Bug#1057843: linux: ext4 data corruption in 6.1.64-1

[Salvatore Bonaccorso]

Source: linux
Version: 6.1.64-1
Severity: grave
Tags: upstream
Justification: causes non-serious data loss
X-Debbugs-Cc: debian-...@lists.debian.org, car...@debian.org, ad...@debian.org

Hi

I'm filling this for visibility.

There might be a ext4 data corruption issue with the kernel released
in the 12.3 bookworm point release (which is addressed in 6.1.66
upstream already).

The report about the regression and some details:

https://lore.kernel.org/stable/20231205122122.dfhhoaswsfscuhc3@quack3/

Regards,
Salvatore

[Salvatore Bonaccorso]

Running the single test with ext4:

# LTP_SINGLE_FS_TYPE=ext4 LTP_DEV_FS_TYPE=ext4 ./preadv03_64
tst_device.c:96: TINFO: Found free device 0 '/dev/loop0'
tst_test.c:1690: TINFO: LTP version: 20230929-194-g5c096b2cf
tst_test.c:1574: TINFO: Timeout per run is 0h 00m 30s
tst_supported_fs_types.c:149: TINFO: WARNING: testing only ext4
tst_supported_fs_types.c:90: TINFO: Kernel supports ext4
tst_supported_fs_types.c:55: TINFO: mkfs.ext4 does exist
tst_test.c:1650: TINFO: === Testing on ext4 ===
tst_test.c:1105: TINFO: Formatting /dev/loop0 with ext4 opts='' extra opts=''
mke2fs 1.47.0 (5-Feb-2023)
tst_test.c:1119: TINFO: Mounting /dev/loop0 to /tmp/LTP_preWBHd7l/mntpoint fstyp=ext4 flags=0
preadv03.c:102: TINFO: Using block size 512
preadv03.c:77: TFAIL: Buffer wrong at 0 have 62 expected 61
preadv03.c:77: TFAIL: Buffer wrong at 0 have 62 expected 61
preadv03.c:66: TFAIL: preadv(O_DIRECT) read 0 bytes, expected 512

Summary:
passed 0
failed 3
broken 0
skipped 0
warnings 0

[Stephan Verbücheln]

Are there any guidelines for affected users who already updated before
they got the warning?

Interesting questions for affected users:
- Is it safe to assume that other filesystems (like BTRFS) are not
affected?
- Does this cause filesystem corruption or only file corruption?
- Does this affect metadata or only file contents?
- Is there a way to detect corrupted files?
- If metadata is not affected, is it safe to assume that all files with
a modification date older than the update are fine?
- Does it help to shut down services (such as Apache) or the whole
machine until the fix is available?

[Umberto Zappi]

Are there any guidelines for affected users who already updated before they got the warning?

For example, is it fine to start with the previous kernel 6.1.0-13-amd64 (version 6.1.55-1) ?

Thanks

UmbertoZ.

[helios....@gmx.ch]

Will a file system check detect the corruptions?
Can it be done online?

Thank you.

[BW]

Online check of ext4:

If your filesystem is located on a logical volume (LVM) then I assume you can make a snapshot and do a check of that.

Make SS:
lvcreate --snapshot --size 1G --name lv_root_SS --chunksize 4k /dev/VG1/lv_root

EXT4 check:
e2fsck -f /dev/dm-3

Remove SS:

lvremove --yes VG1/lv_root_SS

[Jamirais Ismail]

Feedback 

Jamirais Bin Ismail