Hi All,

 

Up until I have self-served with the great content on these forums, however I haven't found the answer for this issue!

I have suddenly found that my cache drive is read-only. This was reported by 'Fix Common Problems' and I also noticed I wasn't able to delete some of my dockers etc... which resided on the cache.

My cache drives consist of 2 1TB NVMe drives in Raid 1 they are formatted as BTRFS. I tried to scrub the main drive in the pool to correct any errors but it always auto-aborts about 5 seconds in.

I had a look at the disk logs for the cache and can see lots of the following:

 

Jan 12 03:36:55 Unraid kernel: BTRFS critical (device nvme0n1p1): corrupt leaf: root=7 block=734560256 slot=5, bad key order, prev (18446744073709551606 128 195796140032) current (18446744073709551606 128 195762913280)
Jan 12 03:36:55 Unraid kernel: BTRFS critical (device nvme0n1p1): corrupt leaf: root=7 block=734560256 slot=5, bad key order, prev (18446744073709551606 128 195796140032) current (18446744073709551606 128 195762913280)
Jan 12 20:19:13 Unraid kernel: BTRFS critical (device nvme0n1p1): corrupt leaf: root=7 block=734560256 slot=5, bad key order, prev (18446744073709551606 128 195796140032) current (18446744073709551606 128 195762913280)
Jan 12 20:19:13 Unraid kernel: BTRFS critical (device nvme0n1p1): corrupt leaf: root=7 block=734560256 slot=5, bad key order, prev (18446744073709551606 128 195796140032) current (18446744073709551606 128 195762913280)

Weirdly after trying to scrub and restarting my Unraid box, it now doesn't seem to be read-only... but I darn't do anything for fear of losing some information.

 

Can anyone advise what is best to do? Can I repair the BTRFS filesystem errors? If I can't how would it be best to save my data to the array and start over?

 

Also attached my full diag-logs for anyone that needs it.

 

Thanks,
Steve

 

unraid-diagnostics-20200112-2021.zip

11 hours ago, pearce1340 said:

Can I repair the BTRFS filesystem errors?

With btrfs it's usually better to reformat and start over since fsck isn't very reliable yet, if after that it gets corrupt again without an apparent reason it might be a good idea to run mestest, also make sure RAM isn't overclocked.

Ah ok, so you can't run with RAM overclocked? I have enabled an XMP profile on my RAM as I didn't think this would be a problem.

I think I will move everything off of it and then reformat etc...

Will keep you updated.

Thanks,

Steve

16 minutes ago, pearce1340 said:

Ah ok, so you can't run with RAM overclocked? I have enabled an XMP profile on my RAM as I didn't think this would be a problem.

Overclocking is not recommended with any 24/7 server, particularly with storage server.

It's not a can vs cannot.

It's a risk vs reward assessment.