btrfs-convert fails

btrfs-convert fails

[Helmut Hullen]

Hallo, linux-btrfs,

I've tried to convert a 12 GByte ext2 partition (nearly full, 280 MByte  
free) with btrfs-convert.

After about 15 minutes (700-MHz-CPU) the system tells

  ...
  creating ext2fs image file
  cleaning up system chunk
  btrfs-convert: extent-tree.c:2529: btrfs_reserve_extent: Assertion
     `!(ret)' failed
  Abgebrochen

"syslogd" tells

Oct 31 17:43:16 ElNath kernel: ------------[ cut here ]------------
Oct 31 17:43:17 ElNath kernel: kernel BUG at fs/btrfs/volumes.c:2831!
Oct 31 17:43:17 ElNath kernel: invalid opcode: 0000 [#1]
Oct 31 17:43:17 ElNath kernel: last sysfs file: /sys/devices/pci0000:00/0000:00:04.0/0000:08:02.0/0000:09:00.0/class
Oct 31 17:43:17 ElNath kernel: Modules linked in: sg nf_nat_ftp nf_conntrack_ftp ipt_MASQUERADE iptable_nat nf_nat xt_DSCP xt_multiport xt_recent nf_conntrack_ipv4 nf_defrag_ipv4 xt_state nf_conntrack xt_tcpudp ipt_REJECT iptable_filter iptable_mangle ip_tables xt_iprange x_tables nfsd exportfs ipv6 8139too 8139cp r8169 savagefb fb_ddc vgastate i2c_piix4 piix e100 mii intel_agp agpgart cmd64x video thermal_sys output ac battery yenta_socket pcmcia_rsrc pcmcia pcmcia_core thinkpad_acpi hwmon led_class nvram fuse
Oct 31 17:43:17 ElNath kernel:
Oct 31 17:43:17 ElNath kernel: Pid: 6396, comm: mount Not tainted 2.6.35.8-h1 #1 26478EG/26478EG
Oct 31 17:43:17 ElNath kernel: EIP: 0060:[<c1231d1a>] EFLAGS: 00010206 CPU: 0
Oct 31 17:43:17 ElNath kernel: EIP is at btrfs_rmap_block+0x3ba/0x3f0
Oct 31 17:43:17 ElNath kernel: EAX: 00000000 EBX: 3df00000 ECX: 00000000 EDX: 00000000
Oct 31 17:43:17 ElNath kernel: ESI: 3df00000 EDI: d7bf76c0 EBP: d1befca8 ESP: d1befc24
Oct 31 17:43:17 ElNath kernel:  DS: 007b ES: 007b FS: 0000 GS: 00e0 SS: 0068
Oct 31 17:43:17 ElNath kernel: Process mount (pid: 6396, ti=d1bee000 task=c23767a0 task.ti=d1bee000)
Oct 31 17:43:17 ElNath kernel: Stack:
Oct 31 17:43:17 ElNath kernel:  00000001 00000000 c10252f8 00000000 00000004 000003fa 00000000 d1befc48
Oct 31 17:43:17 ElNath kernel: <0> c1025383 d1befc80 c1228f3d 00001000 c1b90100 d75bd7e0 00000003 d1befc78
Oct 31 17:43:17 ElNath kernel: <0> 00001000 00000000 00000000 00000000 00000000 00000004 d75bdca8 d1befcb0
Oct 31 17:43:17 ElNath kernel: Call Trace:
Oct 31 17:43:17 ElNath kernel:  [<c10252f8>] ? kmap_atomic_prot+0x48/0xc0
Oct 31 17:43:17 ElNath kernel:  [<c1025383>] ? kmap_atomic+0x13/0x20
Oct 31 17:43:17 ElNath kernel:  [<c1228f3d>] ? map_private_extent_buffer+0x8d/0x120
Oct 31 17:43:17 ElNath kernel:  [<c1229091>] ? map_extent_buffer+0xc1/0xd0
Oct 31 17:43:17 ElNath kernel:  [<c11f13ac>] ? exclude_super_stripes+0xac/0x140
Oct 31 17:43:17 ElNath kernel:  [<c11fa78a>] ? btrfs_read_block_groups+0x4aa/0x700
Oct 31 17:43:17 ElNath kernel:  [<c1205ae5>] ? open_ctree+0xfe5/0x1430
Oct 31 17:43:17 ElNath kernel:  [<c12a0577>] ? strlcpy+0x37/0x60
Oct 31 17:43:17 ElNath kernel:  [<c11e728a>] ? btrfs_get_sb+0x5ca/0x790
Oct 31 17:43:17 ElNath kernel:  [<c10c70e1>] ? dput+0x91/0x120
Oct 31 17:43:17 ElNath kernel:  [<c10bde6e>] ? path_to_nameidata+0x1e/0x50
Oct 31 17:43:17 ElNath kernel:  [<c10cc23d>] ? alloc_vfsmnt+0x6d/0x110
Oct 31 17:43:17 ElNath kernel:  [<c109a093>] ? kstrdup+0x43/0x60
Oct 31 17:43:17 ElNath kernel:  [<c10b7942>] ? vfs_kern_mount+0x72/0x1c0
Oct 31 17:43:17 ElNath kernel:  [<c10ca745>] ? get_fs_type+0x35/0xc0
Oct 31 17:43:17 ElNath kernel:  [<c10b7aee>] ? do_kern_mount+0x3e/0xe0
Oct 31 17:43:17 ElNath kernel:  [<c10cd2a8>] ? do_mount+0x488/0x6d0
Oct 31 17:43:17 ElNath kernel:  [<c1099e72>] ? strndup_user+0x62/0xa0
Oct 31 17:43:17 ElNath kernel:  [<c10cd57c>] ? sys_mount+0x8c/0xb0
Oct 31 17:43:17 ElNath kernel:  [<c17495cc>] ? syscall_call+0x7/0xb
Oct 31 17:43:17 ElNath kernel: Code: 44 8b 55 dc 89 34 da 89 7c da 04 8b 4d d8 83 c3 01 89 5d ec 8b 49 18 89 4d bc e9 70 fd ff ff 89 d0 31 d2 f7 f7 89 d1 89 c6 eb 80 <0f> 0b eb fe 66 90 ba 2c 0b 00 00 b8 a6 a4 87 c1 e8 21 e5 df ff
Oct 31 17:43:17 ElNath kernel: EIP: [<c1231d1a>] btrfs_rmap_block+0x3ba/0x3f0 SS:ESP 0068:d1befc24
Oct 31 17:43:17 ElNath kernel: ---[ end trace f68618b8246ff4aa ]---

What goes wrong?
How much free place needs "btrfs-convert"?

Viele Gruesse!
Helmut

Re: btrfs-convert fails

[Helmut Hullen]

Hallo,

I wrote am 31.10.10 zum Thema btrfs-convert fails:

> I've tried to convert a 12 GByte ext2 partition (nearly full, 280
> MByte free) with btrfs-convert.

> After about 15 minutes (700-MHz-CPU) the system tells

>   ...
>   creating ext2fs image file
>   cleaning up system chunk
>   btrfs-convert: extent-tree.c:2529: btrfs_reserve_extent: Assertion
>      `!(ret)' failed
>   Abgebrochen

Next try: about 11 GByte data, about 690 MByte free: same crash.

> "syslogd" tells

> Oct 31 17:43:16 ElNath kernel: ------------[ cut here ]------------
> Oct 31 17:43:17 ElNath kernel: kernel BUG at fs/btrfs/volumes.c:2831!
> Oct 31 17:43:17 ElNath kernel: invalid opcode: 0000 [#1]
> Oct 31 17:43:17 ElNath kernel: last sysfs file:
> /sys/devices/pci0000:00/0000:00:04.0/0000:08:02.0/0000:09:00.0/class

[etc.]

These syslogd messages come when I try to mount the incompletely  
converted partition.

And then the system needs to be restarted ... mounting doesn't work,  
"halt" or "reboot" doesn't work.

------------------------------------------------

Another experiment: when the partition (about 11.7 GByte) is filled with  
about 9 GByte data, converting works fine.

How much free space needs "btrfs-convert"?

Does it need xx GByte, or does it need yy percent?

Depends this free space on the number of files? I have filled my testing  
partition with *.iso images, 2 to 4 files.

Viele Gruesse!
Helmut

Re: btrfs-convert fails

[Yan, Zheng]

On Sun, Oct 31, 2010 at 11:55 PM, Helmut Hullen <Hullen@t-online.de> wr=
ote:
> Hallo, linux-btrfs,
>
> I've tried to convert a 12 GByte ext2 partition (nearly full, 280 MBy=
te
> free) with btrfs-convert.
>
> After about 15 minutes (700-MHz-CPU) the system tells
>
> =A0...
> =A0creating ext2fs image file
> =A0cleaning up system chunk
> =A0btrfs-convert: extent-tree.c:2529: btrfs_reserve_extent: Assertion
> =A0 =A0 `!(ret)' failed
> =A0Abgebrochen

Try btrfs-convert -r /dev/xxx, hopefully it will recover your ext2.

Yan, Zheng
--
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

Re: btrfs-convert fails

[Helmut Hullen]

Hallo, Yan,

Du meintest am 01.11.10:

>> I've tried to convert a 12 GByte ext2 partition (nearly full, 280
>> MByte free) with btrfs-convert.
>>
>> After about 15 minutes (700-MHz-CPU) the system tells
>>
>> =A0...
>> =A0creating ext2fs image file
>> =A0cleaning up system chunk
>> =A0btrfs-convert: extent-tree.c:2529: btrfs_reserve_extent: Assertio=
n
>> =A0 =A0 `!(ret)' failed
>> =A0Abgebrochen

> Try btrfs-convert -r /dev/xxx, hopefully it will recover your ext2.

Restoring is (now) no problem, I'm still testing. But in some days (or =
=20
weeks) I want to convert a partition with 1,5 TByte (which has no =20
backup) ...

Viele Gruesse!
Helmut
--
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

Re: btrfs-convert fails

[Felix Blanke]

Hi Helmut,

don't use btrfs for important files if you got no backup yet. It's
quite stable and a lot of ppl don't have problem with it, but there are
some cases where it could be very ugly, like a powerloss while writing
etc.


Regards,
=46elix

Am 01 Nov 2010 09:02:00 +0100
schrieb "Helmut Hullen" <Hullen@t-online.de>:

> Hallo, Yan,
>=20
> Du meintest am 01.11.10:
>=20
> >> I've tried to convert a 12 GByte ext2 partition (nearly full, 280
> >> MByte free) with btrfs-convert.
> >>
> >> After about 15 minutes (700-MHz-CPU) the system tells
> >>
> >> =A0...
> >> =A0creating ext2fs image file
> >> =A0cleaning up system chunk
> >> =A0btrfs-convert: extent-tree.c:2529: btrfs_reserve_extent: Assert=
ion
> >> =A0 =A0 `!(ret)' failed
> >> =A0Abgebrochen
>=20
> > Try btrfs-convert -r /dev/xxx, hopefully it will recover your ext2.
>=20
> Restoring is (now) no problem, I'm still testing. But in some days
> (or weeks) I want to convert a partition with 1,5 TByte (which has
> no backup) ...
>=20
> Viele Gruesse!
> Helmut
> --
> 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

--
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

Re: btrfs-convert fails

[Helmut Hullen]

Hallo, Felix,

Du meintest am 01.11.10:

>>>> =A0btrfs-convert: extent-tree.c:2529: btrfs_reserve_extent:
>>>> Assertion =A0 =A0 `!(ret)' failed
>>>> =A0Abgebrochen

>>> Try btrfs-convert -r /dev/xxx, hopefully it will recover your ext2.

>> Restoring is (now) no problem, I'm still testing. But in some days
>> (or weeks) I want to convert a partition with 1,5 TByte (which has
>> no backup) ...

> don't use btrfs for important files if you got no backup yet. It's
> quite stable and a lot of ppl don't have problem with it, but there
> are some cases where it could be very ugly, like a powerloss while
> writing etc.

In these special cases: no big problem. The 1.5 TByte partition contain=
s =20
my videos (at least most of them), and there is a kind of backup on =20
another machine in the family (some kilometers away).

And my machine works behind an UPS. But what about a shutdown initiated=
 =20
by this UPS after (p.e.) 3 minutes powerloss?

Viele Gruesse!
Helmut
--
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

Re: btrfs-convert fails

[cwillu]

On Mon, Nov 1, 2010 at 5:55 AM, Helmut Hullen <Hullen@t-online.de> wrot=
e:
> Hallo, Felix,
>
> Du meintest am 01.11.10:
>
>>>>> =A0btrfs-convert: extent-tree.c:2529: btrfs_reserve_extent:
>>>>> Assertion =A0 =A0 `!(ret)' failed
>>>>> =A0Abgebrochen
>
>>>> Try btrfs-convert -r /dev/xxx, hopefully it will recover your ext2=
=2E
>
>>> Restoring is (now) no problem, I'm still testing. But in some days
>>> (or weeks) I want to convert a partition with 1,5 TByte (which has
>>> no backup) ...
>
>> don't use btrfs for important files if you got no backup yet. It's
>> quite stable and a lot of ppl don't have problem with it, but there
>> are some cases where it could be very ugly, like a powerloss while
>> writing etc.
>
> In these special cases: no big problem. The 1.5 TByte partition conta=
ins
> my videos (at least most of them), and there is a kind of backup on
> another machine in the family (some kilometers away).
>
> And my machine works behind an UPS. But what about a shutdown initiat=
ed
> by this UPS after (p.e.) 3 minutes powerloss?

Problems resulting from powerloss should be limited to faulty hardware
(in which case you're screwed regardless of the filesystem) and lack
of barrier support (which should show up in dmesg soon after the first
write goes to disk).  I'm told that the devs do quite a bit of crash
testing, and I have yet to lose a btrfs filesystem to a powerloss on a
variety of crappy hardware.
--
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

Re: btrfs-convert fails

[Chris Samuel]

On 01/11/10 23:20, cwillu wrote:

> Problems resulting from powerloss should be limited to faulty hardware
> (in which case you're screwed regardless of the filesystem) and lack
> of barrier support (which should show up in dmesg soon after the first
> write goes to disk).

Note that for until 2.6.31 the DM layer didn't support barrier
operations fully, according to this email:

http://www.spinics.net/lists/linux-ext4/msg15789.html

# Barrier infrastructure started to get added to the DM
# core in 2.6.30, see: http://git.kernel.org/linus/af7e466a1ace
#
# But barriers were not enabled for all DM targets (_except_
# dm-multipath) until 2.6.31.  So 2.6.31's dm-crypt does support
# barriers, see: http://git.kernel.org/linus/647c7db14ef9
#
# If the underlying device(s) support barriers DM should
# faithfully pass them on (again except for dm-multipath).
# Also, requests with barriers that result in -EOPNOTSUPP
# are retried without the barrier, see:
# http://git.kernel.org/linus/51aa32284958

cheers,
Chris
-- 
 Chris Samuel  :  http://www.csamuel.org/  :  Melbourne, VIC