On 8/4/2024 at 11:49 AM, JorgeB said:

You can try, but the problem may only manifest itself with multiple drives running.

 

@JorgeB

Found out it was a freaking bad USB Stick..

 

I got a new JetFlash 780 today so hopefully that will last a bit longer.

 

https://docs.unraid.net/unraid-os/manual/storage-management/#rebuilding-a-drive-onto-itself

 

On step "Step 6" it now says "New Device" and left of the "Start" it says "Stopped. Replacement disk installed." Is that correct?

 

 

Edited August 13, 20241 yr by DigitalLF

Did a new config. tried to start without paritys. Disk 7 was not mounted.. "Unmountable: Unsupported or no file system"

 

So because of that 2th try failed after 10h 30m i now have a drive that is Unmountable..

Edited August 6, 20241 yr by DigitalLF

9 hours ago, DigitalLF said:

Did a new config.

You cannot use a new config to rebuild a failed drive.

1 hour ago, JorgeB said:

You cannot use a new config to rebuild a failed drive.

 

Is it possible to try to repair the file system for BTRFS? I also have a USB backup so what if i restore that and try again? Me and @DiscoverIt talked quite a lot on Discord last night and because i can't trust my Parity 1 and Parity 2 is Offline i went with "new config" and Disk 7 won't mount ("Unmountable: Unsupported or no file system").

 

i also tried to mount it as btrfs

 

mount -t btrfs /dev/sdi1 /mnt/test

[  291.887042] BTRFS info (device sdi1): first mount of filesystem d9c9a06f-7a8e-424e-9f5c-665d1ba0cb5c
[  291.887065] BTRFS info (device sdi1): using crc32c (crc32c-intel) checksum algorithm
[  291.887075] BTRFS info (device sdi1): using free space tree
[  292.014565] BTRFS error (device sdi1): parent transid verify failed on logical 8036978900992 mirror 1 wanted 20143 found 19737
[  292.020624] BTRFS error (device sdi1): parent transid verify failed on logical 8036978900992 mirror 2 wanted 20143 found 19737
[  292.020632] BTRFS warning (device sdi1): couldn't read tree root
[  292.023728] BTRFS error (device sdi1): open_ctree failed

 

Edited August 7, 20241 yr by DigitalLF

1 hour ago, DigitalLF said:

parent transid verify failed on logical 8036978900992 mirror 1 wanted 20143 found 19737

This error is fatal, but you may be able to recover some data using the options here, btrfs restore is usually the best one for this error.

8 minutes ago, JorgeB said:

This error is fatal, but you may be able to recover some data using the options here, btrfs restore is usually the best one for this error.

 

root@unRAID:~# mount -o rescue=all,ro /dev/sdi1 /mnt/test
mount: /mnt/test: wrong fs type, bad option, bad superblock on /dev/sdi1, missing codepage or helper program, or other error.
       dmesg(1) may have more information after failed mount system call.
root@unRAID:~# 

 

Seems like it can't even see a partition so i'm screwed screwed..

 

fdisk tells me:

Device     Start         End     Sectors  Size Type
/dev/sdi1     64 35156656094 35156656031 16.4T Linux filesystem

 

Edited August 7, 20241 yr by DigitalLF

2 hours ago, DigitalLF said:

Seems like it can't even see a partition

sdi1 is the partition, like mentioned, btrfs restore is the option you should try, option 2

26 minutes ago, JorgeB said:

sdi1 is the partition, like mentioned, btrfs restore is the option you should try, option 2

So with that command i would require another drive. I can't afford that so i will just have to loose the data and hope it's nothing i will miss..

@JorgeB I thought i solved it with replacing the USB drive with a JetFlash 780. I did a rebuild with the two parity drives and that took around 36h and after that i was able to use the server for a few days but this morning i had a new crash with all drives going offline...

 

In the logs it states error:

Aug 13 09:12:08 unRAID kernel: mpt3sas_cm0: log_info(0x311201ff): originator(PL), code(0x12), sub_code(0x01ff)
Aug 13 09:12:08 unRAID kernel: mpt3sas_cm0: log_info(0x31110d01): originator(PL), code(0x11), sub_code(0x0d01)

 

And from that i found another post here where you also answered.

 

 

I really don't know what would be my next step..

 

I could possibly borrow disks from a friend and connect drives directly to SATA on motherboard (X9SRL-F) and see if i get the same errors.

 

So what i think right now.. One of the "HBA", "Cables", "Backplane" must be the failing part..

 

Any thoughts on what i should do next? I haven't even tried to start the array this time..

Please post the current diags.

4 hours ago, JorgeB said:

Please post the current diags.

I also sent you a stripped syslog and a dmesg and a image from Proxmox for this LSI card on Discord.

unraid-diagnostics-20240813-2144.zip

Edited August 13, 20241 yr by DigitalLF

The server was rebooted twice since the errors you mentioned, so they are gone from the logs, post new ones if it happens again.

 

11 hours ago, DigitalLF said:

on Discord.

Better to use the forum to contact me, I don't usually check Discord

Just now, JorgeB said:

The server was rebooted twice since the errors you mentioned, so they are gone from the logs, post new ones if it happens again.

 

Better to use the forum to contact me, I don't usually check Discord

Sorry for that! Just sent you the logs here.

 

 

Post them in the thread please.

6 hours ago, JorgeB said:

Post them in the thread please.

Could it be the backplane? I don't think Gooxi stand for quality.

It could.

On 8/14/2024 at 5:21 PM, JorgeB said:

It could.

I now gotten 4 drives where i can test around.. Should i just try "New Config" again and just add the 4 new (used) drives? I can connect the drives directly to the motherboard and test if it's the motherboard/cpu/ram.. I'm just unsure what would be the best next step.. Otherwise i could try with breakout cables (miniSAS-4xSATA).

 

The dmesg file did include some codes that i tried to google but i did not get to much information out of it..

1 hour ago, DigitalLF said:

Should i just try "New Config" again and just add the 4 new (used) drives?

Do you mean create a new array with just the test drives or add them to the existing one?

4 minutes ago, JorgeB said:

Do you mean create a new array with just the test drives or add them to the existing one?

Sorry for being unclear. just the "new" drives in a array to not mess up my data even worse... 3 disconnects with my real array is really stressful.

 

I have just now connected all the "new" drives plus a extra 1tb i had so 5 in total directly to the motherboard to test the motherboard, cpu, ram, psu. So i removed the HBA, dual cables, Backplane from this equation for now.. I have a gut feeling it must be the backplane like in my old case from the same company called Gooxi. Where i asked the company to buy a new backplane and he just sent me two new once for free so i think they know its low quality and this case is 6 years old...

Yes, that's fine and a good way to test without risking your data.