Comment 6 for bug 1857914

Hello
You suggested that I use a file to simulate a partition. I could not highlight the incident.

However, the commands you offer can also be used with a partition. So I no longer need to use gparted.
This highlights that it is the resize command that destroys the partition. Here is the trace below and also in attached file

______________

UNDER UBUNTU1910
.
sudo -i
export LC_ALL=POSIX
mount /dev/sda1 /mnt
ls -ls /mnt
       total 28
        4 drwx------ 20 root root 4096 Jan 5 12:54 P2
        4 drwx------ 20 root root 4096 Jan 5 12:59 P3
        4 drwx------ 20 root root 4096 Jan 5 13:04 P4
       16 drwx------ 2 root root 16384 Jan 5 12:38 lost+found
umount /dev/sda1
e2fsck -f /dev/sda1
          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
          bug_1857914: 11765/3932160 files (0.4% non-contiguous), 11842526/15728640 blocks
resize2fs -p /dev/sda1 50G
          resize2fs 1.45.3 (14-Jul-2019)
          Resizing the filesystem on /dev/sda1 to 13107200 (4k) blocks.
          Begin pass 2 (max = 2580320)
          Relocating blocks XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
          Begin pass 3 (max = 480)
          Scanning inode table XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
          Begin pass 4 (max = 5606)
          Updating inode references XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
          The filesystem on /dev/sda1 is now 13107200 (4k) blocks long.
e2fsck /dev/sda1
          e2fsck 1.45.3 (14-Jul-2019)
          Superblock has an invalid journal (inode 8).
          Clear<y>? cancelled!
          e2fsck: The journal superblock is corrupt while checking journal for bug_1857914
          e2fsck: Cannot proceed with file system check
dumpe2fs -h /dev/sda1
          dumpe2fs 1.45.3 (14-Jul-2019)
          Filesystem volume name: bug 1857914
          ....................
          dumpe2fs: Inode checksum does not match inode while reading journal inode
mount /dev/sda1 /mnt
          mount: /mnt: wrong fs type, bad option, bad superblock on /dev/sda1, missing codepage or helper program, or other error.
dmesg | tail -2
      [ 2319.145382] EXT4-fs error (device sda1): ext4_get_journal_inode:4751: inode #8: comm mount: iget: checksum invalid
      [ 2319.176425] EXT4-fs (sda1): no journal found