Re: misc-next and for-next: kernel BUG at fs/btrfs/extent_io.c:2350! during raid5 recovery

[Wang Yugui <wangyugui@e16-tech.com>]

Hi,

With a modified script based on this reproducer,
'uncorrectable errors' is easy to reproduce

kernel:
	5.20.0(2022/08/09) + some patches in misc-next
	0*-btrfs-scrub-properly-report-super-block-errors-in-sy.patch
	0*-btrfs-scrub-try-to-fix-super-block-errors.patch
	0*-btrfs-fix-lost-error-handling-when-looking-up-extend.patch

script:
	source ~/xfstests/local.config

	SCRATCH_DEV_ARRAY=($SCRATCH_DEV_POOL)
	umount $SCRATCH_MNT
	set -e
	uname -a

	mkfs.btrfs -f -draid5 -mraid1 ${SCRATCH_DEV_ARRAY[@]}
	mount ${SCRATCH_DEV_ARRAY[0]} $SCRATCH_MNT # -o
	compress=zstd,noatime
	mkdir -p $SCRATCH_MNT/dir1

	/bin/cp -a /usr/hpc-bio $SCRATCH_MNT/dir1/
	sync
	du -sh $SCRATCH_MNT
	cd $SCRATCH_MNT
	while true; do
		find -type f -exec cat {} + > /dev/null
	done &

	for((i=0;i>=0;++i)); do

		sync; sleep 2; sync; sleep 4; sync; sleep 20; # change the
	device to discard in every loop
		j=$(( i % ${#SCRATCH_DEV_ARRAY[@]} ))
		/usr/sbin/blkdiscard -f ${SCRATCH_DEV_ARRAY[$j]} >/dev/null 2>&1

		btrfs scrub start -Bd $SCRATCH_MNT | grep
'summary\|Uncorrectable'
done &
wait

Result:
	'uncorrectable errors' is reported in 2nd loop.

Is this a problem of this test script or btrfs kernel?

Best Regards
Wang Yugui (wangyugui@e16-tech.com)
2022/08/12


> > Any chance you could share a script for your reproducer?
> 
> The simplest reproducer is some variant of:
> 
> 	mkfs.btrfs -draid5 -mraid1 /dev/vdb /dev/vdc /dev/vdd
> 	mount /dev/vdb /mnt -ocompress=zstd,noatime
> 	cd /mnt
> 	cp -a /40gb-test-data .
> 	sync
> 	while true; do
> 		find -type f -exec cat {} + > /dev/null
> 	done &
> 	while true; do
> 		cat /dev/zero > /dev/vdb
> 	done &
> 	while true; do
> 		btrfs scrub start -Bd /mnt
> 	done &
> 	wait
> 
> but it can take a long time to hit a failure with something that gentle.
> I throw on some extra test workload (e.g. lots of rsyncs) to keep the
> page cache full and under memory pressure, which seems to speed up the
> failure rate to once every few hours.