I have an AWS linux instance with a root drive of 32GB.
I have an EBS mount of 20GB

On my root I ran out of space, so I cleared out some files. However my root drive is still full. I can't find out why because when I look at sizes of the directories using du and ncdu they show the drive should have a lot of space.


I get the following results

Filesystem     1K-blocks     Used Available Use% Mounted on
/dev/xvda1      32894812 31946092    848472  98% /
devtmpfs         2017224       60   2017164   1% /dev
tmpfs            2025364        0   2025364   0% /dev/shm
/dev/xvdh       20511356  4459276  15003504  23% /mnt/ebs

My /dev/xvda1 is still full

After some research I installed a great tool ncdu to display disk space and the results are:

ncdu 1.10 ~ Use the arrow keys to navigate, press ? for help                                                                                                                                                                                                                                                                                                    
--- / -------------------------------------           
     4.2GiB [##########] /mnt                          
     1.5GiB [###       ] /var
     1.2GiB [##        ] /usr
   684.9MiB [#         ] /opt
   464.3MiB [#         ] /home
   141.7MiB [          ] /lib
    53.5MiB [          ] /boot
    21.2MiB [          ] /lib64
    10.8MiB [          ] /sbin
     8.1MiB [          ] /bin
     7.1MiB [          ] /etc
     2.7MiB [          ] /tmp
    60.0KiB [          ] /dev
    48.0KiB [          ] /root
 e  16.0KiB [          ] /lost+found
 e   4.0KiB [          ] /srv
 e   4.0KiB [          ] /selinux
 e   4.0KiB [          ] /media
 e   4.0KiB [          ] /local
 .   0.0  B [          ] /proc
     0.0  B [          ] /sys
     0.0  B [          ]  .autofsck

If I du -h my total is 8.3G /

So why would my disk be 95% full when it clearly has a lot of space. Am I missing something to do with the mounts and is there any other tool I can run to find out why it is 95% full?


What did you delete? If you remove a file that is still in use by a running process (e.g., a daemon), that disk space is only released when the process is shut down/restarted.

For example, if you removed current Apache log files, the space will still be in use until you restart Apache. Similarly for system logs (those in /var/log).

You can either:

  • Restart the process(es) in question (e.g. Apache, syslogd, etc).
  • Restart your system.

Once you do one of the above, you should see more available space.

| improve this answer | |
  • You are right. After restarting a process that I know I removed some files from the used space has gone to 16% /dev/xvda1 32894812 5133296 27661268 16% / I can see how my process went wrong. I changed the home location of a build server to a different mount and restarted the build server and once I confirmed it was running correctly I deleted the old directory. There must have still been a lock on that directory as the logs still get written there. – Shawn Vader Jul 26 '15 at 11:41

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.