Skip to contentSkip to conversation

Btrfs scrub on single ssd reports non existing errors on Linux 6.4

DISCUSSION Bagas Sanjaya 1 participant 1 message Aug 3, 2023

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

Hi,

I notice a regression report on Bugzilla [1]. Quoting from it:

Quoted 34 lines

My distribution is a Manjaro 23.0.0, CPU is AMD A10. System disk is an SSD PNY_CS900 120Gb

With Linux 6.4.6.1-MANJARO, a scrub on then system disk root reports errors and some times freeze the linux :

sudo btrfs scrub status /

UUID:             53e62983-fed7-46ed-b97c-23d19af4f26f
Scrub started:    Tue Aug  1 13:26:14 2023
Status:           running
Duration:         0:03:21
Time left:        1:12:50
ETA:              Tue Aug  1 14:42:25 2023
Total to scrub:   85.39GiB
Bytes scrubbed:   3.75GiB  (4.40%)
Rate:             19.13MiB/s
Error summary:    read=80
  Corrected:      34
  Uncorrectable:  46
  Unverified:     0

In this case, this report is the last one before the Linux freeze.
Scrub not always freese, but when it freeze only low level functionnalities as ping or ps are working.

If Y downgrade the linux kernel to 6.1.41-1-MANJARO there is no errors :

sudo btrfs scrub status /

UUID:             53e62983-fed7-46ed-b97c-23d19af4f26f
Scrub started:    Thu Aug  3 14:17:11 2023
Status:           finished
Duration:         0:04:41
Total to scrub:   66.72GiB
Rate:             225.62MiB/s
Error summary:    no errors found
Quoted 1 line

Ask me if you want more details.

See Bugzilla for the full thread.

Anyway, I'm adding this regression to be tracked by regzbot:

#regzbot introduced: v6.1.41..v6.4.6 https://bugzilla.kernel.org/show_bug.cgi?id=217758

Thanks.

[1]: https://bugzilla.kernel.org/show_bug.cgi?id=217758

Signature · 1 line
An old man doll... just what I always wanted! - Clara