Storage Pool Status Error (Not Active) in RAID 10 on TVS-EC1080

[Jayden67]

In Storage & Snapshots, one of the storage pools has Status Error (Not Active). The UI showed that 2 of the 4 disks in the RAID 10 configuration were missing. Replace the 2 "bad" disks with higher capacity disks assuming that would resolve the problem. It now shows 4 good disks in the RAID group, but the storage pool and volume are still unusable in Error state. What commands do I need to run in SSH or options to select in the UI to restore the storage pool?

qnap error.png

Md_checker output (abridged):

Code: Select all

RAID metadata found!
UUID:           1a7a74e9:459085d1:16c1c681:c48f731c
Level:          raid10
Devices:        4
Name:           md3
Chunk Size:     512K
md Version:     1.0
Creation Time:  Feb 17 18:58:50 2018
Status:         OFFLINE
===============================================================================
 Disk | Device | # | Status |   Last Update Time   | Events | Array State
===============================================================================
 --------------  0  Missing   -------------------------------------------
 --------------  1  Missing   -------------------------------------------
   1  /dev/sdj3  2   Active   Aug 15 19:15:03 2023     1122   A.AA                                                                                                                                                                           
   4  /dev/sdk3  3   Active   Aug 15 19:15:03 2023     1122   A.AA                                                                                                                                                                           
===============================================================================

Complete SMART test on both new disks still running, will probably take another day to finish on the 20TB disk. FYI, when the storage pool is back online, I will be swapping out that 4TB drive. Replace Disks options is disabled.

I don't know what caused the 2 drives to fail or how long the storage pool has been in error state. Luckily, the volumes on that pool weren't being used.
Data is not backed up, but I don’t care about the data. It contains a virtual machine I haven't booted up in years. I realize that it would probably be faster/easier to destroy and recreate the storage pool instead of trying to troubleshoot, but I want to learn how to fix this problem in case it happens to the other storage pool on the NAS which has important data I don’t want to lose.

Units is TVS-EC1080 with firmware 4.3.4.0483 Build 20180213. All 10 disk bays plus cache is populated, all disks are utilized.

[dolbyman]

If two drives of a mirror pair of a RAID10 are busted /swapped..then the RAID is toast.

If it had been a RAID6 it would still be working

[Jayden67]

Thanks for the advice. I'll recreate the storage pool, but still with RAID10 because I value the enhanced performance over durability.
Follow up question: Is there a way to know which disks are part of which pair?

[dolbyman]

The mirror pairs are apparently not easy to find on old md versions (that QNAP uses) see here

https://superuser.com/questions/567267/ ... d-10-array