Linux Btrfs filesystem development
 help / color / mirror / Atom feed
* Unrecoverable filesystem (ERROR: child eb corrupted: parent bytenr=1106952192 item=75 parent level=1 child level=1)
@ 2020-12-11 14:25 Ulrich Windl
  2020-12-15 18:18 ` Zygo Blaxell
  0 siblings, 1 reply; 6+ messages in thread
From: Ulrich Windl @ 2020-12-11 14:25 UTC (permalink / raw)
  To: linux-btrfs

Hi!

While configuring a VM environment in a cluster I had setup an SLES15 SP2 test VM using BtrFS. Due to some problem with libvirt (or the VirtualDomain RA) the VM was active on more than one cluster node at a time, corrupting the filesystem beyond repair it seems:
hvc0:rescue:~ # btrfs check /dev/xvda2
Opening filesystem to check...
Checking filesystem on /dev/xvda2
UUID: 1b651baa-327b-45fe-9512-e7147b24eb49
[1/7] checking root items
ERROR: child eb corrupted: parent bytenr=1107230720 item=75 parent level=1 child level=1
ERROR: failed to repair root items: Input/output error
hvc0:rescue:~ # btrfsck -b /dev/xvda2
Opening filesystem to check...
Checking filesystem on /dev/xvda2
UUID: 1b651baa-327b-45fe-9512-e7147b24eb49
[1/7] checking root items
ERROR: child eb corrupted: parent bytenr=1106952192 item=75 parent level=1 child level=1
ERROR: failed to repair root items: Input/output error
hvc0:rescue:~ # btrfsck --repair /dev/xvda2
enabling repair mode
Opening filesystem to check...
Checking filesystem on /dev/xvda2
UUID: 1b651baa-327b-45fe-9512-e7147b24eb49
[1/7] checking root items
ERROR: child eb corrupted: parent bytenr=1107230720 item=75 parent level=1 child level=1
ERROR: failed to repair root items: Input/output error

Two questions arising:
1) Can't the kernel set some "open flag" early when opening the filesystem, and refuse to open it again (the other VM) when the flag is set? That could avoid such situations I guess
2) Can't btrfs check try somewhat harder to rescue anything, or is the fs structure in a way that everything is lost?

What really puzzles me is this:
There are several snapshots and subvolumes on the BtFS device. It's hard to believe that absolutely nothing seems to be recoverable.

I have this:
hvc0:rescue:~ # btrfs inspect-internal dump-super /dev/xvda2
superblock: bytenr=65536, device=/dev/xvda2
---------------------------------------------------------
csum_type               0 (crc32c)
csum_size               4
csum                    0x659898f3 [match]
bytenr                  65536
flags                   0x1
                        ( WRITTEN )
magic                   _BHRfS_M [match]
fsid                    1b651baa-327b-45fe-9512-e7147b24eb49
metadata_uuid           1b651baa-327b-45fe-9512-e7147b24eb49
label
generation              280
root                    1107214336
sys_array_size          97
chunk_root_generation   35
root_level              0
chunk_root              1048576
chunk_root_level        0
log_root                0
log_root_transid        0
log_root_level          0
total_bytes             10727960576
bytes_used              1461825536
sectorsize              4096
nodesize                16384
leafsize (deprecated)           16384
stripesize              4096
root_dir                6
num_devices             1
compat_flags            0x0
compat_ro_flags         0x0
incompat_flags          0x163
                        ( MIXED_BACKREF |
                          DEFAULT_SUBVOL |
                          BIG_METADATA |
                          EXTENDED_IREF |
                          SKINNY_METADATA )
cache_generation        280
uuid_tree_generation    40
dev_item.uuid           2abdf93e-2f2d-4eef-a1d8-9325f809ebce
dev_item.fsid           1b651baa-327b-45fe-9512-e7147b24eb49 [match]
dev_item.type           0
dev_item.total_bytes    10727960576
dev_item.bytes_used     2436890624
dev_item.io_align       4096
dev_item.io_width       4096
dev_item.sector_size    4096
dev_item.devid          1
dev_item.dev_group      0
dev_item.seek_speed     0
dev_item.bandwidth      0
dev_item.generation     0

Regards,
Ulrich Windl



^ permalink raw reply	[flat|nested] 6+ messages in thread

end of thread, other threads:[~2020-12-18  7:01 UTC | newest]

Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2020-12-11 14:25 Unrecoverable filesystem (ERROR: child eb corrupted: parent bytenr=1106952192 item=75 parent level=1 child level=1) Ulrich Windl
2020-12-15 18:18 ` Zygo Blaxell
2020-12-16 11:46   ` Antw: [EXT] " Ulrich Windl
2020-12-17 13:48   ` Ulrich Windl
2020-12-18  1:51     ` Zygo Blaxell
2020-12-18  7:00       ` Ulrich Windl

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox