bad crc and ro serial console unbrick?

I tried to restore uboot to facotry and ended up with bad checksum. I messed up the pad by the rx pin so i cant write from serial console. here is what happens on boot:

U-Boot 1.1.4  Cloud Engines 1.1.2 (3.4.22) SATA PHYADDR=0

U-Boot code: 00600000 -> 0067FFF0  BSS: -> 00691750

Soc: 88F6281 A1 (DDR2)
CPU running @ 1200Mhz L2 running @ 400Mhz
SysClock = 400Mhz , TClock = 200Mhz

DRAM CAS Latency = 5 tRP = 5 tRAS = 18 tRCD=6
DRAM CS[0] base 0x00000000   size 128MB
DRAM Total size 128MB  16bit width
Addresses 8M - 0M are saved for the U-Boot usage.
Mem malloc Initialization (8M - 7M): Done
NAND:256 MB
Flash:  0 kB

CPU : Marvell Feroceon (Rev 1)
CLOUD ENGINES BOARD (GUESSED): DISCOVERY:0.1

Streaming disabled
Write allocate disabled


USB 0: host mode
PEX 0: interface detected no Link.
Net:   egiga0 [PRIME]
Hit any key to stop autoboot:  0

NAND read: device 0 offset 0x100000, size 0x300000

Reading data from 0x3ff800 -- 100% complete.
 3145728 bytes read: OK
## Booting image at 00800000 ...
   Image Name:   Linux-2.6.32.18-dockstar
   Created:      2011-02-17   8:10:10 UTC
   Image Type:   ARM Linux Kernel Image (uncompressed)
   Data Size:    3236180 Bytes =  3.1 MB
   Load Address: 00008000
   Entry Point:  00008000
   Verifying Checksum ... Bad Data CRC
CE>>

Is there any way to unbrick?

Yes, setup tftp on another Linux or OSX box and flash the current uBoot image.

In the example below,

192.168.0.100 is the ip address of the other Linux box
192.168.0.200 is the ip address of this Dockstar (assuming it is the Dockstar)
uboot.kwb is the uBoot image that is retrieved by Jeff script (http://projects.doozan.com/uboot/install_uboot_mtd0.sh) for the Dockstar.

and all commands are executed in serial console prompt (interrupt serial console at the countdown).

setenv serverip 192.168.0.100
setenv ipaddr 192.168.0.200
tftp 0x800000 uboot.kwb
nand erase 0x0 0x80000
nand write.e 0x800000 0x0 0x80000

Please note that this instruction left out the details of how to setup tftp. Please make sure tftp is setup and load successfully. And before running the last 2 commands (nans erase and and write.e) try running help at serial console prompt to check if these commands are available.

If you are not sure, please post the output of the first 3 commands before proceeding to the 2 nand commands! And then post the output of the last 2 commands so we can check before rebooting the box.

Update: sorry I've missed the part that you said you can't write to serial console. Now it is an interesting problem :) is the box the Dockstar or other Kirkwood plug?

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)



Edited 2 time(s). Last edit at 01/29/2014 10:58AM by bodhi.

I didn't stress the point that I destroyed the through hole of the rx pin so I cannot interupt anything via serial console. I am going to try to solder to the trace or something but I was wondering if there was any tricks I was missing.

Thanks for everything on this forum...

OK I missed that update. (and I looked for it when I saw that the post was updated.... ) yes it is interesting. Its a kirkwood (goflex home).

thanks

earl,

If it's a GoFlex Home then try UART booting:
http://forum.doozan.com/read.php?3,7852,7852

And you can use uBoot image from this thread for UART booting:
http://forum.doozan.com/read.php?3,12381

After verifying that it booted, put this rootfs on a USB thumb drive:
http://forum.doozan.com/read.php?2,12096

If everything boots OK into Debian, then while in Debian, flash this uBoot to NAND.

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)

ok! :) up and running. mostly. I got into this trouble because I was trying (no good reason [which is in itself a good reason]) to restore to factory uboot. still can't, there is something different about this version of goflexhome i think because I needed to use --no-uboot-check in the script. I used tftp btw. It was just easier for me than uart.
I have a new question. I cannot access the rescue system when no usb is plugged in. I am more concerned now since my whole serial connection is ahnging on a thread literally. It boots into rescue but eth0 does not get an ip. if i ifdown and ifup then it gets. could it have something to do with the fact that the leds are different and part of /etc/network/interfaces ifup echo fails?

rescue:~# ifdown eth0
/bin/sh: can't create /sys/class/leds/dockstar:green:health/trigger: nonexistent directory
rescue:~# ifup eth0
udhcpc (v1.17.1) started
Sending discover...
Sending discover...
Sending select for 192.168.0.230...
Sending select for 192.168.0.230...
Sending select for 192.168.0.230...
Lease of 192.168.0.230 obtained, lease time 604800
deleting routers
route: SIOCDELRT: No such process
adding dns 192.168.0.1
/bin/sh: can't create /sys/class/leds/dockstar:green:health/trigger: nonexistent directory
rescue:~#

thanks for the help so far!

Your rescue system arcNumber must be something other than the GoFlex Home (3338). Looks like it's a Dockstar. Check it with:
fw_printenv arcNumber

But before setting this arcNumber, make sure you can boot it with a kernel that supports it (I don't know if your rescue system supports it) . But first, make sure you can boot with the rootfs above (it works for most Kirwood plugs).

In any case, you should set up netconsole before doing anything else, since you might not have serial console later:
http://forum.doozan.com/read.php?3,14

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)



Edited 1 time(s). Last edit at 01/29/2014 08:02PM by bodhi.

I set up net console and have my serial console hotglued in and back in the case just to hope. I already set my arcnumber to 3089 and it recognizes sata drives. but i am not sure if it will boot from it yet.

thanks again.

with arcNumber 3888 I get:

Error: unrecognized/unsupported machine ID (r1 = 0x00000d0a).

Available machine support:

ID (hex)        NAME
00000690        Marvell DB-88F6281-BP Development Board
00000691        Marvell RD-88F6192-NAS Development Board
00000692        Marvell RD-88F6281 Reference Board
0000078c        Marvell 88F6281 GTW GE Board
00000831        Marvell SheevaPlug Reference Board
00000bb6        Seagate FreeAgent DockStar
00000c11        Seagate Goflex Net
0000085b        QNAP TS-119/TS-219
00000915        Marvell OpenRD Base Board

Please check your kernel config and/or bootloader.

and simillarly usb also does not boot.

with arc 3089 it boots fine and can boot from sata.

The leds still are not right though (and maybe NAND mapping?).
I also still cant get to rescue because it wont pick up an ip.
should I make a new thread for that?



Edited 3 time(s). Last edit at 01/30/2014 10:34AM by earl.

Earl,

The reason is that you are running GoFlex Net uBoot by davygravy (installed by Jeff's script). This works because the GoFlex Home is a closed cousin of the GF Net. The LEDs must be set in Debian using Goflex Net settings, you can find the trigger in /sys/class/leds. If you're not sure, post the output of

ls -lR /sys/class/leds

Rescue system is another story, since you have set arcNumber to the GoFlex Net, and the rescue kernel is not GoFlex Net, so the arcNumber mismatch might give you problem like this. Which rescue kernel version do you have? is it from Jeff or from davygravy?

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)

Iwould at least like the leds to tell me when i can power off.
here's the output of ls -lR /sys/class/leds:

/sys/class/leds:
total 0
lrwxrwxrwx 1 root root 0 Feb  1 20:05 status:green:health -> ../../devices/platform/leds-gpio/leds/status:green:health
lrwxrwxrwx 1 root root 0 Feb  1 20:05 status:orange:fault -> ../../devices/platform/leds-gpio/leds/status:orange:fault
lrwxrwxrwx 1 root root 0 Feb  1 20:05 status:white:left0 -> ../../devices/platform/leds-gpio/leds/status:white:left0
lrwxrwxrwx 1 root root 0 Feb  1 20:05 status:white:left1 -> ../../devices/platform/leds-gpio/leds/status:white:left1
lrwxrwxrwx 1 root root 0 Feb  1 20:05 status:white:left2 -> ../../devices/platform/leds-gpio/leds/status:white:left2
lrwxrwxrwx 1 root root 0 Feb  1 20:05 status:white:left3 -> ../../devices/platform/leds-gpio/leds/status:white:left3
lrwxrwxrwx 1 root root 0 Feb  1 20:05 status:white:right0 -> ../../devices/platform/leds-gpio/leds/status:white:right0
lrwxrwxrwx 1 root root 0 Feb  1 20:05 status:white:right1 -> ../../devices/platform/leds-gpio/leds/status:white:right1
lrwxrwxrwx 1 root root 0 Feb  1 20:05 status:white:right2 -> ../../devices/platform/leds-gpio/leds/status:white:right2
lrwxrwxrwx 1 root root 0 Feb  1 20:05 status:white:right3 -> ../../devices/platform/leds-gpio/leds/status:white:right3

the rescue system is from jeff.

I just want to point out that connected via serial console I see that eth0 is not getting an ip. If i ifdown eth0 and then ifupeth0 it gets an ip.

cd /tmp
wget http://jeff.doozan.com/debian/rescue/install_rescue.sh
chmod +x install_rescue.sh
./install_rescue.sh

Edited 2 time(s). Last edit at 02/01/2014 09:18PM by earl.

Earl,

1. To turn the status LED to green for normal operation, put these statement in your /etc/rc.local (before the exit 0):

# turn on LED
if [ -d /sys/class/leds/status:green:health ]; then
   echo default-on > /sys/class/leds/status:green:health/trigger
   if [ -d /sys/class/leds/status:orange:fault ]; then
      echo none > /sys/class/leds/status:orange:fault/trigger
   fi
fi
exit 0

To turn off the status LED before shutting down, put these in your /etc/rc0.d/Kxxhalt (xx could be 07, 08,… depending on each system installation) before the halt command. Note that it's prudent to wait about 10 seconds for the HDD to spin down before power off.

        log_action_msg "Will now halt"

if [ -d /sys/class/leds/status:green:health ]; then
   echo none  > /sys/class/leds/status:green:health/trigger
fi
	halt -d -f $netdown $poweroff $hddown

}

2. Regarding the rescue system (while in the rescue system using serial console) what is your output of these 2 commands (70 and 75 could be different, depending each system installation):

cat /etc/udev/rules.d/70-persistent-net.rules | grep SUBSYSTEM
cat /lib/udev/rules.d/75-persistent-net-generator.rules | grep -i -C3 'device name whitelist'

See this post for a possible solution if it is related to udev rules: http://forum.doozan.com/read.php?2,13053,13070#msg-13070

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)

I'll hopefully look at that link soon but I am going to put this on hold for a little while my usb ttl was not working so great and i ordered a new one. I'm going to wait till the new one comes to try connecting serial console.

thanks again :)