Dear all,
"btrfs rescue chunk-recover" also crashes. I wonder if having this crash fixed
in the tools is preventing me from restoring the filesystem? Or maybe manually
fixing the data corruption that causes this?
##
Chunk: start = 61070229110784, len = 4294967296, type = 9, num_stripes = 4
Stripes list:
[ 0] Stripe: devid = 0, offset = 0
[ 1] Stripe: devid = 0, offset = 0
[ 2] Stripe: devid = 0, offset = 0
[ 3] Stripe: devid = 0, offset = 0
Block Group: start = 61070229110784, len = 4294967296, flag = 9
Device extent list:
[ 0]Device extent: devid = 4, start = 15122580897792, len =
1073741824, chunk offset = 61070229110784
[ 1]Device extent: devid = 3, start = 15122580897792, len =
1073741824, chunk offset = 61070229110784
[ 2]Device extent: devid = 1, start = 15122635423744, len =
1073741824, chunk offset = 61070229110784
[ 3]Device extent: devid = 2, start = 15122614452224, len =
1073741824, chunk offset = 61070229110784
Unrecoverable Chunks:
Chunk: start = 6329730072576, len = 33554432, type = 2, num_stripes = 0
Stripes list:
Block Group: start = 6329730072576, len = 33554432, flag = 2
No device extent.
Total Chunks: 14126
Recoverable: 14125
Unrecoverable: 1
Orphan Block Groups:
Orphan Device Extents:
No block group[61074524078080, 4294967296]
No block group[61065934143488, 4294967296]
No block group[61070229110784, 4294967296]
No block group[6329730072576, 33554432]
Failed to find [55432747909120, 168, 16384]
btrfs unable to find ref byte nr 55432747909120 parent 0 root 10 owner 1
offset 0
transaction.c:168: btrfs_commit_transaction: BUG_ON `ret` triggered, value -5
btrfs[0x43ebef]
btrfs(btrfs_commit_transaction+0x89)[0x43f07a]
btrfs(btrfs_recover_chunk_tree+0x327a)[0x46a85f]
btrfs[0x465b08]
btrfs(handle_command_group+0x5d)[0x40c7b3]
btrfs(cmd_rescue+0x15)[0x465b98]
btrfs(main+0x24a)[0x40ca02]
/lib64/libc.so.6(__libc_start_main+0xf5)[0x7fc9a26c3b35]
btrfs[0x40c509]
[1] 2750 abort btrfs rescue chunk-recover -y -vv /dev/sdh
###
Sent with ProtonMail Secure Email.
‐‐‐‐‐‐‐ Original Message ‐‐‐‐‐‐‐
On Sunday, 31. March 2019 16:38, berodual_xyz <berodual_...@protonmail.com>
wrote:
> Adding to last email, being more brave to test out commands now:
>
> Different crash than on last report.
>
> Command:
>
> btrfs check --init-extent-tree --mode=lowmem -p /dev/sdh
>
> =========================================================
>
> #########
>
> $ btrfs check --init-extent-tree --mode=lowmem -p /dev/sdh
> WARNING: low-memory mode repair support is only partial
> Opening filesystem to check...
> Checking filesystem on /dev/sdh
> UUID: 8b19ff46-3f42-4f51-be6b-5fc8a7d8f2cd
> Creating a new extent tree
> Failed to find [6815744, 168, 16384]
> btrfs unable to find ref byte nr 55431977238528 parent 0 root 10 owner 2
> offset 0
> Failed to find [6815744, 168, 16384]
> btrfs unable to find ref byte nr 55431977254912 parent 0 root 10 owner 1
> offset 0
> Failed to find [6815744, 168, 16384]
> btrfs unable to find ref byte nr 55431977271296 parent 0 root 10 owner 0
> offset 0
> Failed to find [6815744, 168, 16384]
> btrfs unable to find ref byte nr 55431977287680 parent 0 root 10 owner 1
> offset 0
> Failed to find [6815744, 168, 16384]
> btrfs unable to find ref byte nr 55431977304064 parent 0 root 10 owner 0
> offset 0
> parent transid verify failed on 55431977353216 wanted 60235 found 60236
> Ignoring transid failure
> Failed to find [55431977238528, 168, 16384]
> btrfs unable to find ref byte nr 55431977320448 parent 0 root 1 owner 1
> offset 0
> Failed to find [55431977238528, 168, 16384]
> btrfs unable to find ref byte nr 55431977336832 parent 0 root 1 owner 0
> offset 0
> Failed to find [55431977353216, 168, 16384]
> btrfs unable to find ref byte nr 55431977369600 parent 0 root 1 owner 0
> offset 0
> [1/7] checking root items... skipped
> ERROR: extent[1048576 16384] backref lost (owner: 3, level: 1) root 3
>
> (shortened - same output repeating several times)
>
> --------------------------------------------------
>
> Added an extent item [1048576 16384]
> Added one tree block ref start 1048576 root 3
> ERROR: extent[1064960 16384] backref lost (owner: 3, level: 0) root 3
> Added an extent item [1064960 16384]
> Added one tree block ref start 1064960 root 3
> ERROR: Dev extent's total-byte 15123666173952 is not equal to bytes-used
> 15121518690304 in dev[1, 204, 1]
> ERROR: Dev extent's total-byte 15123687145472 is not equal to bytes-used
> 15121539661824 in dev[1, 204, 2]
> ERROR: Dev extent's total-byte 15123653591040 is not equal to bytes-used
> 15121506107392 in dev[1, 204, 3]
> ERROR: Dev extent's total-byte 15123653591040 is not equal to bytes-used
> 15121506107392 in dev[1, 204, 4]
> ERROR: extent[1163264 16384] backref lost (owner: 3, level: 0) root 3
> Added an extent item [1163264 16384]
> Added one tree block ref start 1163264 root 3
> ERROR: extent[1179648 16384] backref lost (owner: 3, level: 0) root 3
> Added an extent item [1179648 16384]
> Added one tree block ref start 1179648 root 3
> ERROR: extent[1196032 16384] backref lost (owner: 3, level: 0) root 3
> Added an extent item [55431977877504 16384]
> Added one tree block ref start 55431977877504 root 1
> ERROR: extent[55431977861120 16384] backref lost (owner: 10, level: 0) root 1
> Added an extent item [55431977861120 16384]
> Added one tree block ref start 55431977861120 root 1
> ERROR: extent[54142438047744 16384] backref lost (owner: 10, level: 0) root 1
> Added an extent item [54142438047744 16384]
> Added one tree block ref start 54142438047744 root 1
> ERROR: extent [54142438047744 16384] referencer bytenr mismatch, wanted:
> 54142438047744, have: 55432766324736
> Delete backref in extent [54142438047744 16384]
> Failed to find [55432521760768, 168, 16384] (0:00:06 elapsed, 225 items
> checked)
> btrfs unable to find ref byte nr 55432762015744 parent 0 root 10 owner 0
> offset 0
> transaction.c:168: btrfs_commit_transaction: BUG_ON `ret` triggered, value -5
> btrfs[0x43ebef]
> btrfs(btrfs_commit_transaction+0x89)[0x43f07a]
> btrfs[0x47baec]
> btrfs(check_chunks_and_extents_lowmem+0x1a4)[0x47c8b2]
> btrfs(cmd_check+0x1e34)[0x460cd4]
> btrfs(main+0x24a)[0x40ca02]
> /lib64/libc.so.6(__libc_start_main+0xf5)[0x7f6ce4825b35]
> btrfs[0x40c509]
> [1] 1049 abort btrfs check --init-extent-tree --mode=lowmem -p /dev/sdh
>
> ###
>
> And on another run of the same command, yet another crash:
>
> ###
>
> ERROR: extent[34096057876480 16384] backref lost (owner: 1, level: 0) root 1
> Added an extent item [34096057876480 16384]
> Added one tree block ref start 34096057876480 root 1
> ERROR: extent[3543721410560 16384] backref lost (owner: 1, level: 0) root 1
> Added an extent item [3543721410560 16384]
> Added one tree block ref start 3543721410560 root 1
> ERROR: extent[55432521760768 16384] backref lost (owner: 10, level: 0) root 1
> Added an extent item [55432521760768 16384]
> Added one tree block ref start 55432521760768 root 1
> ERROR: extent[55431977877504 16384] backref lost (owner: 10, level: 0) root 1
> Added an extent item [55431977877504 16384]
> Added one tree block ref start 55431977877504 root 1
> check/mode-lowmem.c:4588: walk_down_tree: Warning: assertion `1` failed,
> value 1
> btrfs[0x478fb4]
> btrfs(check_chunks_and_extents_lowmem+0x4b)[0x47c759]
> btrfs(cmd_check+0x1e34)[0x460cd4]
> btrfs(main+0x24a)[0x40ca02]
> /lib64/libc.so.6(__libc_start_main+0xf5)[0x7f348d9a4b35]
> btrfs[0x40c509]
> check/mode-lowmem.c:4588: walk_down_tree: Warning: assertion `1` failed,
> value 1
> btrfs[0x478fb4]
> btrfs(check_chunks_and_extents_lowmem+0x4b)[0x47c759]
> btrfs(cmd_check+0x1e34)[0x460cd4]
> btrfs(main+0x24a)[0x40ca02]
> /lib64/libc.so.6(__libc_start_main+0xf5)[0x7f348d9a4b35]
> btrfs[0x40c509]
> ERROR: extent[55432747909120 16384] backref lost (owner: 10, level: 1) root 1
> Added an extent item [55432747909120 16384]
> Added one tree block ref start 55432747909120 root 1
> ctree.c:1725: leaf_space_used: Warning: assertion `data_len < 0` failed,
> value 1
> btrfs[0x40cb13]
> btrfs(btrfs_leaf_free_space+0x7f)[0x40ee06]
> btrfs[0x474092]
> btrfs[0x47906f]
> btrfs(check_chunks_and_extents_lowmem+0x4b)[0x47c759]
> btrfs(cmd_check+0x1e34)[0x460cd4]
> btrfs(main+0x24a)[0x40ca02]
> /lib64/libc.so.6(__libc_start_main+0xf5)[0x7f348d9a4b35]
> btrfs[0x40c509]
> leaf is not a leaf 55432747909120
> ERROR: extent [55431977877504 16384] referencer bytenr mismatch, wanted:
> 55431977877504, have: 55432766324736
> Delete backref in extent [55431977877504 16384]
> ERROR: extent [55432747909120 16384] referencer bytenr mismatch, wanted:
> 55432747909120, have: 55432766324736
> Delete backref in extent [55432747909120 16384]
> Failed to find [55432521760768, 168, 16384]
> btrfs unable to find ref byte nr 55432746516480 parent 0 root 10 owner 0
> offset 0
> transaction.c:168: btrfs_commit_transaction: BUG_ON `ret` triggered, value -5
> btrfs[0x43ebef]
> btrfs(btrfs_commit_transaction+0x89)[0x43f07a]
> btrfs[0x47baec]
> btrfs(check_chunks_and_extents_lowmem+0x1a4)[0x47c8b2]
> btrfs(cmd_check+0x1e34)[0x460cd4]
> btrfs(main+0x24a)[0x40ca02]
> /lib64/libc.so.6(__libc_start_main+0xf5)[0x7f348d9a4b35]
> btrfs[0x40c509]
> [1] 1652 abort btrfs check --init-extent-tree --mode=lowmem -p /dev/sdh
>
> ‐‐‐‐‐‐‐ Original Message ‐‐‐‐‐‐‐
> On Sunday, March 31, 2019 4:21 PM, berodual_xyz berodual_...@protonmail.com
> wrote:
>
> > Dear list,
> > following from earlier this weeks case of attempting data rescue on a
> > corrupt FS, we have now cloned all devices and can do potentially dangerous
> > rescue attempts (since we can re-clone the original disks).
> > On kernel 4.20.17 and btrfs-progs 4.20.2:
> > --
> > $ btrfs inspect-internal tree-stats /dev/sdh
> > parent transid verify failed on 1048576 wanted 60234 found 60230
> > parent transid verify failed on 1048576 wanted 60234 found 60230
> > Ignoring transid failure
> > volumes.c:1762: btrfs_chunk_readonly: BUG_ON `!ce` triggered, value 1
> > btrfs[0x426fdc]
> > btrfs(btrfs_chunk_readonly+0x98)[0x429acd]
> > btrfs(btrfs_read_block_groups+0x1c1)[0x41cd44]
> > btrfs(btrfs_setup_all_roots+0x368)[0x416540]
> > btrfs[0x416a8a]
> > btrfs(open_ctree_fs_info+0xd0)[0x416bcc]
> > btrfs(open_ctree+0x75)[0x416c61]
> > btrfs(cmd_inspect_tree_stats+0xfa)[0x4706ad]
> > btrfs(handle_command_group+0x5d)[0x40c7b3]
> > btrfs(cmd_inspect+0x15)[0x4499f1]
> > btrfs(main+0x24a)[0x40ca02]
> > /lib64/libc.so.6(__libc_start_main+0xf5)[0x7f0f26c7cb35]
> > btrfs[0x40c509]
> > [1] 15273 abort btrfs inspect-internal tree-stats /dev/sdh
> >
> > ###
> >
> > $ btrfs rescue zero-log /dev/sdh
> > parent transid verify failed on 1048576 wanted 60234 found 60230
> > parent transid verify failed on 1048576 wanted 60234 found 60230
> > Ignoring transid failure
> > volumes.c:1762: btrfs_chunk_readonly: BUG_ON `!ce` triggered, value 1
> > btrfs[0x426fdc]
> > btrfs(btrfs_chunk_readonly+0x98)[0x429acd]
> > btrfs(btrfs_read_block_groups+0x1c1)[0x41cd44]
> > btrfs(btrfs_setup_all_roots+0x368)[0x416540]
> > btrfs[0x416a8a]
> > btrfs(open_ctree_fs_info+0xd0)[0x416bcc]
> > btrfs(open_ctree+0x75)[0x416c61]
> > btrfs[0x4657ff]
> > btrfs(handle_command_group+0x5d)[0x40c7b3]
> > btrfs(cmd_rescue+0x15)[0x465b98]
> > btrfs(main+0x24a)[0x40ca02]
> > /lib64/libc.so.6(__libc_start_main+0xf5)[0x7f4507ab1b35]
> > btrfs[0x40c509]
> > [1] 30211 abort btrfs rescue zero-log /dev/sdh
> >
> > ###
> >
> > Even if we will not be able to recover the data, adding these traces here
> > hopefully helps enhancing the tools in the future as a crash is most
> > definitely not the "expected behavior".
> > Have a nice weekend and I hope to receive some feedback.
> > Thanks in advance,
> > Marcel
