commandlinefu.com is the place to record those command-line gems that you return to again and again.
You can sign-in using OpenID credentials, or register a traditional username and password.
Subscribe to the feed for:
Also shows files as they are found. Only works from a tty.
In OSX you would have to make sure that you "sudo -s" your way to happiness since it will give a few "Permission denied" errors before finally spitting out the results. In OSX the directory structure has to start with the "Users" Directory then it will recursively perform the operation.
Your Lord and master,
the -h option of du and sort (on appropriate distrib) makes output "Human" readable and still sorted by "reversed size" (sort -rh)
This command give a human readable result without messing up the sorting.
I had the problem that our monitoring showed that the "/" filesystem is >90% full. This command helped me to find out fast which subdirs are the biggest. The system has many NFS-mounts therefore the -x.
Search for files and list the 20 largest.
find . -type f
gives us a list of file, recursively, starting from here (.)
-print0 | xargs -0 du -h
separate the names of files with NULL characters, so we're not confused by spaces
then xargs run the du command to find their size (in human-readable form -- 64M not 64123456)
| sort -hr
use sort to arrange the list in size order. sort -h knows that 1M is bigger than 9K
| head -20
finally only select the top twenty out of the list
from my bashrc ;)
This one line Perl script will display the smallest to the largest files sizes in all directories on a server.
All folders, human-readable, no subfolder, with a total. Even shorter.
i'm using gawk, you may get varying mileage with other varieties. You might want to change the / after du to say, /home/ or /var or something, otherwise this command might take quite some time to complete. Sorry it's so obsfucated, I had to turn a script into a one-liner under 255 characters for commandlinefu. Note: the bar ratio is relative, so the highest ratio of the total disk, "anchors" the rest of the graph. EDIT: the math was slightly wrong, fixed it. Also, made it compliant with older versions of df.
This command simply outputs 10 files in human readable, that takes most space on your disk in current directory.
In this case I'm just grabbing the next level of subdirectories (and same level regular files) with the --max-depth=1 flag. leaving out that flag will just give you finer resolution. Note that you have to use the -h switch with both 'du' and with 'sort.'
as per eightmillion's comment.
Simply economical :)
6 characters counting whitespace!
Shows the size of the directory the command is ran in.
The size is in MB and GB.
There is no need to type the path, its the current working directory.