1

Not sure why but the result from two command below are not matched. I am trying to identify which folder use that much disk space. Looks to me only 2.5G is used, but DF says 592G is used.

[root@w10 /]# df -h
Filesystem            Size  Used Avail Use% Mounted on
/dev/mapper/VolGroup00-LogVol00
                      855G  592G  220G  73% /
/dev/mapper/isw_djeiehjhh_Volume0p1
                       99M   19M   75M  21% /boot
tmpfs                  24G     0   24G   0% /dev/shm
[root@w10 /]# du -hs /
du: cannot access `/proc/2819': No such file or directory
du: cannot access `/proc/2820': No such file or directory
du: cannot access `/proc/2821': No such file or directory
du: cannot access `/proc/2822': No such file or directory
2.5G    /

1 Answer 1

4

(possible explanation) On a posix filesystem file may be deleted while it is still in use. When this happens, data has to stay on the disk, until all filedescriptors leading to that file are closed.

You can try doing lsof | grep deleted to see, if you have such file on your system.

Another explanation may be a filesystem corruption, or compromised machine.

Late edit: yet another explanation may be, you're doing it wrong ... with the du -hs / command, you're counting in all files in /dev, and /proc, some of which can get quite large. you should add -x to count files just on one filesystem. so the command looks more like du -xhs /.

3
  • You are right. I have couple of deleted huge files in there. I guess I may need a reboot to clean them up. They are VNCServer log files. Thanks!
    – starchx
    Dec 21, 2011 at 5:31
  • No probs. btw. i just realized, yet another explanation may be, you're doing it wrong ... see the answer
    – Fox
    Dec 22, 2011 at 6:01
  • Thanks for that. It is not my case though. It is definitely some deleted huge files in my case. Thanks.
    – starchx
    Dec 22, 2011 at 23:06

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .