, it's not clear to me which box this actually is. The EBD in the wiki has an EdminiV2 board with a 2nd sata port. This one is different. And the bootloader says it's a 'D2 Next'. So maybe it is a D2, but AFAIK that box is symmetric, so the bootloader should be able to boot from both disks.

Quote:
The fact the this arm-arch kernel won't work is simple it doesn't support this board.
Which kernel do you mean?

@heavyarms: Clearly the webserver is running. Else it couldn't tell you the system is still booting. On other Lacie's a factory reset on this point can help.


Top
 Profile  
 
PostPosted: Sat Sep 28, 2013 10:56 am 
Offline

Joined: Wed Mar 18, 2009 11:13 pm
Posts: 773
yeah my knowledge of this device is zero ;) but I learned something from other lacie devices.
Although this one always seems to enforce boot from device 1 as you mentioned you still need to care of the the disks definition and device nodes when using a custom kernel. Its a funny way Lacie designed these boards, opposite to what a logical thinking human would expect.

Quote:
Which kernel do you mean?
I mean this one he tried here viewtopic.php?f=151&p=69443#p69303


Top
 Profile  
 
PostPosted: Sat Sep 28, 2013 11:29 am 
Offline

Joined: Mon Jun 16, 2008 10:45 am
Posts: 6039
fvdw wrote:
Quote:
Which kernel do you mean?
I mean this one he tried here viewtopic.php?f=151&p=69443#p69303

That one has EBD support. (The EBD is supported by the vanilla). Well, it could be that this is no EBD, but even in that case *some* kernel output should be visible. There is nothing, that's why I asked you if you known anything about u-boot limitations. I have read that older u-boots cannot boot a recent Kirkwood kernel due to wrong cache initialisation, which makes the decompression piggyback fail. Silently. I have never read such a thing about an Orion, but that doesn't say it isn't there, of course.


Top
 Profile  
 
PostPosted: Sat Sep 28, 2013 12:26 pm 
Offline

Joined: Wed Mar 18, 2009 11:13 pm
Posts: 773
are you sure that it is supported ?
this is the content of mach-orion5x folder in 3.9.5, which one is the setup file for lacie ethernet big disk ?
Also remember that lacie used in the past for the orion boxes most of the time mac number 526 for all of them, so there can not be a kernel suitable for all, it must be adapted to at least correct the mach number.
or.. you are referring to device tree stuff in the newer kernels..which I do not understand (yet), but also those will be asked for a mach number..fitting with that in the boot loader
Code:
07-06-2013  21:54    <DIR>          .
07-06-2013  21:54    <DIR>          ..
07-06-2013  21:54             9.431 common.c
07-06-2013  21:54             3.201 rd88f6183ap-ge-setup.c
07-06-2013  21:54            14.589 pci.c
07-06-2013  21:54             4.903 rd88f5181l-fxo-setup.c
07-06-2013  21:54            18.391 dns323-setup.c
07-06-2013  21:54             4.999 rd88f5181l-ge-setup.c
07-06-2013  21:54               965 mpp.c
07-06-2013  21:54             4.263 wnr854t-setup.c
07-06-2013  21:54             4.444 addr-map.c
07-06-2013  21:54             7.086 mss2-setup.c
07-06-2013  21:54             9.519 d2net-setup.c
07-06-2013  21:54             4.972 edmini_v2-setup.c
07-06-2013  21:54             2.213 board-dt.c
07-06-2013  21:54             9.636 db88f5281-setup.c
07-06-2013  21:54             9.484 terastation_pro2-setup.c
07-06-2013  21:54                84 Makefile.boot
07-06-2013  21:54             5.932 mv2120-setup.c
07-06-2013  21:54             4.205 Kconfig
07-06-2013  21:54             3.269 tsx09-common.c
07-06-2013  21:54             2.590 common.h
07-06-2013  21:54            10.621 kurobox_pro-setup.c
07-06-2013  21:54             8.327 ls-chl-setup.c
07-06-2013  21:54            15.419 ts78xx-setup.c
07-06-2013  21:54             7.211 ls_hgl-setup.c
07-06-2013  21:54               874 irq.c
07-06-2013  21:54            11.678 net2big-setup.c
07-06-2013  21:54             6.139 wrt350n-v2-setup.c
07-06-2013  21:54    <DIR>          include
07-06-2013  21:54             5.819 mpp.h
07-06-2013  21:54             8.327 ts409-setup.c
07-06-2013  21:54             1.047 ts78xx-fpga.h
07-06-2013  21:54             8.764 ts209-setup.c
07-06-2013  21:54             7.280 lsmini-setup.c
07-06-2013  21:54               439 tsx09-common.h
07-06-2013  21:54             1.285 Makefile
07-06-2013  21:54             7.430 rd88f5182-setup.c
              35 bestand(en)          224.836 bytes
               3 map(pen)  1.836.444.385.280 bytes beschikbaar


u-boot limitations
it seem to use this version
Code:
U-Boot 1.1.4 (Feb  9 2009 - 13:15:53) Marvell version: 2.4.9  LaCie : 1.00.0001

I do not have the source code of that one to see what it doing when booting the box.
The first one I have is u-boot-1.1.4 marvel ??? lacie version 1.2.8 for the kirkwood family so no good reference
Means I cannot say what limitations this u-boot has, sorry.


Top
 Profile  
 
PostPosted: Sat Sep 28, 2013 1:30 pm 
Offline

Joined: Wed Aug 21, 2013 2:20 pm
Posts: 38
Mijzelf wrote:
@heavyarms: Clearly the webserver is running. Else it couldn't tell you the system is still booting. On other Lacie's a factory reset on this point can help.


I did factory reset the box but I still get this:

Code:
 unicorn daemon:
  missing or empty sysfs LED directory: /sys/class/leds
 
      ipcommd   [DAEMON_FAIL_START_RCODE]



@fvdw I was also trying to read about your kernel on another forum and it's interesting, though I don't know if its going to work with this board.


Top
 Profile  
 
PostPosted: Sat Sep 28, 2013 2:44 pm 
Offline

Joined: Wed Aug 21, 2013 2:20 pm
Posts: 38
On a second note..I was able to access the Web Admin Page Yey!

But another problem, it can't detect the drive :( although there is hdd activity when I click on the format drive

This is the last part of the output:

Code:
 unicorn daemon:
  modprobe: chdir(2.6.22.7): No such file or directory

      avahi                             [starting]
      avahi                             [started]       ( pid: 1091 )
      netatalk/cnid_metad               [starting]
      netatalk/cnid_metad               [started]       ( pid: 1101 )
      netatalk/afpd                     [starting]
      netatalk/afpd                     [started]       ( pid: 1106 )
      samba/smbd                        [starting]
      samba/initial                     [starting]
      samba/initial                     [done]          ( done in 172ms )
      samba/smbd                        [started]       ( pid: 1143 )
      samba/nmbd                        [starting]
      samba/nmbd                        [started]       ( pid: 1146 )
      samba/winbind                     [starting]
      samba/winbind                     [started]       ( pid: 1150 )
      cups                              [starting]
      cups                              [started]       ( pid: 1159 )

 unicorn daemon:
  missing or empty sysfs LED directory: /sys/class/leds

      ipcommd                           [DAEMON_FAIL_START_RCODE]
      ntpd                              [starting]
      ntpd                              [started]       ( pid: 1259 )
      cups                              [stopping]
      cups                              [stopped]       ( stopped in 37ms )



https://docs.google.com/file/d/0B3RJUVbE6jxWV0JkckE4MFpBeUk/edit?usp=sharing


Top
 Profile  
 
PostPosted: Sat Sep 28, 2013 2:45 pm 
Offline

Joined: Wed Aug 21, 2013 2:20 pm
Posts: 38
If this works then I will stick to the d2 then hehe :shock:


Top
 Profile  
 
PostPosted: Sat Sep 28, 2013 3:40 pm 
Offline

Joined: Mon Jun 16, 2008 10:45 am
Posts: 6039
fvdw wrote:
are you sure that it is supported ?
You can just select it in 'make menuconfig'. I did some analysis, and it appears that d2net-setup.c is the one, with CONFIG_MACH_BIGDISK defined. The D2 and EBD must be almost equal, the only difference is a string.

Quote:
Also remember that lacie used in the past for the orion boxes most of the time mac number 526 for all of them, so there can not be a kernel suitable for all, it must be adapted to at least correct the mach number.
Yes I know. I hoped on a kernel log, so I could see the current machtype, and the supported onces, so I could easily do a grep.

@heavyarms: How are your disks organized? The D2 expects 2 disks, with identical partitions. And also, what does the webinterface say which kind of box you have? The rootfs decides on base of the kernel command line which box it has. Your box says 'D2 Next', but I don't know what a D2 says.

At the moment I am uploading the additional files of the update. You can extract them on partition 8 (as root), overwriting some of the files there already are.

BTW, it might be a good idea to install ssh, while you have partition 8 mounted. Instructions here.


Top
 Profile  
 
PostPosted: Sat Sep 28, 2013 4:38 pm 
Offline

Joined: Wed Mar 18, 2009 11:13 pm
Posts: 773
the output heavyarms posted of u-boot mentions this
Code:
 ** MARVELL BOARD: DB-88F5182-LaCie LE

U-Boot 1.1.4 (Feb  9 2009 - 13:15:53) Marvell version: 2.4.9  LaCie : 1.00.0001

U-Boot code: 00200000 -> 0026FFF0  BSS: -> 00283280

Soc: 88F5182 A2 (DDR2)


There is a setup file in the vanilla kernel for this board with name db88f5182-setup.c
In the setup it mentions this DB88F5281, "Marvell Orion-2 Development Board" it has also a mach number.
The setup of this file is quite different from a D2 network, this booard uses a pci....
So does he really has the device you think.
I don't if there is somewhere a u-boot output of an EBD to compare


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 73 posts ]  Go to page Previous  1, 2, 3, 4, 5  Next

All times are UTC


Who is online

Users browsing this forum: No registered users and 6 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB® Forum Software © phpBB Group