Re: question to btrfs scrub

[Forza <forza@tnonline.net>]

On 2023-06-29 10:26, Bernd Lentes wrote:
> Hi guys,
> 
> i have a BTRFS volume which produces a lot of errors in the syslog.
> Here I got the recommendation to start a “btrfs scrub” on that volume.
> I made an image of that volume (with dd) and started the scrub on that.
> That’s the result:
> 
> ha-idg-1:/mnt/sdc1/ha-idg-1/image # btrfs scrub start -B /mnt/image/
> 
> scrub done for bbcfa007-fb2b-432a-b513-207d5df35a2a
> Scrub started:    Tue Jun 27 20:47:26 2023
> Status:           finished
> Duration:         35:39:48
> Total to scrub:   5.07TiB
> Rate:             40.16MiB/s
> Error summary:    csum=1052
>    Corrected:      0
>    Uncorrectable:  1052
>    Unverified:     0
> ERROR: there are uncorrectable errors
> 
> 1052 checksum errors on a 5TB volume. Is that much, or is that normal ?
> What can I do ?
> Start a btrfs check ? First on the image before on the original ?
> 

Uncorrectable errors means there were some corruptions that Btrfs could 
not correct using a good copy (RAID/DUP profiles). Those corruptions 
could be different things, for example media errors on the disk drive.

Is it just a single disk that you have in this filesystem?

What does smartctl -x /dev/xxx show? Especially look at the table 
containing Uncorrectable_Error_Cnt or Reallocated_Sector_Ct.

You can also issue a drive self-test using `smartctl -t long /dev/xxx`

~F