@Ike: Let me know if the above sounds plausible. I should also make the following notes, which may provide evidence counter to my theory of pre-existing corruption:
It is possible that mkfs.ext4 was ran outside of sudo (e.g. directly in a root shell), and would therefore not be logged in /var/log/auth.log. It's also possible that testing was running in an ssh shell, so the commands I gleaned from the console may not be the full picture.
While it is true that the first mount of /dev/sda2 after applying Ted's patch reported "mounting fs with errors", subsequent mounts do not have that message, and yet there are "bad block bitmap checksum" messages that follow:
Jul 9 08:19:42 d05-4 kernel: [ 139.572607] EXT4-fs (sda2): mounted filesystem with ordered data mode. Opts: (null)
Jul 9 09:17:09 d05-4 kernel: [ 3586.211348] EXT4-fs error (device sda2): ext4_validate_block_bitmap:383: comm stress-ng: bg 4705: bad block bitmap checksum
Jul 9 09:37:34 d05-4 kernel: [ 4810.952360] EXT4-fs (sda2): mounted filesystem with ordered data mode. Opts: (null)
Jul 9 10:34:58 d05-4 kernel: [ 8254.776992] EXT4-fs error (device sda2): ext4_validate_block_bitmap:383: comm stress-ng: bg 4193: bad block bitmap checksum
Does that mean the filesystem was somehow no longer corrupted at that time (e.g. fsck'd, or re-mkfs'd), or is it possible that the fs-has-errors flag was just cleared while corruption persisted?
@Ike: Let me know if the above sounds plausible. I should also make the following notes, which may provide evidence counter to my theory of pre-existing corruption:
It is possible that mkfs.ext4 was ran outside of sudo (e.g. directly in a root shell), and would therefore not be logged in /var/log/auth.log. It's also possible that testing was running in an ssh shell, so the commands I gleaned from the console may not be the full picture.
While it is true that the first mount of /dev/sda2 after applying Ted's patch reported "mounting fs with errors", subsequent mounts do not have that message, and yet there are "bad block bitmap checksum" messages that follow:
Jul 9 08:19:42 d05-4 kernel: [ 139.572607] EXT4-fs (sda2): mounted filesystem with ordered data mode. Opts: (null) block_bitmap: 383: comm stress-ng: bg 4705: bad block bitmap checksum block_bitmap: 383: comm stress-ng: bg 4193: bad block bitmap checksum
Jul 9 09:17:09 d05-4 kernel: [ 3586.211348] EXT4-fs error (device sda2): ext4_validate_
Jul 9 09:37:34 d05-4 kernel: [ 4810.952360] EXT4-fs (sda2): mounted filesystem with ordered data mode. Opts: (null)
Jul 9 10:34:58 d05-4 kernel: [ 8254.776992] EXT4-fs error (device sda2): ext4_validate_
Does that mean the filesystem was somehow no longer corrupted at that time (e.g. fsck'd, or re-mkfs'd), or is it possible that the fs-has-errors flag was just cleared while corruption persisted?