Btrfs devel - Jun 2013 - [3.10-rc6] WARNING: at fs/btrfs/inode.c:7961 btrfs_destroy_inode+0x265/0x2e0 [btrfs]()

Hit this while running this script in a loop..
https://github.com/kernelslacker/io-tests/blob/master/setup.sh

[34385.251507] ------------[ cut here ]------------
[34385.254068] WARNING: at fs/btrfs/inode.c:7961 btrfs_destroy_inode+0x265/0x2e0
[btrfs]()
[34385.257275] Modules linked in: vmw_vsock_vmci_transport vmw_vmci vsock bridge
8021q garp stp mrp raid456 async_raid6_recov async_memcpy async_pq async_xor
async_tx raid1 xfs btrfs xor raid6_pq libcrc32c zlib_deflate tun hidp fuse cmtp
kernelcapi rfcomm bnep l2tp_ppp l2tp_netlink l2tp_core nfnetlink ipt_ULOG
can_raw phonet scsi_transport_iscsi ipx p8023 p8022 irda af_802154 caif_socket
caif crc_ccitt rds af_key appletalk psnap pppoe pppox af_rxrpc ppp_generic slhc
nfc rose netrom atm llc2 ax25 llc can_bcm x25 can raid0 kvm_amd
snd_hda_codec_realtek kvm snd_hda_intel snd_hda_codec microcode snd_pcm pcspkr
serio_raw btusb edac_core snd_page_alloc snd_timer bluetooth snd soundcore
rfkill r8169 mii sr_mod cdrom pata_atiixp radeon backlight drm_kms_helper ttm
[34385.287768] CPU: 3 PID: 1677385 Comm: umount Not tainted 3.10.0-rc6+ #6
[loadavg: 21.16 26.58 27.93 2/170 1677389]
[34385.290479] Hardware name: Gigabyte Technology Co., Ltd.
GA-MA78GM-S2H/GA-MA78GM-S2H, BIOS F12a 04/23/2010
[34385.292974]  ffffffffa05eaffa ffff88011fed3cd8 ffffffff816aee0f
ffff88011fed3d10
[34385.295432]  ffffffff810432b0 ffff8801101172b8 ffff8801101172b8
ffff8800a3cb1148
[34385.297745]  ffffffffa05e47a0 ffff8800bbe1c020 ffff88011fed3d20
ffffffff8104339a
[34385.300664] Call Trace:
[34385.302368]  [<ffffffff816aee0f>] dump_stack+0x19/0x1b
[34385.305202]  [<ffffffff810432b0>] warn_slowpath_common+0x70/0xa0
[34385.308010]  [<ffffffff8104339a>] warn_slowpath_null+0x1a/0x20
[34385.310529]  [<ffffffffa057f7d5>] btrfs_destroy_inode+0x265/0x2e0
[btrfs]
[34385.313020]  [<ffffffff811ccd78>] destroy_inode+0x38/0x60
[34385.315214]  [<ffffffff811cceae>] evict+0x10e/0x1a0
[34385.317198]  [<ffffffff811cd7f5>] iput+0x105/0x190
[34385.319169]  [<ffffffffa055c77f>] btrfs_put_block_group_cache+0x8f/0xc0
[btrfs]
[34385.321464]  [<ffffffffa0569794>] close_ctree+0xe4/0x290 [btrfs]
[34385.323603]  [<ffffffff811cdc8c>] ? evict_inodes+0xfc/0x110
[34385.326237]  [<ffffffffa053eb69>] btrfs_put_super+0x19/0x20 [btrfs]
[34385.328880]  [<ffffffff811b2f71>] generic_shutdown_super+0x61/0xf0
[34385.331527]  [<ffffffff811b31c2>] kill_anon_super+0x12/0x20
[34385.334146]  [<ffffffffa0540dc6>] btrfs_kill_super+0x16/0x90 [btrfs]
[34385.336457]  [<ffffffff811b371d>] deactivate_locked_super+0x4d/0x80
[34385.338666]  [<ffffffff811b3796>] deactivate_super+0x46/0x60
[34385.340782]  [<ffffffff811d3b27>] mntput_no_expire+0xd7/0x130
[34385.342899]  [<ffffffff811d49ec>] SyS_umount+0x9c/0x3b0
[34385.345778]  [<ffffffff816bf3d4>] tracesys+0xdd/0xe2
[34385.348560] ---[ end trace 0d6b7f873e402e88 ]---

--
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

Quoting Dave Jones (2013-06-17 09:49:55)
> Hit this while running this script in a loop..
> https://github.com/kernelslacker/io-tests/blob/master/setup.sh
> [34385.251507] ------------[ cut here ]------------
> [34385.254068] WARNING: at fs/btrfs/inode.c:7961
btrfs_destroy_inode+0x265/0x2e0 [btrfs]()
Thanks Dave, how long did you have to run the script to trigger it?

-chris

On Mon, Jun 17, 2013 at 01:39:42PM -0400, Chris Mason wrote:
 > Quoting Dave Jones (2013-06-17 09:49:55)
 > > Hit this while running this script in a loop..
 > > https://github.com/kernelslacker/io-tests/blob/master/setup.sh
 > > [34385.251507] ------------[ cut here ]------------
 > > [34385.254068] WARNING: at fs/btrfs/inode.c:7961
btrfs_destroy_inode+0x265/0x2e0 [btrfs]()
 > 
 > Thanks Dave, how long did you have to run the script to trigger it?
 > 
 > -chris

Judging by the timestamp, about 9 hours.  This is on a 3 disk (sata)
(oldish) quad opteron. Might repro faster on a more modern machine.

	Dave

Quoting Dave Jones (2013-06-17 14:20:06)
> On Mon, Jun 17, 2013 at 01:39:42PM -0400, Chris Mason wrote:
>  > Quoting Dave Jones (2013-06-17 09:49:55)
>  > > Hit this while running this script in a loop..
>  > > https://github.com/kernelslacker/io-tests/blob/master/setup.sh
>  > > [34385.251507] ------------[ cut here ]------------
>  > > [34385.254068] WARNING: at fs/btrfs/inode.c:7961
btrfs_destroy_inode+0x265/0x2e0 [btrfs]()
>  > 
>  > Thanks Dave, how long did you have to run the script to trigger it?
>  > 
>  > -chris
> 
> Judging by the timestamp, about 9 hours.  This is on a 3 disk (sata)
> (oldish) quad opteron. Might repro faster on a more modern machine.
Exactly how did you run it?  I want to make sure I''m matching your
config.

-chris

On Mon, Jun 17, 2013 at 02:42:27PM -0400, Chris Mason wrote:
 > Quoting Dave Jones (2013-06-17 14:20:06)
 > > On Mon, Jun 17, 2013 at 01:39:42PM -0400, Chris Mason wrote:
 > >  > Quoting Dave Jones (2013-06-17 09:49:55)
 > >  > > Hit this while running this script in a loop..
 > >  > >
https://github.com/kernelslacker/io-tests/blob/master/setup.sh
 > >  > > [34385.251507] ------------[ cut here ]------------
 > >  > > [34385.254068] WARNING: at fs/btrfs/inode.c:7961
btrfs_destroy_inode+0x265/0x2e0 [btrfs]()
 > >  > 
 > >  > Thanks Dave, how long did you have to run the script to trigger
it?
 > >  > 
 > >  > -chris
 > > 
 > > Judging by the timestamp, about 9 hours.  This is on a 3 disk (sata)
 > > (oldish) quad opteron. Might repro faster on a more modern machine.
 > 
 > Exactly how did you run it?  I want to make sure I''m matching
your
 > config.

while [ 1 ];
do
  setup.sh
done

You''ll need to set DISK1 etc variables at the top of the script to
point to
at least 3 disks for it to scribble over.

you''ll also need fsx and fsstress in /usr/local/bin.

Think that''s all.

	Dave

--
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

Quoting Dave Jones (2013-06-17 14:58:10)
> On Mon, Jun 17, 2013 at 02:42:27PM -0400, Chris Mason wrote:
>  > Quoting Dave Jones (2013-06-17 14:20:06)
>  > > On Mon, Jun 17, 2013 at 01:39:42PM -0400, Chris Mason wrote:
>  > >  > Quoting Dave Jones (2013-06-17 09:49:55)
>  > >  > > Hit this while running this script in a loop..
>  > >  > >
https://github.com/kernelslacker/io-tests/blob/master/setup.sh
>  > >  > > [34385.251507] ------------[ cut here ]------------
>  > >  > > [34385.254068] WARNING: at fs/btrfs/inode.c:7961
btrfs_destroy_inode+0x265/0x2e0 [btrfs]()
>  > >  > 
>  > >  > Thanks Dave, how long did you have to run the script to
trigger it?
>  > >  > 
>  > >  > -chris
>  > > 
>  > > Judging by the timestamp, about 9 hours.  This is on a 3 disk
(sata)
>  > > (oldish) quad opteron. Might repro faster on a more modern
machine.
>  > 
>  > Exactly how did you run it?  I want to make sure I''m
matching your
>  > config.
> 
> while [ 1 ];
> do
>   setup.sh
> done
> 
> You''ll need to set DISK1 etc variables at the top of the script to
point to
> at least 3 disks for it to scribble over.
> 
> you''ll also need fsx and fsstress in /usr/local/bin.
I''ve tried with and without memory pressure, and let it run for about
30
hours.  So far, nothing here.  Have you seen this again?

-chris

On Wed, Jun 19, 2013 at 02:02:33PM -0400, Chris Mason wrote:
 > Quoting Dave Jones (2013-06-17 14:58:10)
 > > On Mon, Jun 17, 2013 at 02:42:27PM -0400, Chris Mason wrote:
 > >  > Quoting Dave Jones (2013-06-17 14:20:06)
 > >  > > On Mon, Jun 17, 2013 at 01:39:42PM -0400, Chris Mason
wrote:
 > >  > >  > Quoting Dave Jones (2013-06-17 09:49:55)
 > >  > >  > > Hit this while running this script in a loop..
 > >  > >  > >
https://github.com/kernelslacker/io-tests/blob/master/setup.sh
 > >  > >  > > [34385.251507] ------------[ cut here
]------------
 > >  > >  > > [34385.254068] WARNING: at
fs/btrfs/inode.c:7961 btrfs_destroy_inode+0x265/0x2e0 [btrfs]()
 > >  > >  > 
 > >  > >  > Thanks Dave, how long did you have to run the script
to trigger it?
 > >  > > 
 > >  > > Judging by the timestamp, about 9 hours.  This is on a 3
disk (sata)
 > >  > > (oldish) quad opteron. Might repro faster on a more modern
machine.
 > >  > 
 > >  > Exactly how did you run it?  I want to make sure I''m
matching your
 > >  > config.
 > > 
 > > while [ 1 ];
 > > do
 > >   setup.sh
 > > done
 > > 
 > > You''ll need to set DISK1 etc variables at the top of the
script to point to
 > > at least 3 disks for it to scribble over.
 > > 
 > > you''ll also need fsx and fsstress in /usr/local/bin.
 > 
 > I''ve tried with and without memory pressure, and let it run for
about 30
 > hours.  So far, nothing here.  Have you seen this again?

yeah, one time I hit it within 30 minutes.

	Dave

Quoting Dave Jones (2013-06-19 14:34:50)
> On Wed, Jun 19, 2013 at 02:02:33PM -0400, Chris Mason wrote:
>  > Quoting Dave Jones (2013-06-17 14:58:10)
>  > > On Mon, Jun 17, 2013 at 02:42:27PM -0400, Chris Mason wrote:
>  > >  > Quoting Dave Jones (2013-06-17 14:20:06)
>  > >  > > On Mon, Jun 17, 2013 at 01:39:42PM -0400, Chris Mason
wrote:
>  > >  > >  > Quoting Dave Jones (2013-06-17 09:49:55)
>  > >  > >  > > Hit this while running this script in a
loop..
>  > >  > >  > >
https://github.com/kernelslacker/io-tests/blob/master/setup.sh
>  > >  > >  > > [34385.251507] ------------[ cut here
]------------
>  > >  > >  > > [34385.254068] WARNING: at
fs/btrfs/inode.c:7961 btrfs_destroy_inode+0x265/0x2e0 [btrfs]()
>  > >  > >  > 
>  > >  > >  > Thanks Dave, how long did you have to run the
script to trigger it?
>  > >  > > 
>  > >  > > Judging by the timestamp, about 9 hours.  This is on
a 3 disk (sata)
>  > >  > > (oldish) quad opteron. Might repro faster on a more
modern machine.
>  > >  > 
>  > >  > Exactly how did you run it?  I want to make sure
I''m matching your
>  > >  > config.
>  > > 
>  > > while [ 1 ];
>  > > do
>  > >   setup.sh
>  > > done
>  > > 
>  > > You''ll need to set DISK1 etc variables at the top of
the script to point to
>  > > at least 3 disks for it to scribble over.
>  > > 
>  > > you''ll also need fsx and fsstress in /usr/local/bin.
>  > 
>  > I''ve tried with and without memory pressure, and let it run
for about 30
>  > hours.  So far, nothing here.  Have you seen this again?
> 
> yeah, one time I hit it within 30 minutes.
Ok, I''ll try bumping the thread count on both fsstress and fsx

-chris

--
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