berikut cuplikannya
lakukan pencarian dulu supaya tidak salah pilih
#/media/disk# find ./ -name *.scr
kalo sudah pasti kemudian buang langsung
#/media/disk# find ./ -name *.scr -print -exec rm -r {} -f \;
ini ada ti[s yang lain yang bisa dipelajari
http://www.linux.org/lessons/tips/cmndline.html
Misc.
show libraries a binary uses
Sometimes its necessary to see what libraries a binary is using. For example, if you wanted to know what libraries MySQL is using, just issue this command:
ldd /usr/bin/mysql
substitutes for Netcraft
If you want to find out what operating system a server is using, you can go to Netcraft.com and use their 'What's that site running' service. But there are alternatives. With tools available for Linux, you can do it yourself. It requires curl and/or Lynx to be installed.
lynx -mime_header http://www.domain.com | grep Server
curl -sI http://www.domain.com | grep Server
show files changed on a certain date
If you need to find a file you changed on a certain date, this handy one liner will do it
ls -lt * | grep 'May 8' | awk '{print $9}'
change default editor
The default editor on some Linux machines may not be to your liking. This is particularly important for remote logins. If you wanted to change the editor to 'vim', an improved version of vi, you would do this:
export VISUAL=/usr/bin/vim
sort directories from smallest to biggest
Sometimes you check available space on your hard disk and you're surprised to find that you've recently occupied a lot more. If you're curious as to what's taking up the space, you can find out this way:
du -s -k * | sort -n
strings
If you've got some files that were created by proprietary software that's not longer supported and you need to get data out of them, you might try the 'strings' utility. It will find plain text in binary or other types of non-text formatted files.
strings file > newfile
date in YYYY-MM-DD
This comes in handy for shell scripts, especially if you're making backups. This date format is easily understood by all.
date +%Y-%m-%e
Here's an example of it in use:
tar -zcvpf backup_`date +%Y-%m-%e`.tar.gz *.*
change time stamp of a file
For some reason, you may need to timestamp a file. To simply change the date of a file to the current date and time, do this:
touch file
To change the timestamp to some time in the pass, issue this command:
touch -t YYYYMMDDHHMM file
Where YYYY = year, MM = month, DD = day, HH = hour and MM = minutes.
list all files except
This will list all the files in a directory except the wildcard you specify.
ls -I '*.html'
what's using memory
You may find that your computer is running a bit slower. You can easily find out what's using up your memory:
ps -aux | awk '{print $5,$6,$11}' | sort +1n
complete memory information
To see complete information about the memory your machine is using:
cat /proc/meminfo
change text colors
You can change the color of text in an xterm. This can come in handy it you're writing shell scripts. Try this example:
echo -e "\033[42;1m Pretty colors \033"
Erase the contents of a file
This will erase the contents of a file without eliminating the file
cat /dev/null > some.file
'shred' a file
Just like destroying documents with a paper shredder, computer files with sensitive information in them should also be 'shredded'. To do this, there is a command line utility on Linux systems called 'shred'. What this does is to overwrite the file multiple times with random output. This is secure, because if you simply erase a file with 'rm', all this does is to tell the operating system that this part of your hard disk is free to use, but the information still exists until that space is used. Special tools can be used to recover information from a file that's simply been deleted, but if you've 'shred' it, it would be nearly impossible to get that information back. So shred would be especially useful if you're going to sell or give your computer away to somebody.
The following command will shred a file, "zero" it (to hide shredding) and then remove it.
shred -zuv some.file
You could also 'manually' shred a file by doing the following:
cat /dev/urandom > some.file
This procedure also writes random information to a file. Unlike shred, you'll need to stop this process after a few seconds or some.file will begin to grow until it takes up all available space. When you've finished, simply erase (rm) the file.
delete files with 'bad' names
If you've ever had to delete a file you inadvertantly made, starting with a dash, for example, as you might have found out, it's not as easy as rm -file. You can delete it though - like this:
rm -- -file
or
rm ./-file
pop-up a reminder
You can pop up a message on your Linux desktop with this command:
(sleep 60; xmessage -near One Minute has gone by) &
This will pop up the message 'One Minute has gone by' after 60 seconds. Adjust to your needs accordingly.
Using 'find'
find files bigger than 1mb
This will find files in your home directory that are bigger than 1 megabyte
find /tmp -size +1000k -print
find recently modified files
This will show files that were modified within the last 24 hours in your home directory (if your name is mike)
find ./ -ctime -1 -user mike -print
find with size and access time
You can combine file size and access time with find
find ./ -size +1000k -and -atime +7 -print
Finds files larger than 1 MB that haven't been accessed in more than 7 days.
find and copy into multiple dirs
If you ever need to copy the same file into multiple directories and you can't use a symbolic link, then this trick will work:
find . -type d -name "2004*" -exec cp /file.html {} \;
remove unwanted dirs
Using a modification of the above example, you can remove directories. The following example will remove the temporary directories that the GIMP leaves behind.
find . -type d -name .xvpics -print -exec rm -r {} -f \;
Using 'grep'
Grep is a command line utility that's used mostly to find words in files. It is very powerful. After getting up to speed with it, you'll find that you can't work without it. The basic syntax is:
grep word file, or to use a real example, grep kiwi fruits.txt. This will show the line where the word appears in the file.
You don't have to restrict this to one word. You can search for entire sentences if you like. Just put more than one word inside single quotes.
Show only file name
The following will only show the names of the files where a given word appears
grep -wl 'word' *.*
grep file in a directory
This will search for a given word in an entire directory, including subdirectories
grep -r kiwi ~/my_files/*.rtf
You've got mail!
You can use grep to keep track of recently arrived mail. This one-liner will show you who's sent you mail
grep -c '^From:' /var/spool/mail/bob
The caret '^' tells grep to look for any line beginning with what you specify after.
End of the line
Here's an example of the reverse of the previous example. We can look for lines that end with a particular word or words. The following will show us the users on a Linux system that aren't real people
grep nologin$ /etc/passwd
Accounts for programs and daemons will normally end with 'nologin'.
grep this and this and this
Grep will also let you look for words that appear separately in a file. The following will look for the information for users bob, ted and joe in the /etc/passwd file:
grep "\(bob\|ted\|joe\)" /etc/passwd
Directory Assistance
If you have entered telephone numbers in files, you can use grep to look for them. The following will show you all of the telephone numbers in a given file:
grep '[0-9]\{3\}-[0-9]\{4\}' members.txt
This takes for granted that you've entered them in 000-000-0000 format. Different cultures write down telephone numbers in different ways, so you may have to substitute the the number of digits in the curly braces {2\} instead of {3\} ,for example, or subsitute the separating dash '-' with the character that's more common in your locale (a period '.', for example)
grep and grep
You can use grep more than once if you want to apply different options to two different things that you're looking for. The following pipes the Apache webserver log file to two different instances of grep. We're looking to see who's logged in at a page that are outside of our local network.
cat access | grep login.php | grep -v 192.168
