[GT-AXE11000] NAND MTD10 손상 (오류-74) - 공장 파티션 수리에 도움이 필요합니다

[jr lee]

Hi everyone,

I am an ASUS GT-AXE11000 user living in South Korea.
I'm posting this because I've hit a dead end while trying to repair my router's hardware failure, and I'm hoping to get some technical advice from the experts here.

The Situation:

My router's NAND chip had bad blocks, so I had it replaced professionally.
However, the original backup of my mtd10 (misc1 / Factory) partition seems to have bit-flips (corruption).
Even after running ubiformat /dev/mtd10 -y, when I restore my backup using nandwrite, the kernel logs immediately show error -74 (ECC error) and the system becomes unstable (freezing/reboots).
Current Status:

I am currently running the router with the mtd10 partition wiped (empty), using wl_nocal=1 in NVRAM.
The router works, but I have no LEDs and no App connectivity due to the missing board structure.
Why no RMA?

Since I am in Korea using a US unit, the shipping and inspection fees ($35+) for RMA are prohibitively expensive for me.
My Question:
Is there any way to "repair" a corrupted mtd10 image manually?
I suspect the header or the empty space padding is corrupted.
If anyone has knowledge about the structure of the GT-AXE11000's factory partition, or can point me in the right direction to fix the ECC errors, I would be extremely grateful.

(Note: I am NOT looking to clone a device. I have my original stickers with MAC/Serial and just want to restore my own router's functionality.)

Thank you.

 

[ColinTaylor]

How did you restore the other partitions on the flash memory, e.g. the boot loader?

 

[jr lee]

How did you restore the other partitions on the flash memory, e.g. the boot loader?

I purchased a T48 programmer (XGecu), EX-2 and a new NAND chip myself, and entrusted the work to a local hardware expert.

He desoldered the old NAND and used the T48 programmer to dump the full image and clone it to the new chip.
Since the router boots up fine and I can access the CFE mini-web server, I believe the Bootloader (CFE) was copied successfully.

The issue is that the source NAND had bad blocks specifically in the mtd10 area. It seems the T48 programmer copied the corrupted data (bit-flips) exactly as it was to the new chip, which is why I am seeing ECC errors now.

 

[ColinTaylor]

I have sent you a direct message.