Bug 2101045

Did you first run `e2fsck -fvy` before attempting the conversion? Could you run it now and provide the output? And also attach the file created from `debugfs -R show_super_stats /dev/mapper/vg_backup-lv_backup > debugfsshowsuper.txt`

I ran "e2fsck -f -y /dev/mapper/vg_backup-lv_backup" before the conversion attempt, and it didn't report any errors. Below is a run of "e2fsck -f -v -y /dev/mapper/vg_backup-lv_backup" after the attempt.

# e2fsck -f -v -y /dev/mapper/vg_backup-lv_backup
e2fsck 1.46.5 (30-Dec-2021)
Pass 1: Checking inodes, blocks, and sizes
Pass 2: Checking directory structure
Pass 3: Checking directory connectivity
Pass 4: Checking reference counts
Pass 5: Checking group summary information

    12266588 inodes used (1.34%, out of 915660800)
       37031 non-contiguous files (0.3%)
       17584 non-contiguous directories (0.1%)
             # of inodes with ind/dind/tind blocks: 0/0/0
             Extent depth histogram: 12083505/16214/69
  3107368091 blocks used (42.42%, out of 7325267968)
           0 bad blocks
        1391 large files

     4370504 regular files
     7679524 directories
          13 character device files
           0 block device files
          62 fifos
    69886733 links
      216349 symbolic links (166590 fast symbolic links)
         127 sockets
------------
    82153312 files

Created attachment 1892740 [details]
Output of `debugfs -R show_super_stats /dev/mapper/vg_backup-lv_backup > debugfsshowsuper.txt`

>Reported stats:
>  Total space:    12412111552512
>  Free space:     16709885534208 (134.63%)

What is this? There's more free than total? That's not right.

Total is 12T according to this but then from the super

>Block count:              7325267968

27T?

I'm confused, which is it? What's the size of the LV?

I'm seeing feature flag meta_bg on this ext4, but not with default mkfs.ext4 on a 28T loop device. And RAID stride and stripe width are set in the super, consistent with mkfs.ext4 detecting an underlying RAID. Not immediately seeing how this is getting tripped up but it's definitely confused somehow, maybe Qu will have an idea tomorrow.

I get a similar result on a loop mounted 28T ext4 using default mkfs options. 

>  Total space:    12412111552512
>  Free space:     29531792404480 (237.93%)

It does complete without error, however it's also an empty file system.

Some more info on the filesystem.

[root@magneto ~]# df -H /backup
Filesystem                       Size  Used Avail Use% Mounted on
/dev/mapper/vg_backup-lv_backup   30T   13T   16T  44% /backup
[root@magneto ~]# df /backup
Filesystem                        1K-blocks        Used   Available Use% Mounted on
/dev/mapper/vg_backup-lv_backup 29069097456 12197497948 15582365232  44% /backup
[root@magneto ~]# lvdisplay
  --- Logical volume ---
  LV Path                /dev/vg_backup/lv_backup
  LV Name                lv_backup
  VG Name                vg_backup
  LV UUID                dMsxzi-m2eH-0Apt-gWga-BXOk-JWYz-F4Rs60
  LV Write Access        read/write
  LV Creation host, time magneto.discordia, 2019-06-21 15:30:10 -0400
  LV Status              available
  # open                 1
  LV Size                <27.29 TiB
  Current LE             7153582
  Segments               3
  Allocation             inherit
  Read ahead sectors     auto
  - currently set to     12288
  Block device           253:0

[root@magneto ~]# pvdisplay
  --- Physical volume ---
  PV Name               /dev/md0
  VG Name               vg_backup
  PV Size               21.83 TiB / not usable 0
  Allocatable           yes (but full)
  PE Size               4.00 MiB
  Total PE              5722974
  Free PE               0
  Allocated PE          5722974
  PV UUID               Nkh8ZO-djsP-VKJ1-VuS8-KIVO-plVf-ZIwoKU

  --- Physical volume ---
  PV Name               /dev/md1
  VG Name               vg_backup
  PV Size               <5.46 TiB / not usable 5.00 MiB
  Allocatable           yes (but full)
  PE Size               4.00 MiB
  Total PE              1430608
  Free PE               0
  Allocated PE          1430608
  PV UUID               fSr655-OeFV-ORJZ-5fmd-Jbh5-2LQb-zHlR7f

[root@magneto ~]# cat /proc/mdstat
Personalities : [raid6] [raid5] [raid4]
md1 : active raid6 sdi1[1] sdp1[7] sdo1[6] sdn1[3] sdj1[0] sdk1[2] sdl1[4] sdm1[5]
      5859775488 blocks super 1.2 level 6, 512k chunk, algorithm 2 [8/8] [UUUUUUUU]
      bitmap: 0/8 pages [0KB], 65536KB chunk

md0 : active raid6 sdg1[14] sdh1[15] sdf1[13] sde1[12] sdd1[11] sdc1[10] sdb1[9] sda1[8]
      23441307648 blocks super 1.2 level 6, 512k chunk, algorithm 2 [8/8] [UUUUUUUU]
      [=================>...]  check = 87.4% (3417230460/3906884608) finish=131.7min speed=61945K/sec
      bitmap: 0/8 pages [0KB], 262144KB chunk

unused devices: <none>

https://github.com/kdave/btrfs-progs/issues/487

https://github.com/kdave/btrfs-progs/issues/487

With this change....

diff --git a/convert/source-ext2.c b/convert/source-ext2.c
index 9fad4c50..01b630ba 100644
--- a/convert/source-ext2.c
+++ b/convert/source-ext2.c
@@ -92,8 +92,8 @@ static int ext2_open_fs(struct btrfs_convert_context *cctx, const char *name)
 
        cctx->fs_data = ext2_fs;
        cctx->blocksize = ext2_fs->blocksize;
-       cctx->block_count = ext2_fs->super->s_blocks_count;
-       cctx->total_bytes = (u64)ext2_fs->super->s_blocks_count * ext2_fs->blocksize;
+       cctx->block_count = ext2fs_blocks_count(ext2_fs->super);
+       cctx->total_bytes = cctx->block_count * ext2_fs->blocksize;
        cctx->label = strndup((char *)ext2_fs->super->s_volume_name, 16);
        cctx->first_data_block = ext2_fs->super->s_first_data_block;
        cctx->inodes_count = ext2_fs->super->s_inodes_count;

the total space and block count is reported correctly. But the conversion still fails.

[root@magneto ~]# ./btrfs-convert /dev/mapper/vg_backup-lv_backup
btrfs-convert from btrfs-progs v5.18.1

Source filesystem:
  Type:           ext2
  Label:          backup
  Blocksize:      4096
  UUID:           8407aaf9-3225-4b7d-b6fd-af75e97822c7
Target filesystem:
  Label:
  Blocksize:      4096
  Nodesize:       16384
  UUID:           1b2340e7-3b46-4e4c-a702-a36d9abdb909
  Checksum:       crc32c
  Features:       extref, skinny-metadata, no-holes (default)
    Data csum:    yes
    Inline data:  yes
    Copy xattr:   yes
Reported stats:
  Total space:    30004297596928
  Free space:     16711655530496 (55.70%)
  Inode count:       915660800
  Free inodes:       903394212
  Block count:      7325267968
Create initial btrfs filesystem
Create ext2 image file
Create btrfs metadata
convert/source-fs.c:277: record_file_blocks: BUG_ON `cur_off - key.offset >= extent_num_bytes` triggered, value 1
./btrfs-convert[0x414d36]
./btrfs-convert[0x414dba]
./btrfs-convert(record_file_blocks+0x2d6)[0x41598a]
./btrfs-convert(block_iterate_proc+0xfe)[0x4152d4]
./btrfs-convert[0x4166aa]
/lib64/libext2fs.so.2(+0x13c8f)[0x7fe937dddc8f]
/lib64/libext2fs.so.2(+0x1f743)[0x7fe937de9743]
/lib64/libext2fs.so.2(ext2fs_block_iterate2+0x30)[0x7fe937de9b70]
./btrfs-convert[0x41679c]
./btrfs-convert[0x417e2c]
./btrfs-convert[0x4180b0]
./btrfs-convert[0x40d33f]
./btrfs-convert[0x410087]
./btrfs-convert(main+0x78c)[0x41199e]
/lib64/libc.so.6(+0x29550)[0x7fe937b7c550]
/lib64/libc.so.6(__libc_start_main+0x89)[0x7fe937b7c609]
Aborted (core dumped)

Also, is there any ETA on someone taking a look at this? I ask because I can only hold onto the filesystem in its current state for so long. At some point I'll need to put the disk array back in service, and I don't have enough hardware to keep a copy of the filesystem, so I'll lose the ability to reproduce the bug.

I suggest using `e2image -Q` to create a qcow2 image of the ext4 file system metadata (excludes data), if you can save it somewhere in a file sharing service until a dev can get around to looking at it. This will contain directory and file names. If you need that scrubbed you can add -s option, but it might make it less useful to Btrfs developers (not sure).

This message is a reminder that Fedora Linux 36 is nearing its end of life.
Fedora will stop maintaining and issuing updates for Fedora Linux 36 on 2023-05-16.
It is Fedora's policy to close all bug reports from releases that are no longer
maintained. At that time this bug will be closed as EOL if it remains open with a
'version' of '36'.

Package Maintainer: If you wish for this bug to remain open because you
plan to fix it in a currently maintained version, change the 'version' 
to a later Fedora Linux version. Note that the version field may be hidden.
Click the "Show advanced fields" button if you do not see it.

Thank you for reporting this issue and we are sorry that we were not 
able to fix it before Fedora Linux 36 is end of life. If you would still like 
to see this bug fixed and are able to reproduce it against a later version 
of Fedora Linux, you are encouraged to change the 'version' to a later version
prior to this bug being closed.

Fedora Linux 36 entered end-of-life (EOL) status on 2023-05-16.

Fedora Linux 36 is no longer maintained, which means that it
will not receive any further security or bug fix updates. As a result we
are closing this bug.

If you can reproduce this bug against a currently maintained version of Fedora Linux
please feel free to reopen this bug against that version. Note that the version
field may be hidden. Click the "Show advanced fields" button if you do not see
the version field.

If you are unable to reopen this bug, please file a new report against an
active release.

Thank you for reporting this bug and we are sorry it could not be fixed.