Disk problem - BTRFS-Error

[knarf0007]

Hi All,

 

It seems that I have a problem with one of my SSDs. What should be the next steps? Thanks.

 

Frank.

unraid-server-diagnostics-20220926-1850.zip

[knarf0007]

here some additional information:

root@UnRaid-Server:/mnt# btrfs dev stats /mnt/cache_docker_vm
[/dev/mapper/sdi1].write_io_errs    0
[/dev/mapper/sdi1].read_io_errs     0
[/dev/mapper/sdi1].flush_io_errs    0
[/dev/mapper/sdi1].corruption_errs  0
[/dev/mapper/sdi1].generation_errs  0
[/dev/mapper/sde1].write_io_errs    0
[/dev/mapper/sde1].read_io_errs     0
[/dev/mapper/sde1].flush_io_errs    0
[/dev/mapper/sde1].corruption_errs  0
[/dev/mapper/sde1].generation_errs  0
root@UnRaid-Server:/mnt# btrfs dev stats /mnt/cache
[/dev/mapper/sdj1].write_io_errs    0
[/dev/mapper/sdj1].read_io_errs     0
[/dev/mapper/sdj1].flush_io_errs    0
[/dev/mapper/sdj1].corruption_errs  1298
[/dev/mapper/sdj1].generation_errs  0
[/dev/mapper/sdk1].write_io_errs    0
[/dev/mapper/sdk1].read_io_errs     0
[/dev/mapper/sdk1].flush_io_errs    0
[/dev/mapper/sdk1].corruption_errs  1010
[/dev/mapper/sdk1].generation_errs  0

[trurl]

Why do you have 50G libvirt.img? Default 1G is more than enough. 60G docker.img is also at least twice what you need.

 

24 minutes ago, knarf0007 said:

a problem with one of my SSDs

syslog seemed to indicate corruption on both cache disks during mount.

 

Have you done memtest lately?

[knarf0007]

No, I didn't perform any memtests.

[andreidelait]

The fact that you have corruption on two different drives makes me think that could be some RAM related issue, as mentioned above.
I had this happen before. One of my four DIMMs was defective. Memtest worked for me. I removed the faulty RAM and it works great since.

[knarf0007]

I made a mem-test with works out fine. I changed to two new SSDs. Until now, no further problems. I confess, I downgraded from 6.11.0 to 6.10.3.

[JorgeB]

See here for better pool monitoring, so you know as soon as possible if more issues are found.