Article ID: 126087, created on Jul 5, 2015, last review on Jun 17, 2016

  • Applies to:
  • Virtuozzo


  • After increasing disk space of ploop container, used space indication was also increased, for example:

    # vzctl exec 333 df -h /
    dev/ploop51945p1  178G  695M  168G   1% /
    # vzctl set 333 --diskspace 360G --save
    # vzctl exec 333 df -h /
    /dev/ploop51945p1  355G  702M  338G   1% /


This behavior originates in ext4 files system technology. Ploop is a virtual drive with an individual ext4 file system. ext4 pre-allocates FS metadata so it can manage it much better than a growing one.

The reserved block count can be obtained with dumpe2fs command on ploop device:

[root@pcs ~]# grep 333 /proc/mounts
/dev/ploop15333p1 /vz/root/333 ext4 rw,relatime,barrier=1,data=ordered,balloon_ino=12,pfcache_csum,pfcache=/vz/pfcache 0 0
[root@pcs ~]# dumpe2fs -h /dev/ploop15333p1 | grep "Reserved block count"
Reserved block count:     91698
[root@pcs ~]# vzctl set 333  --save --diskspace 360G --offline
[root@pcs ~]# dumpe2fs -h /dev/ploop15333p1| grep "Reserved block count"
Reserved block count:     104805

In its turn, shrinking of ploop will decrease the overhead:

[root@pcs ~]# vzctl set 333  --save --diskspace 178G --offline
[root@pcs ~]#  dumpe2fs -h /dev/ploop15333p1| grep "Reserved block count"
Reserved block count:     91698

Additinal information:

  • Downsize of a large running container may not be possible, offline resize should be performed.

  • df -h always shows smaller amount of filesystem size than actual disk has due to space reservation for root.

Search Words


container disk space

dh -h

df -h

space consumption


0dd5b9380c7d4884d77587f3eb0fa8ef 2897d76d56d2010f4e3a28f864d69223

Email subscription for changes to this article
Save as PDF