Recovering a non-bootable RAID 5 array with bad blocks.

[ccapon]

The QNAP device is a TS-451 (4.14.24-qnap) with 4 x 4GB configured as a RAID-5 array.
After receiving a firmware upgrade request, a restart has left the device unbootable.
I am trying to recover it without losing the data.

The first hard drive shows a solid red light.

Attached a monitor and keyboard.

Tried rebooting:
Decompressing Linux... Parsing ELF... done.
Booting the kernel.
(it just locks up here)

1. Ejected all 4 hard drives.
2. Tried rebooting. It boots successfully to the Setup firmware OS.
3. Inserted all 4 hard drives.
4. Exit the System Install software to a command shell.
5. Log on as admin using default password.

Run:
fdisk -l
/dev/sdb1
/dev/sdc1
/dev/sdd1
/dev/sde1
All 4 drives and their single partitions are present (System: EFI GPT)

Run
badblocks -v /dev/sdb (the first hard drive)
Errors on blocks 1756, 1757, 1758, 1759

There are no bad blocks on the other drives.

(a) Is there a way to recover this system without losing all the data?
(b) Why can't it boot? Shouldn't the RAID-5 array compensate for bad sectors on only one drive?
(c) I'm worried that just replacing the first drive on a non-bootable system will not recover the RAID-5 array. Maybe it can be done from the Setup firmware OS shell, but I don't know how.

Thanks.
Chris.

[dolbyman]

sounds like the LPC degradation.. try the 100 Ohm resistor fix (at least partially .. red LED, stuck on decompressing kernel, etc)

viewtopic.php?t=157459

and as always .. external backups are needed at all times .. not in hindsight when it's too late !

[ccapon]

@dolbyman,
Not the same issue. This device can boot past the 1st beep and in to the System firmware OS. I'm looking for help recovering a RAID-5 array from the system shell.

[dolbyman]

what does

Code: Select all

md_checker

say ?

Here is some hints on how to mount QNAP disks in a foreign system (not that easy)
viewtopic.php?f=25&t=143408&start=15#p702846

[ccapon]

Code: Select all

Welcome to MD superblock checker (v2.0) - have a nice day~
Scanning system...
RAID metadata found!
UUID:           fe4d1583:93bc854e:e8c85a3d:902d05d7
Level:          raid5
Devices:        4
Name:           md1
Chunk Size:     512K
md Version:     1.0
Creation Time:  Aug 7 16:56:47 2016
Status:         OFFLINE
=========================
Enclosure | Port | Block Dev Name | # | Status |    Last Update Time | Events | Array State
=========================
NAS_HOST     1       /dev/sdb3      0   Rebuild  Oct 21 10:39:43 2021  49990     AAAA
NAS_HOST     2       /dev/sdc3      1   Active   Oct 22 01:01:06 2021  51446     .AAA
NAS_HOST     3       /dev/sdd3      2   Active   Oct 22 01:01:06 2021  51446     .AAA
NAS_HOST     4       /dev/sde3      3   Active   Oct 22 01:01:06 2021  51446     .AAA

Odd... it looks like the block device names have a (3) on the end where fdisk -l shows a (1)