Checking disk space in Linux can done with the command line progams df and du. df is used to show usage at the filesystem level, while du can be used to show how much space individual files and directories use.
If you aren’t familiar with how Linux presents storage devices, you will probably want to read the mounting section of: Linux Filesystem Hierarchy.
Checking Disk Usage With df
The command line program df is used to show filesystem usage information. Here is an example:
[root@centos7 ~]# df
Filesystem 1K-blocks Used Available Use% Mounted on
/dev/mapper/centos-root 6981632 2505032 4476600 36% /
devtmpfs 929916 0 929916 0% /dev
tmpfs 941312 0 941312 0% /dev/shm
tmpfs 941312 8664 932648 1% /run
tmpfs 941312 0 941312 0% /sys/fs/cgroup
/dev/sda1 508588 293572 215016 58% /boot
tmpfs 188264 0 188264 0% /run/user/0
With no arguments df displays all mounted filesystems’ usage statistics in 1K blocks. In this context, 1K is 1024.
Common df Options
Here are a few common options for df.
| Option | Description |
|---|---|
| -h | Display the disk usage in human readable sizes using an appropriate power of 1024. |
| -H | Display the disk usage in human readable sizes using an appropriate power of 1000. |
| -k | Display the disk usage in KiB |
| –total | Displays a totals line at the bottom of the output. |
Checking Usage of a Particular Filesystem
If you wish to see how much space is available to a particular file or directory, you can pass it to df as a command line argument. When you do this, df determines which device the file or directory is stored on and displays its usage stats.
[root@centos7 ~]# df -h /boot/grub
Filesystem Size Used Avail Use% Mounted on
/dev/sda1 497M 287M 210M 58% /bootChecking File and Directory Disk Usage Withdu
Sometimes you may want to check how much space a file or directory is using. ls can do it, but du is much easier in many cases. Also, ls won’t show you recursively much space a directory with subdirectories is using.
The example below demonstrates checking the usage of the current directory:
[root@centos7 sysconfig]# pwd
/etc/sysconfig
[root@centos7 sysconfig]# du
8 ./cbq
0 ./console
0 ./modules
224 ./network-scripts
316 .By default, du recursively outputs the disk usage of your current directory and all of its subdirectories in 1K blocks. As with df, du has a -h option that will display the statistics in human-readable powers of 1024:
[root@centos7 sysconfig]# pwd
/etc/sysconfig
[root@centos7 sysconfig]# du -h
8.0K ./cbq
0 ./console
0 ./modules
224K ./network-scripts
316K .Summarizing Output
The output of du can get large. If you only care about the total usage of the current directory, you can suppress the output with -s.
[root@centos7 bin]# pwd
/usr/bin
[root@centos7 usr]# du -hs
1.1G .Specifying Files and Directories
You can specify files and directories by naming them on the command line:
[root@centos7 etc]# du /etc/ssh /etc/sysconfig /etc/hosts
608 /etc/ssh
8 /etc/sysconfig/cbq
0 /etc/sysconfig/console
0 /etc/sysconfig/modules
224 /etc/sysconfig/network-scripts
316 /etc/sysconfig
4 /etc/hostsCombining df and du to Troubleshoot Problems
I commonly use these two tools to find the source of a disk that is filling up. The following example shows a realistic scenario where the yum cache on a CentOS host is filling up the disk.
[root@centos7 /]# df -h
Filesystem Size Used Avail Use% Mounted on
/dev/mapper/centos-root 6.7G 5.4G 1.4G 80% /
devtmpfs 909M 0 909M 0% /dev
tmpfs 920M 0 920M 0% /dev/shm
tmpfs 920M 8.5M 911M 1% /run
tmpfs 920M 0 920M 0% /sys/fs/cgroup
/dev/sda1 497M 287M 210M 58% /boot
tmpfs 184M 0 184M 0% /run/user/0
[root@centos7 /]# cd /
[root@centos7 /]# du -hs *
0 bin
500M blah
262M boot
0 dev
35M etc
68K home
112K ldapdata
0 lib
0 lib64
0 media
0 mnt
21M opt
du: cannot access ‘proc/4986/task/4986/fd/4’: No such file or directory
du: cannot access ‘proc/4986/task/4986/fdinfo/4’: No such file or directory
du: cannot access ‘proc/4986/fd/4’: No such file or directory
du: cannot access ‘proc/4986/fdinfo/4’: No such file or directory
0 proc
589M root
8.5M run
0 sbin
0 srv
0 sys
0 tmp
1.1G usr
3.2G var
[root@centos7 /]# cd var/
[root@centos7 var]# du -hs *
0 adm
3.0G cache
0 crash
8.0K db
0 empty
0 games
0 gopher
0 kerberos
124M lib
0 local
0 lock
4.6M log
0 mail
0 nis
0 opt
0 preserve
0 run
24K spool
0 tmp
0 yp
[root@centos7 var]# cd cache/
[root@centos7 cache]# du -hs *
20K ldconfig
1012K man
3.0G yum
[root@centos7 cache]# cd yum
[root@centos7 yum]# du -hs *
3.0G x86_64
[root@centos7 yum]# cd x86_64/
[root@centos7 x86_64]# du -hs *
3.0G 7
[root@centos7 x86_64]# cd 7
[root@centos7 7]# du -hs *
36M base
8.6M centosplus
1.2M extras
4.0K timedhosts
4.0K timedhosts.txt
3.0G tylersguides
21M updates
[root@centos7 7]# rm -rf tylersguides
[root@centos7 7]# df -h /
Filesystem Size Used Avail Use% Mounted on
/dev/mapper/centos-root 6.7G 2.4G 4.3G 36% /