Csdrhrt

I am writing a bash script to install a program for different users.

For that I want to make sure that each user has at least 500Mb available in their $HOME. My $HOME directory looks as follows

jen@ser23:~$ df -h

Filesystem                           Size  Used Avail Use% Mounted on

/dev/sda1                            2,9G  1,1G  1,7G  40% /

udev                                 10M     0   10M   0% /dev

tmpfs                                13G   826M   12G   7% /run

/dev/sda3                            15G   9,8G   4,1G   71% /usr

tmpfs                                32G     0   32G   0% /dev/shm

tmpfs                                5,0M    0   5,0M  0% /run/lock

tmpfs                                32G     0   32G   0% /sys/fs/cgroup

/dev/mapper/local_disk_1-tmp         3,7G   21M  3,7G   1% /tmp

/dev/mapper/local_disk_1-opt         20G   2,0G   18G   10% /opt

/dev/mapper/local_disk_1-project1    401G  287G  114G  72% /project1

/dev/mapper/local_disk_1-var         3,8G  1,7G  1,7G  50% /var

/dev/mapper/local_disk_1-project2    99G   70G   29G   71% /project2

/dev/mapper/local_disk_1-usr_local   2,0G  3,4M  1,9G   1% /usr/local

nfs4.sf0.ise.fhg.de:/g/6/TSB/Archiv  632T  349T  276T  56% /net/p/600/groupdrives/TSB/Archiv

nfs4.sf0.ise.fhg.de:/home            632T  349T  276T  56% /net/home

tmpfs                                6,4G     0  6,4G   0% /run/user/12419

tmpfs                                6,4G  4,0K  6,4G   1% /run/user/13471

tmpfs                                6,4G  4,0K  6,4G   1% /run/user/9351

tmpfs                                6,4G     0  6,4G   0% /run/user/13142

My idea is to use df -h /path/to/home | awk but I am not sure how I can get the actual available space from df -h. Any help please? Thanks, Jen.

jen@ser23:~$ df -P /net/home/j/jen

Filesystem                1024-blocks    Used Available Capacity Mounted on

nfs4.sf0.dfd.fhg.de:/home     5242880 1026048   4216832      20% /net/home

The important point to remember is that df operates on filesystems, which may be attached to particular folders, and if you specify a path or file, it will resolve to usage of the filesystem on which file/folder resides. So df -P /net/home/j/jen operates on the filesystem mounted at /net/home, which is nfs4.sf0.ise.fhg.de network attached storage apparently.

The usage of a directory and all the files requires a recursive solution that will traverse files and directories within particular directory. The tool that you seek then, is du and in particular du -shx /net/home/user.

Keypoints (for more read man du):

• 
  du is recursive by default


• 
  -s provides summary instead of listing filesizes individually


• 
  -h provides human readable output. If you require further processing on data, -b might be more preferable.


• 
  -x is to keep du descending into another. For instance, you could have another network server attached to /net/home/user/anotherplace, so processing that directory is undesirable as it will give incorrect filesystem usage results.

Considering that this is an assignment, further processing and manipulations on output of du are left to the reader to implement

It looks like you're heading towards a workable solution already in the comments, but I'm going to throw this in:

DFHOME=$( df $HOME | awk {'print $4;'} | tail -n 1 )

if [ $DFHOME -lt 500000 ]; then 

  echo "You don't got enough space!";

fi

I'd recommend not using -h because if the size is small enough, the G will turn to an M and your solution might break.

No need to parse df output with awk, you can use --output switch:

df --output=avail "$HOME" | tail -n1

--output[=FIELD_LIST]

use the output format defined by FIELD_LIST, or print all fields if FIELD_LIST is omitted.

FIELD_LIST is a comma-separated list of columns to be included. Valid field names are: 'source', 'fstype', 'itotal', 'iused', 'iavail', 'ipcent', 'size', 'used', 'avail', 'pcent', 'file' and 'tar‐
get'