Data-Rebuild stuck due to parity2 being disabled from errors...

[Acps]

I have been having issues with moving large amounts of data from my cache drives to the array for the last several months. Started shortly after doubling my ram from 2 to 4 sticks. Today I thought I would remove 2 ram sticks to see if the disk errors would clear up. However I clearly botched what I thought was a simple task into a possible data loss incident? 🤕

 

I'll attach my logs and hope someone could take a look and help me try and dig myself out of this pickle im in now.

 

Ty in advance

~Acps

unraid-diagnostics-20230217-1149.zip

[Squid]

Your disk 4 I'd say is shot

197 Current_Pending_Sector  -O--C-   001   001   000    -    28288

 

Also, what I'm assuming to be parity 2 ST8000DM004-2CX188_ZCT2X75X-20230217-1149 probably had it's SATA connection disturbed when you played with the RAM.  Reseat both ends (and power) 

[Acps]

How do I stop the data-rebuild. The menu buttons are not responding.

Will Parity 2 need to be rebuilt too, aka unassigned and reassigned too? Also do I want to leave disk 4 disabled for now or just unassign the bad drive until I order a replacement?

-Acps


Sent from my iPhone using Tapatalk

[JorgeB]

Parity2 dropped offline but the rebuild appears to still be going, slowly, post new diags to see if it advanced since the last ones.

 

 

[Acps]

The gui was acting like the array was read/writing at 160Mbs, however it never moved passed 0% at 1.97gb after 2 hours. I wasnt able to pause or cancel the rebuild, however I was able to do a powerdown -r, to get the system rebooted and powered down. Server is off site and I plan on grabbing it after work today to work on it this weekend. Snagged the diag after reboot. I also should have nearly 99% of all my syslogs going back atleast to 2021.

 

Any input and how to go recover parity2 while i replaced disk 4 so i dont botch it 😁

unraid-diagnostics-20230217-1323.zip

[JorgeB]

Check/replace cables on parity2 then unassign it, start array, stop array, re-assign it, start array, it can be synced at the same time disk4 is rebuilt.

[Acps]

I likely won't have a replacement for disk 4 for several days. Can I reassign Parity2 and leave disk 4 unassigned to rebuild that disk? Or do both disks need to be assigned together?


I've been wanting to replace my 4-5tb drives with 8tb cmr, likely ST8000VNZ04.

-Acps


Sent from my iPhone using Tapatalk

[trurl]

Since you have dual parity, you can rebuild 2 drives at the same time, or you can just rebuild 1 disk now and rebuild 1 disk later. That applies whether the 2 disks are parity or data or both.

[Acps]

Over the weekend I was able to successfully unassign disk4 and rebuild pairty2. Ran another extended smart test on disk 4 and it completed with read failure again. My replacement drive should be arriving this week.

 

I ran a parity check after my parity rebuild, which also completed with 0 errors. But now parity 2 is disabled again after having many read errors. I have most dockers turned off right now, save Plex.

 

Attached my diags and will run an extended smart test on parity2 overnight, is it likely I have another disk going bad? Parity2 is only 2 years old, while disk 4 is 8 years.

 

Appreciate any help!

 

Ty in advance

~Acps

unraid-diagnostics-20230221-1827.zip

[JorgeB]

It's not logged as disk problem and SMART looks OK, so possibly a power/connection issue, also worth updating the LSI firmware to latest.