Re: 3.2-rc4: scrubbing locks up the kernel, then hung tasks on boot

From: Martin Steigerwald <hidden>
Date: 2012-03-19 08:31:24

Am Sonntag, 18. M=C3=A4rz 2012 schrieb Greg Kroah-Hartman:

On Sat, Mar 17, 2012 at 10:43:25AM +0100, Martin Steigerwald wrote:

quoted

Hi Greg, Chris, Arne, hi everyone,
=20
Am Freitag, 16. M=C3=A4rz 2012 schrieb Martin Steigerwald:

quoted

Am Donnerstag, 15. M=C3=A4rz 2012 schrieb Chris Mason:

quoted

On Thu, Mar 15, 2012 at 06:32:19PM +0100, Martin Steigerwald wr=

ote:

quoted

Am Samstag, 25. Februar 2012 schrieb Arne Jansen:

quoted

On 02/24/12 16:51, Martin Steigerwald wrote:

quoted

Am Samstag, 21. Januar 2012 schrieb Martin Steigerwald:

quoted

Am Samstag, 21. Januar 2012 schrieb Martin Steigerwald:

quoted

I still have this with 3.2.0-1-pae - which is a debian =

kernel

quoted
quoted
quoted
quoted
quoted
quoted
quoted
quoted
based on 3.2.1.
=20
When I do btrfs scrub start / the machine locks immedia=

tely

quoted
quoted
quoted
quoted
quoted
quoted
quoted
quoted
up hard.
=20
Then usually on next boot it stops on space_cache enabl=

ed

quoted
quoted
quoted
quoted
quoted
quoted
quoted
quoted
message, but not  the one for /, but the one for /home =

which

quoted

quoted
quoted
quoted
quoted
quoted
is mounted later.

=20
[=E2=80=A6]
=20

quoted

I now tested scrubbing /home which is a different BTRFS
filesystem on the same machine.
=20
Then the scrub is started, scrub status tells me so, but
nothing happens, no block in/out activity in vmstat, no =

CPU

quoted
quoted
quoted
quoted
quoted
quoted
quoted
related activity in top.
=20
btrfs scrub cancel then hangs, but not the complete mach=

ine,

quoted
quoted
quoted
quoted
quoted
quoted
quoted
only the process.
=20
I had this once on my T520 with the internal Intel SSD 3=

20 as

quoted
quoted
quoted
quoted
quoted
quoted
quoted
well. The other time it worked.
=20
Well maybe that is due to BTRFS doing something else on =

my T23

quoted

now:
=20
deepdance:~>  ps aux | grep ino-cache | grep -v grep
root      1992  5.5  0.0      0     0 ?        D    12:1=

quoted
quoted
quoted
quoted
quoted
quoted
quoted
0:09 [btrfs- ino-cache]
=20
[=E2=80=A6]
=20
quoted
quoted
quoted
quoted
quoted
At least it doesn=C2=B4t lock up hard, so there might re=

ally be

quoted
quoted
quoted
quoted
quoted
quoted
quoted
something strange with /.
=20
FWIW a btrfs filesystem balance / does work. After this a=

 btrfs

quoted

scrub start / still locks the kernel.

=20
Hi Martin,
=20
I just sent 2 patches to the list. Could you please test if=

 these

quoted

fix your problem with scrub?

=20
I didn=C2=B4t yet test it but I tried the first balance then =

scrub stuff

quoted

=20

quoted

again:

Looks like you're on a 32 bit machine.  The current for-linus b=

ranch

quoted

has an important fix for scrub on 32 bit that should solve this=

=2E

quoted

=20
So finally - the machine did make-kpkg over night and complained =

about

quoted

missing Documentation lguest, then I switched off lots from distr=

quoted

default config and just did the usual make stuff - I was able to =

scrub

quoted

both partitions on that ThinkPad T23:
=20
deepdance:~> btrfs scrub status /
scrub status for 2bf5b1dc-1d89-4f0d-a561-1a5551a27275
=20
        scrub started at Fri Mar 16 11:56:12 2012 and finished af=

ter

quoted

=20
741 seconds total bytes scrubbed: 9.62GB with 0 errors
=20
deepdance:~> btrfs scrub status /home
scrub status for a600de65-e1ab-4cbf-b150-bbaeaf9fa98d
=20
        scrub started at Fri Mar 16 12:00:31 2012 and finished af=

ter

quoted

=20
1708 seconds total bytes scrubbed: 36.63GB with 0 errors
=20
Thanks a lot for fixing this.
=20
Tested-by: Martin Steigerwald <redacted>

=20
[=E2=80=A6]
=20
Would that patch
=20
Btrfs: fix casting error in scrub reada code
=20
http://git.kernel.org/?p=3Dlinux/kernel/git/mason/linux-
btrfs.git;a=3Dcommit;h=3Da175423c831ea582c06784d1e172d2ce1d79923a
=20
(sorry for line break. KMail insists on it even when I disable line
breaks due to the minus sign in the URL.)
=20
that I tested above be something for stable?

=20
<formletter>
=20
This is not the correct way to submit patches for inclusion in the
stable kernel tree.  Please read Documentation/stable_kernel_rules.tx=

for how to do this properly.
=20
</formletter>

What point of the rules do you refer to? Is it

 - It or an equivalent fix must already exist in Linus' tree (upstream)=
=2E

?

Or do you want the patch quoted in the mail?

Anyways its not yet in linus tree AFAIR and from my side it was a quest=
ion=20
whether it should be included at all. I thought it might be good ccing =
you on=20
that question already.

Chris, how about this patch for stable once it hits Linus' tree?

Thanks,
--=20
Martin Steigerwald - teamix GmbH - http://www.teamix.de
gpg: 19E3 8D42 896F D004 08AC A0CA 1E10 C593 0399 AE90
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" =
in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

`h`	back out one level
`j`	next message in thread
`k`	previous message in thread
`l`	drill in
`Esc`	close help / fold thread tree
`?`	toggle this help