Hi!
I recently aquired for just a few € a Pogoplug V3 (POGO-P24). It had been modified in an unknown way, but I managed to use the serial console to remove some strange static IP configuration and reset the SSH password so that I now have SSH access.
I tried to boot the rootfs from 17 May 2020 but it didn't work (I didn't try to find out why because this wasn't a priority).

Then I tried to proceed to install uBoot following the great instructions provided in these forums. However, when checking the contents of proc/mtd I found the following:

/tools # cat /proc/mtd
dev:    size   erasesize  name
mtd0: 08000000 00020000 "NAND 128MiB 3,3V 8-bit"
mtd1: 00e00000 00020000 "boot"
mtd2: 07200000 00020000 "rootfs"

Should it be safe to proceed? I think I managed to backup there mtd but I cannot backup current envs (fw_printenv returns a memory allocation error and blparam is nowhere to be found).

Also, am I right to assume that bad blocks should pose no problem?

<6>[    1.670000] Scanning device for bad blocks
<4>[    1.710000] Bad eraseblock 653 at 0x0000051a0000
<4>[    1.740000] Bad eraseblock 954 at 0x000007740000

Thanks!

Nuno,

> However, when checking the contents of proc/mtd I
> found the following:
>

> /tools # cat /proc/mtd
> dev:    size   erasesize  name
> mtd0: 08000000 00020000 "NAND 128MiB 3,3V 8-bit"
> mtd1: 00e00000 00020000 "boot"
> mtd2: 07200000 00020000 "rootfs"
>

> Should it be safe to proceed?

Yes it is safe. The mtds are actually the same. This is stock definition, where mtd0 is the whole flash. As in overlay, mtd1 starts at 0, mtd2 starts at 07200000.

But to be absolutely certain, you can adjust the mtd0 in the command to become mtd1. Like below.

Quote

3. Be sure to save your current uBoot image in mtd0, or have a backup image saved somewhere as precaution.

Dumping NAND mtd0:
nanddump --noecc --omitoob -f mtd1 /dev/mtd1

/usr/sbin/flash_erase /dev/mtd1 0x0 6
/usr/sbin/nandwrite /dev/mtd1 uboot.spl.2013.10.ox820.850mhz.mtd0.img
/usr/sbin/nandwrite -s 262144 /dev/mtd1 uboot.2015.10-tld-2.ox820.mtd0.img
/usr/sbin/flash_erase /dev/mtd1 0x00100000 1
/usr/sbin/nandwrite -s 1048576 /dev/mtd1 uboot.2015.10-tld-2.ox820.environment.img

> Also, am I right to assume that bad blocks should
> pose no problem?
> <4>[ 1.710000] Bad eraseblock 653 at
> <4>[ 1.740000] Bad eraseblock 954 at

Yes it is safe to flash, these bad blocks are far away in NAND, not in the 1st 2MB.

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

bodhi,

Thanks!

Quote

The mtds are actually the same.

I was suspecting that, because I noticed the size of mtd0 was the sum of mtd1 and mtd2... :-)

Before doing what you suggested, I tried to modify fw_env.config to reference mtd1 instead of mtd0. Now there's no memory allocation error, and fw_printenv appears to work, but I do see a warning/error:

/tools # ./fw_printenv
Warning: Bad CRC, using default environment
bootcmd=bootp; setenv bootargs root=/dev/nfs nfsroot=${serverip}:${rootpath} ip=${ipaddr}:${serverip}:${gatewayip}:${netmask}:${hostname}::off; bootm
bootdelay=5
baudrate=115200

Should I worry about this or can I just disregard this message? I do not expect to have any use for default envs except for the MAC, which I already have from the serial console boot log, which I previously saved.

Nuno,

> Should I worry about this or can I just disregard
> this message?

No need to worry. After you flash the env image, fw_printenv will be OK.

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

Thanks!
I did what you suggested and everything is fine now. The Pogoplug booted to Debian and I was able to set a swapfile and update all packages.
Now I just have to update the kernel and maybe try updating Debian to bullseye...