kernel BUG

kernel BUG

[Dipl.-Ing. Michael Niederle]

Hi!

I got a kernel BUG at

fs/btrfs/disk-io.c:2311

when testing the 2.6.37-rc5 kernel today.

The system crashed while doing really nothing for a few minutes.

Greetings, Michael

Re: kernel BUG

[Chris Ball]

Hi,

   > Hi!  I got a kernel BUG at
   > 
   > fs/btrfs/disk-io.c:2311
   > 
   > when testing the 2.6.37-rc5 kernel today.

It's good to mention which line this is, since the line numbers are
always changing.  This is:

        if (total_errors > max_errors) {
                printk(KERN_ERR "btrfs: %d errors while writing supers\n",
                       total_errors);
                BUG();
        }
                                                                       
in write_all_supers().  Does dmesg say anything about the nature of
the errors above the BUG() line?

- Chris.
-- 
Chris Ball   <cjb@laptop.org>
One Laptop Per Child

Re: kernel BUG

[Michael Niederle]

Hi!

>    > Hi!  I got a kernel BUG at
>    > 
>    > fs/btrfs/disk-io.c:2311
>    > 
>    > when testing the 2.6.37-rc5 kernel today.  
> 
> It's good to mention which line this is, since the line numbers are
> always changing.  This is:
> 
>         if (total_errors > max_errors) {
>                 printk(KERN_ERR "btrfs: %d errors while writing supers\n",
>                        total_errors);
>                 BUG();
>         }
>                                                                        
> in write_all_supers().  Does dmesg say anything about the nature of
> the errors above the BUG() line?

The message was output to the console of the crashed system - so I'm sorry I
cannot supply further informations.

But exactly the same crash happened again some minutes ago. (I'm now back to my
old kernel.)

Greetings, Michael

Re: kernel BUG

[Chris Ball]

Hi Michael,

   >> in write_all_supers().  Does dmesg say anything about the nature
   >> of the errors above the BUG() line?

   > The message was output to the console of the crashed system - so
   > I'm sorry I cannot supply further informations.

Ah, okay.  You could consider installing kgdb/kdb -- it will let you
drop to the kdb shell after the crash and run dmesg through a pager.

Just:

CONFIG_HAVE_ARCH_KGDB=y
CONFIG_KGDB=y
CONFIG_KGDB_SERIAL_CONSOLE=y
CONFIG_KGDB_TESTS=y
# CONFIG_KGDB_TESTS_ON_BOOT is not set
CONFIG_KGDB_LOW_LEVEL_TRAP=y
CONFIG_KGDB_KDB=y
CONFIG_KDB_KEYBOARD=y

and then boot with "kgdboc=kms,kbd" added.

   > But exactly the same crash happened again some minutes ago. (I'm
   > now back to my old kernel.)

Not sure I understand.  Do you mean that the same crash happens on
your older kernel now, or that you've reverted to your older kernel
because you had another -rc5 crash?  If the crash happens on -rc5 but
not the older kernel (what revision is the older kernel?), that's
extremely interesting.

Josef/Chris, any knowledge about hitting this BUG() on -rc5?

- Chris.
-- 
Chris Ball   <cjb@laptop.org>
One Laptop Per Child

Re: kernel BUG

[Michael Niederle]

Hi, Chris!

>    > But exactly the same crash happened again some minutes ago. (I'm
>    > now back to my old kernel.)  
> 
> Not sure I understand.  Do you mean that the same crash happens on
> your older kernel now, or that you've reverted to your older kernel
> because you had another -rc5 crash?  If the crash happens on -rc5 but
> not the older kernel (what revision is the older kernel?), that's
> extremely interesting.

It happened again on -rc5.

But the reason for the crashes could also be a hardware problem - today I used
(unintentionally) another USB-port (USB 2.0) for the drive containing the
btrfs-file-system then I used (for several month) before (USB 3.0).

Currently I'm running with -rc7. If the system still runs tomorrow morning I
will do another test with -rc5 - this time on the USB 3.0 port. This should show
whether this is a hardware or software problem.

Greetings, Michael

Re: kernel BUG

[Chris Ball]

Hi,

   > Currently I'm running with -rc7. If the system still runs
   > tomorrow morning I will do another test with -rc5 - this time on
   > the USB 3.0 port. This should show whether this is a hardware or
   > software problem.

Thanks, sounds good.  Yes, the location of the BUG() does suggest a
hardware problem, assuming it's coded correctly.  dmesg would tell us
for sure, because the I/O error would be reported above the BUG()
output.

- Chris.
-- 
Chris Ball   <cjb@laptop.org>
One Laptop Per Child

Re: kernel BUG

[Michael Niederle]

Hi!

It seems that in my case the bug (fs/btrfs/disk-io.c:2311) was caused by a
hardware problem.

I plugged by accident an usb-3.0-drive (that worked quite well for several
weeks) into an usb-2.0-port (which worked quite well for several years). It
seems that the two didn't like each other and caused these crashes after some
time of idling (it seems that the idling is necessary for the crash, cause it
never crashed while I was actually working).

Greetings, Michael

Re: kernel BUG

[Christian Thaeter]

Am Sat, 25 Dec 2010 10:50:28 +0100
schrieb "Michael Niederle" <mniederle@gmx.at>:

> Hi!
> 
> It seems that in my case the bug (fs/btrfs/disk-io.c:2311) was caused
> by a hardware problem.
> 
> I plugged by accident an usb-3.0-drive (that worked quite well for
> several weeks) into an usb-2.0-port (which worked quite well for
> several years). It seems that the two didn't like each other and
> caused these crashes after some time of idling (it seems that the
> idling is necessary for the crash, cause it never crashed while I was
> actually working).
> 
> Greetings, Michael

yes looks the same case here too, I only wanted to point
out that this has to be fixed *someday*, Retrying after a bus reset,
mounting the filesystem readonly or block all IO on the filesystem, but
not bring down the kernel.

	Christian

kernel bug

[Roman Kapusta]

Segfault and kernel bug after reflinking several files (20-30 files)
each of size around 1GB through following command:

cp -r --reflink sorce_dir .
Segmentation fault

I'm on Fedora 15, kernel 3.0.4
# uname -a
Linux bruno.localdomain 2.6.40.4-5.fc15.i686.PAE #1 SMP Tue Aug 30 14:43:52 UTC
2011 i686 i686 i386 GNU/Linux

[ 4982.973679] ------------[ cut here ]------------
[ 4982.973691] kernel BUG at fs/btrfs/delayed-inode.c:1693!
[ 4982.973697] invalid opcode: 0000 [#1] SMP
[ 4982.973703] Modules linked in: btrfs zlib_deflate libcrc32c sha256_generic dm
_crypt fuse vboxpci vboxnetadp vboxnetflt vboxdrv 8021q cpufreq_ondemand garp st
p llc powernow_k8 mperf nf_conntrack_netbios_ns nf_conntrack_broadcast ip6t_REJE
CT nf_conntrack_ipv6 nf_defrag_ipv6 nf_conntrack_ipv4 nf_defrag_ipv4 ip6table_fi
lter ip6_tables xt_state nf_conntrack bnep bluetooth w83627ehf hwmon_vid snd_hda
_codec_realtek snd_hda_codec_hdmi snd_hda_intel snd_hda_codec snd_hwdep snd_seq
snd_seq_device snd_pcm snd_timer snd soundcore snd_page_alloc eeepc_wmi asus_wmi
 sparse_keymap rfkill sp5100_tco r8169 mii microcode serio_raw fglrx(P) k10temp
i2c_piix4 i2c_core ipv6 firewire_ohci firewire_core crc_itu_t wmi [last unloaded
: scsi_wait_scan]
[ 4982.973779]
[ 4982.973786] Pid: 10895, comm: cp Tainted: P            2.6.40.4-5.fc15.i686.P
AE #1 System manufacturer System Product Name/E35M1-M
[ 4982.973795] EIP: 0060:[<fa48b15c>] EFLAGS: 00010286 CPU: 0
[ 4982.973847] EIP is at btrfs_delayed_update_inode+0xdf/0xe1 [btrfs]
[ 4982.973853] EAX: ffffffe4 EBX: e3456c98 ECX: 00018000 EDX: 00000000
[ 4982.973858] ESI: eac03c60 EDI: ea7e4800 EBP: e69ebd80 ESP: e69ebd60
[ 4982.973863]  DS: 007b ES: 007b FS: 00d8 GS: 00e0 SS: 0068
[ 4982.973869] Process cp (pid: 10895, ti=e69ea000 task=e8b01940 task.ti=e69ea00
0)
[ 4982.973873] Stack:
[ 4982.973876]  00018000 00000000 00018000 e3456cc4 e34c901c e34c901c ea7e4800 e
a7e4800
[ 4982.973888]  e69ebda4 fa452f21 00000000 e69ebd94 eac03c60 e69ebda4 e34c901c e
a7e4800
[ 4982.973898]  00000000 e69ebe88 fa4751ff a1c00000 0000002f 2bbe9000 00000000 0
0000000
[ 4982.973908] Call Trace:
[ 4982.973954]  [<fa452f21>] btrfs_update_inode+0x40/0x12c [btrfs]
[ 4982.973997]  [<fa4751ff>] btrfs_ioctl_clone+0xa38/0xb27 [btrfs]
[ 4982.974010]  [<fa47761e>] ? btrfs_ioctl_trans_end+0x52/0x52 [btrfs]
[ 4982.974010]  [<fa477f56>] btrfs_ioctl+0x938/0xbda [btrfs]
[ 4982.974010]  [<c05a8f88>] ? inode_has_perm+0x3f/0x46
[ 4982.974010]  [<c05a9031>] ? file_has_perm+0xa2/0xbb
[ 4982.974010]  [<fa47761e>] ? btrfs_ioctl_trans_end+0x52/0x52 [btrfs]
[ 4982.974010]  [<c05009d7>] do_vfs_ioctl+0x428/0x488
[ 4982.974010]  [<c05aa90b>] ? selinux_file_ioctl+0xac/0xb0
[ 4982.974010]  [<c0500a7f>] sys_ioctl+0x48/0x69
[ 4982.974010]  [<c08028df>] sysenter_do_call+0x12/0x28
[ 4982.974010] Code: 10 01 00 00 8b 80 e8 1f 00 00 83 c0 14 e8 39 e3 ff ff 8b 45
 ec e8 77 0c 37 c6 89 d8 e8 5c ea ff ff 31 c0 83 c4 14 5b 5e 5f 5d c3 <0f> 0b 55
 89 e5 53 3e 8d 74 26 00 e8 8a e4 ff ff 85 c0 89 c3 74
[ 4982.974010] EIP: [<fa48b15c>] btrfs_delayed_update_inode+0xdf/0xe1 [btrfs] SS
:ESP 0068:e69ebd60
[ 4982.974411] ---[ end trace 9cf7abc81c235a01 ]---


Thanks

Roman Kapusta

Re: kernel bug

[Roman Kapusta]

I forgot to mention that every process writing to btrfs partition
hanged and cannot be killed by 'kill -9', btrfs filesystem cannot be
unmounted.

Re: kernel bug

[Li Zefan]

Roman Kapusta wrote:
> Segfault and kernel bug after reflinking several files (20-30 files)
> each of size around 1GB through following command:
> 
> cp -r --reflink sorce_dir .
> Segmentation fault
> 
> I'm on Fedora 15, kernel 3.0.4

This has been fixed in Linux 3.1-rc7.