On Tue, Nov 27, 2018 at 10:16:52PM +0800, Qu Wenruo wrote:
>
> But it's less a concerning problem since it doesn't reach latest RC, so
> if you could reproduce it stably, I'd recommend to do a bisect.

No problem to bisect, usually.
But right now it's not possible for me, I explain further.
Anyway, here the rest of the story.

So, in the end I:
a) booted with 4.20.0-rc4
b) updated backup
c) did the btrfs check --read-only
d) seven steps, everything is perfect
e) no complains on screen or in logs (never had)
f) so, started to compile linux-next 20181128 (on another partition)
e) without using (reading or writing) on /home, I started
f) btrfs filesystem defrag -v -r -t 128M /home
g) it worked without complain (in screen or logs)
h) then, reboot with kernel tag 20181128
i) and no way to mount:

------
nov 28 15:44:03 glet kernel: BTRFS: device label home devid 1 transid 37360 /dev/mapper/cry-home
nov 28 15:44:04 glet kernel: BTRFS info (device dm-3): use lzo compression, level 0
nov 28 15:44:04 glet kernel: BTRFS info (device dm-3): turning on discard
nov 28 15:44:04 glet kernel: BTRFS info (device dm-3): enabling auto defrag
nov 28 15:44:04 glet kernel: BTRFS info (device dm-3): disk space caching is enabled
nov 28 15:44:04 glet kernel: BTRFS info (device dm-3): has skinny extents
nov 28 15:44:04 glet kernel: BTRFS error (device dm-3): bad tree block start, want 2150302023680 have 17816181330383341936
nov 28 15:44:04 glet kernel: BTRFS error (device dm-3): failed to read block groups: -5
nov 28 15:44:04 glet kernel: BTRFS error (device dm-3): open_ctree failed
------

l) get back to 4.20.0-rc4
m) mounted, but after a few minutes, I get this:

------
nov 28 15:51:23 glet kernel: BTRFS warning (device dm-3): block group 2199347265536 has wrong amount of free space
nov 28 15:51:23 glet kernel: BTRFS warning (device dm-3): failed to load free space cache for block group 2199347265536, rebuilding it now
nov 28 15:51:23 glet kernel: BTRFS warning (device dm-3): block group 2196126040064 has wrong amount of free space
nov 28 15:51:23 glet kernel: BTRFS warning (device dm-3): failed to load free space cache for block group 2196126040064, rebuilding it now
nov 28 15:52:09 glet kernel: BTRFS warning (device dm-3): block group 2184314880000 has wrong amount of free space
nov 28 15:52:09 glet kernel: BTRFS warning (device dm-3): failed to load free space cache for block group 2184314880000, rebuilding it now
nov 28 15:52:09 glet kernel: BTRFS warning (device dm-3): block group 2183241138176 has wrong amount of free space
nov 28 15:52:09 glet kernel: BTRFS warning (device dm-3): failed to load free space cache for block group 2183241138176, rebuilding it now
nov 28 15:52:53 glet kernel: BTRFS warning (device dm-3): block group 2152102625280 has wrong amount of free space
nov 28 15:52:53 glet kernel: BTRFS warning (device dm-3): failed to load free space cache for block group 2152102625280, rebuilding it now
nov 28 15:54:13 glet kernel: BTRFS warning (device dm-3): block group 2530059747328 has wrong amount of free space
nov 28 15:54:13 glet kernel: BTRFS warning (device dm-3): failed to load free space cache for block group 2530059747328, rebuilding it now
nov 28 15:55:10 glet kernel: BTRFS warning (device dm-3): block group 2151028883456 has wrong amount of free space
nov 28 15:55:10 glet kernel: BTRFS warning (device dm-3): failed to load free space cache for block group 2151028883456, rebuilding it now
nov 28 15:55:48 glet kernel: BTRFS warning (device dm-3): block group 2203642232832 has wrong amount of free space
nov 28 15:55:48 glet kernel: BTRFS warning (device dm-3): failed to load free space cache for block group 2203642232832, rebuilding it now
------

n) and then read-only mode:

------
[ 1058.996960] BTRFS error (device dm-3): bad tree block start, want 2150382092288 have 159161645701828393
[ 1058.996967] BTRFS: error (device dm-3) in __btrfs_free_extent:6831: errno=-5 IO failure
[ 1058.996969] BTRFS info (device dm-3): forced readonly
[ 1058.996971] BTRFS: error (device dm-3) in btrfs_run_delayed_refs:2978: errno=-5 IO failure
[ 1059.002857] BTRFS error (device dm-3): pending csums is 97832960
------

So, ok, for the moment I'm very sorry I can't help you with bisect, because I have to
revert to ext4. This is the laptop I use to work with.

If I can help you investigating, just tell me.

Thanks for your time,
Gelma