Daniel Pocock wrote:
> # strace -v -e trace=statfs df -lh /mnt/btrfs0
> statfs("/mnt/btrfs0", {f_type=0x9123683e, f_bsize=4096,
> f_blocks=488374272, f_bfree=475264720, f_bavail=474749786, f_files=0,
> f_ffree=0, f_fsid={2083217090, -1714407264}, f_namelen=255,
> f_frsize=4096}) = 0
> Filesystem                    Size  Used Avail Use% Mounted on
> /dev/mapper/vg00-btrfsvol0_0  1.9T   51G  1.8T   3% /mnt/btrfs0

Thank you for the additional information.  This telle me:

  used: f_blocks - f_bfree
  488374272 - 475264720 = 13109552

  used: 4096 * 13109552 = 53696724992
  used humanly: 53696724992 / (1024*1024) = 51209 (51G)

  bavail: 4096 * 474749786 = 1944575123456
  bfree:  4096 * 475264720 = 1946684293120
  blocks: 4096 * 488374272 = 2000381018112
  blocks human: 2000381018112 / (1024*1024) = 1907712 (1.9T)

Those numbers seem okay to me.  The 'df' command is operating
correctly.  Which means the error lies in the btrfs filesystem layer
and the numbers it is reporting.  Or perhaps the filesystem really is
2T in size and was created as a striped array instead of a mirrored
array?  I would double check that carefully.

Would you like to reassign this bug to the appropriate place in the
kernel?

Bob

Attachment: signature.asc
Description: Digital signature

Reply via email to