So on another Unraid Server I have a cache in btrfs RAID1 with 2 Nvme Disks. Lets call them A & B. 

 

Disk B failed really hard, with sectors not being read at all. The system really degraded, the webinterface took 20s+ to load and the scrub/check was in the kbit/s speed.

 

So I bought a new nvme and removed drive B and replaced it with the new Disk C. I then started the full btrfs balance thing.

 

Now while doing the balance i am still getting some nvme 0x0 aborts, which chatgpt says is due to disk A also being on life-support. 

 

So here's the question: How do I get Unraid to mirror the Data to the other disk again? The full balance is running, yet the dashboard (and disk log) is showing 0 activity for disk C. Therefore, is it not copying the files?

 

Edit: System is still really slow for some reasons, it took about 15s to generate the diags. But thats not the primary concern now.

 

Edited January 24, 20251 yr by 012315

There are errors but the balance appears to be continuing, post current diags for comparison.

Yeah, now there are actual read errors.

Alright, but what would be the best course of action?

Also judging by Device Info, the new SSD is still completely blank and only had like 20mb written to it. Why is it not doing raid1?

13 hours ago, 012315 said:

Alright, but what would be the best course of action?

Since both pool devices failed I would recommend creating a new pool with a new device(s) and restoring the data from a backup, assuming one exists.

On 1/5/2025 at 12:00 PM, JorgeB said:

Since both pool devices failed I would recommend creating a new pool with a new device(s) and restoring the data from a backup, assuming one exists.

Backup exists, so I will create a new pool and move the data there. I've already restored the backup to the array to reintroduce redundancy.

 

The question remains why unraid did not mirror the RAID1 to the new nvme drive though

Because looks like there were already issues with the device.