Networking: A Beginner’s Guide Fifth Edition- P73 ppsx

For example, to copy index.html to index-orig.html, type the following: [root@ford /root]# cp index.html index-orig.html The cp command has a large number of options, which are detailed

File Management and Manipulation

This section provides an overview of the basic command-line tools for managing files and directories Most of this overview should be familiar if you have used a command-line interface before Basically, you use the same old functions, just with new commands

cp: Copy Files

The cp command is used to copy files By default, this command works silently, displaying

status information only if there is an error condition For example, to copy index.html to index-orig.html, type the following:

[root@ford /root]# cp index.html index-orig.html

The cp command has a large number of options, which are detailed on its man page The most common options are -f to force copy (do not ask for verification) and -1 for an interactive copy (ask for verification before copying) For example, to copy interactively all files ending in html to the /tmp directory, type the following:

[root@ford /root]# cp -i *.html /tmp

mv: Move Files

Use the mv command to move files from one location to another The command can

move files across partitions as well; however, realize that when moving a file between partitions that you are actually copying the file to the other partition, and then erasing the original, so it can take longer than moving a file within a partition Moving a file within a single partition just tells the system that the file is in a different directory; the file isn’t copied or physically moved on the disk

For example, to move a file from /usr/src/myprog/bin/* to /usr/bin, type the following:

[root@ford /root]# mv /usr/src/myprog/bin/* /usr/bin

The most common options are -f to force a move and -l to move interactively.

Although Linux has no explicit rename tool, you can use mv to accomplish this

task To rename /tmp/blah to /tmp/bleck, type the following:

[root@ford /root]# mv /tmp/bleck /tmp/blah

ln: Link Files

The ln command allows you to establish a hard link or a soft link (See the “About Files

and Directories” section earlier in this chapter for additional information.) The general format of this command is as follows:

[root@ford /root]# ln original_file new_file

The ln command has many options, most of which you’ll never need to use The most common option is -s, which creates a symbolic link instead of a hard link For

example, to create a symbolic link so that /usr/bin/myadduser points to /usr/ local/

bin/myadduser, type the following:

[root@ford /root]# ln -s /usr/local/bin/myadduser /usr/bin/myadduser

find: Find a File

The find command enables you to find files based on a number of criteria The

following is the command’s general format:

[root@ford /root]# find start_dir [options]

where start_dir is the directory from which the search should start

find, like the other commands that we have already discussed, has a large number

of options that you can read about on its man page Table 21-5 list the most common

options used with find.

Option Description

-mount Do not search file systems other than the file system from which you started

-atime n Specify that the file was accessed at least n*24 hours ago.

-ctime n Look only for files changed at least n*24 hours ago.

-inum n Find a file that has i-node n.

-amin n Specify that the file was accessed n minutes ago.

-cmin n Look only for files that were changed n minutes ago.

-empty Find empty files

-mmin n Specify that the file was modified n minutes ago.

-mtime n Search only for files modified n*24 hours ago.

-nouser Find files whose UID does not correspond to a real user in /etc/passwd

-nogroup Look only for files whose GID does not correspond to a real group in

/etc/group

-perm mode Specify that the file’s permissions are exactly set to mode.

-size n[bck] Search only for files at least n blocks/characters/kilobytes big One block

equals 512 bytes

-print Print the filenames found

-exec cmd\; On every file found, execute cmd If you are using the bash shell, be sure to

follow every cmd with a \; otherwise, the shell will become very confused.

-name name Specify that the filename should be name You can use regular expressions

here

Table 21-5. Common find Command Options

For example, to find all files in /tmp that have not been accessed in at least seven days, type the following:

[root@ford /root]# find /tmp -atime 7 -print

To find all files in /usr/src whose names are core and then remove them, type the

following:

[root@ford /root]# find /usr/src -name core -exec rm {} \;

To find all files in /home with the extension jpg that are bigger than 100KB, type the following:

[root@ford /root]# find /home -name "*.jpg" -size 100k

dd: Convert and Copy a File

The dd command reads the contents of a file and sends them to another file What makes dd different from cp is that dd can perform on-the-fly conversions on the file and can accept data from a device (such as a tape or floppy drive) When dd accesses a

device, it does not assume anything about the file system and instead pulls the data in

a raw format Thus, you can use the data to generate images of disks, even if the disk is

of foreign format Table 21-6 lists the most common parameters for dd:

For example, to generate an image of a floppy disk (which is especially useful for foreign file formats), type the following:

[root@ford /root]# dd if=/dev/fd0 of=/tmp/floppy_image

Option Description

if=infile Specify the input file as infile.

of=outfile Specify the output file as outfile.

count=blocks Specify blocks as the number of blocks on which dd should

operate before quitting

ibs=size Set the block size of the input device to be size.

obs=size Set the block size of the output device to be size.

seek=blocks Skip blocks number of blocks on the output.

skip=blocks Skip blocks number of blocks on the input.

swab Convert big endian input to little endian, or vice versa

Table 21-6. Common dd Command Options

gzip: Compress a File

In the original distributions of UNIX, a tool to compress a file was appropriately called

compress Unfortunately, an entrepreneur patented the algorithm, hoping to make

a great deal of money Instead of paying out, most sites sought and found gzip, a

different compression tool with a patent-free algorithm Even better, gzip consistently

achieves better compression ratios than compress does Note that gzip compresses

the file “in place,” meaning that after the compression takes place, the original file is

removed, leaving only the compressed file

TIP You can usually distinguish files compressed with gzip from those compressed by compress by

checking their extensions Files compressed with gzip typically end in gz whereas files compressed

with compress end in Z.

Table 21-7 lists the most-used optional parameters to gzip See the man page for a

complete list

For example, to compress a file and then decompress it, type the following:

[root@ford /root]# gzip myfile

[root@ford /root]# gzip -d myfile.gz

To compress all files ending in html using the best compression possible, enter the

following command:

[root@ford /root]# gzip -9 *.html

NOTE The gzip tool does not share file formats with either PkZip or WinZip However, WinZip can

decompress gzip files

Option Description

-c Write compressed file to the standard output device (thereby allowing

the output to be piped to another program)

-d Decompress

-r Recursively find all files that should be compressed

-9 Provide the best compression

-1 Achieve the fastest compression

Table 21-7. Common gzip Command Options

mknod: Make Special Files

As discussed earlier, Linux accesses all of its devices through files To create a file that the system understands as an interface to a device, you must specify that the file is

of type block or character and has a major and minor number To create this kind of

file with the necessary values, you use the mknod command In addition to creating interfaces to devices, you can use mknod to create named pipes.

The command’s format is as follows:

[root@ford /root]# mknod name type [major] [minor]

where name is the name of the file; and type is the character b for block device, c for character device, or p for named pipe If you choose to create a block or character device, you need to specify the major and minor number

The only time you will need to create a block or character device is when installing some kind of device driver that requires it The documentation that comes with your driver should tell you which values to use for the major and minor numbers

For example, to create a named pipe called /tmp/mypipe, type the following:

[root@ford /root]# mknod /tmp/mypipe p

mkdir: Create a Home Directory

The mkdir command in Linux is identical to the one in other flavors of UNIX as well as

in MS-DOS For example, to create a directory called mydir, type the following:

[root@ford /root]# mkdir mydir

The only option available is -p, which creates a parent directory if none exists For

example, if you need to create /tmp/bigdir/subdir/mydir, and the only directory that

exists is /tmp, using -p will automatically create bigdir and subdir along with mydir.

NOTE Under Linux, you cannot abbreviate the mkdir command as md as you can under DOS.

rmdir: Remove Directory

The rmdir command offers no surprises if you are familiar with the DOS version of the

command It simply removes an existing directory For example, to remove a directory called mydir, type the following:

[root@ford /root]# rmdir mydir

One command-line parameter available for this command is -p, which removes

parent directories as well For example, if the directory /tmp/bigdir/subdir/mydir exists and you want to get rid of all of the directories from bigdir to mydir, issue the following command:

[root@ford /tmp]# rmdir -p bigdir/subdir/mydir