I keep having drive failures, on brand new disks. I'm at a loss. 

 

Between drives, and the log occasionally just losing its mind and filling up, breaking the system, Between reboots, seizures and drive failures something is really wrong. I really need some help. 

 

vision-diagnostics-20241211-1015.zip

have you powered off the device and confirmed and cheeked all power and data cables?

this sounds like a bad sata cable.

Edited December 10, 20241 yr by bmartino1

thanks @bmartino1, That's fantastic advice.

I shut down and reseated all the sata cables coming from the HBA and that's fixed the logging issue, and there doesn't immediately seem to be errors on the parity drive any further. As I suspected, there appears to have been multiple concurrent issues though. There are still 2 unmountable disks though which confuses me. 

 

Another diagnostic after reseating the cables if it helps. 

vision-diagnostics-20241211-1051.zip

Edited December 10, 20241 yr by Bakes32

I've not encountered the unmountable disk issue. nor how to fix / emulate when this occurs.
this may be something a more advance user would need to look at. @JorgeB

 

Check filesystem on emulated disk2 and disk3, run it without -n.

Thanks so much for this. 

How long does this process normally take? (they are 2tb drives at about 75% capacity).  Both repair processes have been running for nearly 12 hours. I'm wondering if something is borked. 

 

Both say: 

Phase 1 - find and verify superblock couldn't verify primary superblock -
not enough secondary superblocks with matching geometry !!!
attempting to find secondary superblock ...

 

 

Ahh i'm just being impatient. I'm sure it'll finish.

Edited December 11, 20241 yr by Bakes32

Both disks exited with what appears to be the same result to my eyes. 
 

phase 1 -find and verify superblock

couldn't identify superblock  - not enough secondary superblocks with matching geometry

 

I'm unsure if this means I should attempt to restart the array in normal mode, or perform another action. No clear outcome to my eyes. I'd really appreciate some input from the clever folk. 

Edited December 12, 20241 yr by Bakes32
additional info

new diag since the repairs finished

vision-diagnostics-20241212-1358.zip

22 hours ago, JorgeB said:

Check filesystem on emulated disk2 and disk3, run it without -n.

no clear outcome from that (see post above). Does that mean start the array in normal mode, or perform another step, or i'm boned? 

2 hours ago, Bakes32 said:

Both disks exited with what appears to be the same result to my eyes. 

How did you run the filesystem check?

3 minutes ago, JorgeB said:

How did you run the filesystem check?

From the Disk Settings page > Check Filesystem Status > Check Button (with the default -n argument removed, no arguments)

this is the output of the filesystem check dialog btw (i can't paste it into the forum response as text, the cloudflare proxy on my work laptop blocks the submission as potential scripting lol). 

 

It never gets past Phase 1 though and ultimately says:

Sorry, could not find valid secondary superblock
Exiting now.

unraid feedback.txt

That's odd since there are signs of xfs filesystem for both disks in the diags, reboot and post new diags after array start.

Diags after a reboot

vision-diagnostics-20241212-2320.zip

Just to double check, start the array in maintenance mode, then type:

 

xfs_repair -v /dev/md3p1

 

Do you see the same?

Last time running this from the GUI took more than 12 hours, but the initial run of that command after restarting in maint mode is showing the exact same after a few mins. From what I've read elsewhere, typically that command takes a few mins to run. 

That's odd, but in that case, the filesystem is not repairable by xfs_repair, if you have backups, reformat the disks and restore from a backup, if you don't, you can try a file recovery app like UFS explorer.

That's a shame, but thanks so much all the same, I really appreciate your help.