Introduction to Linux Command Line Prompt

Restart or shutdown command: shutdown

Shutdown and restart commands: halt/poweroff/reboot

pwd [options]

cd [options] [directory]

When using the cd command, if you use the auto-complete function of the "Tab" key on the keyboard, you can improve the input speed and accuracy. The auto-complete function of the "Tab" key also applies to other commands.

To understand the concept of path, for example, a relative path is a path that does not start from "/" (slash), but starts from the current directory or a specified directory, such as: data/, mnt/oldboy; an absolute path is a path that starts from "/" "(slash) path starting from the root, such as: /data/, /mnt/oldboy.

When you need to switch to the directory where the current user was last located, please use "cd-" (note the space); when you need to switch to the current user's home directory, please use "cd~" (note the space); when you need to switch to When specifying the path to the directory above the current directory, please use "cd.." (note the space).

tree [options] [directory]

mkdir [options] [directory]

The mkdir command can create multiple directories at the same time, in the format mkdir dir1 dir2…

Create directories recursively with the -p parameter (mkdir -p oldboy/test)

Add the -v parameter to display the process of creating the directory. (mkdir -pv oldboy2/test)

You can use the -m parameter when creating a directory to set the default permissions for the directory. (mkdir -m 333 dir2)

Create multiple directories and multi-level subdirectories at the same time. (mkdir -pv oldboy/{dir1_1,dir1_2}/{dir2_1,dir2_2} )

The touch command has two functions: one is to create a new empty file; the other is to change the timestamp attribute of an existing file.

touch [options] [file]

Create file (touch a.txt b.txt)

Change the timestamp attribute of a file

Specify time attributes to create/modify files

ls [options] [<file or directory>]

Use the -a parameter to display all files, especially hidden files

Use the -l parameter to display detailed information.

Parameter --time-style=long-iso that displays the complete time attribute.

Execute the ls command with parameters that display the access time attribute of the content.

Execute the ls command with the -F parameter (this is very similar to the -F of the tree command).

Use the -d parameter to only display information about the directory itself.

Use the -R parameter to recursively view directories.

Knowledge about ls command aliases and setting ls aliases.

Find recently updated files.

Back up the production scenario database and obtain a list of database names.

Delete garbage occupying inode nodes in production scenarios.

Extended knowledge of ls-F command

Interpretation of attributes of ls command output content

cp [options] [source file] [destination file]

cp -a file1.txt file5.txt #<==Use the -a parameter to copy file1.txt to file5.txt. Properties copied using the -a parameter are unchanged.

cp -i file1.txt file5.txt #<== Use the -i parameter to copy the file. You will be prompted whether to overwrite the file.

cp -r dir1 dir2/ #<==If you use the -r parameter, copy the directory recursively and copy all subdirectories and folders under the directory.

cp /etc/ssh/sshd_config{,.ori}#<==The principle of this method is bash's expansion operation of curly braces. /etc/ssh/sshd_config{,.ori} expands into /etc/ssh/sshd_config / etc/ssh/sshd_config.ori and then pass it to the cp command.

mv [options] [source file] [destination file]

mv file6.txt file7.txt #<==If file7.txt does not exist, rename file6.txt to file7.txt.

mv file5.txt file7.txt #<==If file7.txt exists, overwrite file5.txt with file7.txt.

\mv file4.txt file7.txt #<==Use \ to shield the system alias and you will not be asked whether to overwrite.

mv file7.txt dir1/ #<==dir1 is a directory and exists, then move file7.txt to dir1. If dir1 does not exist, rename it to an ordinary file in dir1.

mv -t dir1/ file1.txt file2.txt file3.txt file7.txt #<== Use the -t parameter to swap the source and target, -t followed by the directory, and finally the file to be moved.

rm [options] [<file or directory>]

rm -f file3.txt #<==-f parameter forces deletion without prompting.

rm -r dir1 #<==Use -r to delete recursively, but there will be a confirmation prompt. You can use -f to force it.

Practical experience on deletion

rmdir [options] [directory]

The rmdir command is used to delete empty directories. When the directory is not empty, the command does not work.

rmdir -p -v dir1/a/b/ #<== Recursive deletion still requires listing all directory structures.

ln [options] [source file or directory] [destination file or directory]

hard link

soft link

readlink [options] [file]

readlink /usr/bin/awk #<==You can view the true content of this soft link file.

readlink -f /usr/bin/awk #<==Using the -f parameter will display the last non-symbolic link file.

find [options] [path] [operation statement]

find . -atime -2 #<== "." represents the current directory. Use the option atime to find files accessed within two days. -2 represents within two days.

find search time description

find /var/log/ -mtime 5 -name '*.log' #<==Find files ending with ".log" 5 days ago in the /var/log/ directory.

find . ! -type d #<== "!" means negation, search for files that are not directories, pay attention to the position of the exclamation mark.

find /data/ -perm 755 #<==Find files according to file permissions, 755 is permissions

find . -size 1000c #<==Find files with a file size greater than 1000 bytes in the current directory.

find /data -path "/data/dir3" -prune -o -print #<==The parameter -path specifies the path style, and the -prune parameter is used to exclude the specified directory.

find /data \( -path /data/dir2 -o -path /data/dir3 \) -prune -o -print Use parentheses to combine multiple expressions, but parentheses have special meaning in the command line meaning, so "\" is used to escape here, which tells bash not to parse the following character "()", but to leave it to the find command for processing. Moreover, there is a space between the left bracket and path in "\(-path", and there is a space between dir3 and the right bracket in "dir3\)". This is a grammatical requirement.

find . -user nobody #<==Find files whose user is nobody.

find . -nouser#<==Find files that do not correspond to any user.

find . -group nobody #<==This function is similar to the previous example, here it refers to finding files with the user group nobody.

find . -nogroup#<==Find files that do not correspond to any user group.

find . -newer file1.txt ! -newer file2.txt#<==Find files whose change time is newer than file1.txt but older than file2.txt.

find . -maxdepth 1 -type d #<==-maxdepth 1 finds the first-level directory, similar to tree -L 1.

find . -maxdepth 1 -type d ! -name "." #<==Use the exclamation mark (!) to negate and do not output lines whose names are dots.

find . -maxdepth 1 -type d ! -name "." -o -name "oldboy"#<==-o means or, display all directories or files named oldboy except "." .

find . -maxdepth 1 -type d ! -name "." -a -name "ext"#<==-a means and, it searches for directories that are not dots and whose name is ext. The final result is only displayed Directory named ext.

find . -type f -exec ls -l {} \;#<==The find command matches all ordinary files in the current directory and uses the ls -l command in the -exec option to list them.

find . -type f -mtime 14 -exec rm {} \; #<==find command finds files in the directory that were changed more than 14 days ago, and uses the rm command in the -exec option to delete them.

find /var/log/ -name "*.log" -mtime 5 -ok rm {} \;#<==The find command finds all file names ending with ".log" and the change time in the /var/log/ directory files older than 5 days and delete them. So far, the function of -ok is the same as -exec, but -ok also has a function of giving a prompt before deleting. Press the y key to delete the file, and press the n key not to delete the file, which will be safer.

find . -type f|xargs ls -l #<== Pass the ordinary files found by the find command to the ls command through the pipe symbol and the xargs command for execution. Pay attention to the command format. The pipe symbol "|" is used here. xargs is a command and a filter for passing parameters to other commands. You can read the chapter about the xargs command first before reading this part.

find . -name "*.txt"|xargs -i mv {} dir2/ #<==Use the -i parameter of xargs so that {} represents the files found by find. Put these files as parameters after the mv command as The source file to be moved is moved to the dir2 directory. For more methods, please refer to Section 2.13.3 "Expand Knowledge: Several Methods of Moving Found Files to Specified Locations".

find dir2 -name "file*"|xargs -p rm -f#<== Note: Using the -p option of the xargs command will prompt you to confirm whether to execute the following commands.

xargs [options] The xargs command is a filter that passes command line parameters to other commands. It can convert the data passed by the pipe or standard input into the command line parameters of the command that follows the xargs command.

xargs < test.txt#<== turns all numbers into one line. Note that xargs cannot directly connect to files and needs to be combined with the input redirection character "<".

xargs -n 3 < test.txt#<== Output up to 3 per line.

echo splitXsplitXsplitXsplitX|xargs -d X -n 2 #<==Use X as the delimiter and output up to 2 per line.

find . -name "*.log"|xargs -i mv {} dir1/ #<==Using the -i option of xargs allows {} to replace the file or directory found by the previous find command.

find . -name "file*"|xargs -I [] cp [] dir2 The -I option can specify other characters instead of {}, such as [].

find . -type f -name "*.txt" -print0|xargs -0 rm -f #<==xargs mistakenly thinks that their delimiter is a space. The solution is to separate the output with the character null and use the -0 option.

rename from to file

rename "_finished" "" * #<==Replace _finished of all files with empty.

rename .jpg .oldboy *.jpg #<==Replace .jpg in all files with .oldboy.

basename [<file or directory>] [suffix]

basename /data/dir1/file1.txt #<==Remove the path part, that is, only the file name is displayed.

dirname [<file or directory>

dirname /data/dir1/file1.txt #<==Only displays the path where the file is located. /data/dir1

chattr [options] [mode] [<file or directory>]

chattr a test #<== aAdd additional attributes.

chattr i file1.txt #<==Use the i parameter to lock the file.

lsattr [options] [<file or directory>]

lsattr file1.txt #<==View the default extended attributes of the file.

lsattr -d dir2 #<==Use the -d option to view the extended attributes of the directory.

file [options] [<file or directory>]

file oldboy oldboy: directory #<==oldboy is a directory.

md5sum [options] [file]

md5sum oldboy.txt #<==md5sum command can directly connect the file to get the MD5 value of the file.

chown [options] [user:group] [<file or directory>]

chown oldboy file1.txt #<==Authorize oldboy user, oldboy user must be created in advance.

chown .oldboy file1.txt #<==Authorize the oldboy user group. Be careful not to miss the dot. The dot here can also be replaced by a colon.

chown root:root file1.txt #<==You can use ":" or ".".

chown -R oldboy.oldboy dir2/ #<==Use the -R parameter to authorize recursively.

The chmod command is a command used to change file or directory permissions, but only the owner of the file and the super user root can execute this command.

chmod [options] [mode] [<file or directory>]

chmod a= file1.txt #<==Set all (a) permissions to empty (no characters after the equal sign).

chmod u x file1.txt #<==Set the owner execution permission of the user file.

chmod g w file1.txt #<==Set the writable permissions of the group file user group.

chmod o r file1.txt #<==Set other readable permissions for other users.

chmod ug r,o-r file1.txt #<==Multiple permission operations can be used together, separated by commas. ug r is the abbreviation of u r and g r.

chmod u=rwx,g=rx,o=x file1.txt #<== "=" revoke all original permissions, and then grant the given permissions.

chmod 000 file1.txt #<==This has the same effect as the previous example chmod a= file1.txt.

chmod 753 file1.txt #<==Everyone must be proficient in converting numerical permissions and letter permissions.

chmod -R 777 dir2/ #<==Recursively grant permissions to file directory 777.

chgrp [options] [usergroup] [<file or directory>]

chgrp oldboy install.log #<==The user group that modifies the install.log file is oldboy.

chgrp -R root dir1/ #<==Parameter -R recursive authorization.

umask [options] [mode]

cat [options] [file]

cat >test.txt<<EOF EOF #<== You have to press Enter to end it. In addition, EOF must appear in pairs, but it can also be replaced with other paired tags. For example: oldboy character tag, by default, the ending EOF must be written in top case.

cat -n test.txt#<==Note: As can be seen from the above example, the -n option is to number the file content according to the line and print the output, including blank lines.

cat -b test.txt#<==Note: As can be seen from the above example, the -b option is similar to the -n option, but the -b option does not number empty lines.

cat -E test.txt#<== Note: As can be seen from the above example, the -E option displays the hidden end identifier $ symbol at the end of the file. Even if it is a blank line, there is an end identifier at the end, so everyone should pay attention to this.

cat -s test.txt#<==There were originally three blank lines here. Now, due to the -s option, it becomes one blank line.

cat >test3.txt

tac [options] [file]

more [options] [file]

more /etc/services #<==Does not accept any parameters and displays the file content on the full screen.

more -5 /etc/services #<==At this time, the file content is not displayed on the full screen, only 5 lines of content are displayed.

more 888 /etc/services #<==At this time, the file content is displayed directly from line 888.

ls /etc/|more -10 #<==Everyone should know that there are many file directories under /etc. Direct ls view will display too much content, so you can use the more command to display it in pages.

less [options] [file]

less /etc/services #<==Does not accept any parameters and displays the file content on the full screen.

less -N /etc/services #<==There is a line number in front of each line.

ls /etc/|less #<==View the contents of etc directory files in pages.

head [options] [file]

head /etc/passwd #<==When the head command does not receive any parameters, the first 10 lines of the file will be displayed by default.

head -n 5 /etc/passwd #<==The first format specifies the first 5 lines to be displayed.

head -5 /etc/passwd #<==The second format also specifies the first 5 lines to be displayed, but this way of writing is more streamlined.

head -c 10 /etc/passwd #<==Read the first 10 bytes of the file. The previous writing method is in row units, while -c is in bytes. This feature is not commonly used.

head -n -21 /etc/passwd #<==The number here is a negative value, and this writing method is not commonly used.

tail [options] [file]

tail /etc/passwd #<==The tail command does not take parameters and displays the last 10 lines by default.

tail -n 5 /etc/passwd #<==Display the last 5 lines of text

tail -5 /etc/passwd #<==The second way of writing the last 5 lines of the text is displayed. This way of writing is simpler.

tail -n 15 /etc/passwd #<==Display the file starting from line 15. Extended usage, but not commonly used.

tail -f /application/nginx/logs/access.log #<==tail -f monitors file changes in real time. A common scenario in production is to monitor log files.

tail -f oldboy #<==Use the -f parameter. When the file does not exist, an error will be reported and the command will exit.

tail -F oldboy #<==Use the -F parameter. When the file does not exist, an error will be returned, but it will still wait for the file to be generated and will not exit the command.

tailf [options] [file]

The tailf command is almost equivalent to tail-f. The difference with tail-f is that if the file does not grow, it will not access the disk file and will not change the access time of the file.

tailf /application/nginx/logs/access.log #<== can easily check the changing log file.

cut [option] [file]

cut -b 3 oldboy.txt #<==Only output the 3rd byte.

cut -b 3-5,10 oldboy.txt #<==-b supports writing in the form of 3-5, and multiple positions are separated by commas.

cut -b -3 oldboy.txt #<==-3 means from the first byte to the third byte.

cut -b 3- oldboy.txt #<==3- means from the third byte to the end of the line.

cut -b -3,3- oldboy.txt #<==This way of writing will output the entire line, and there will be no two consecutive overlapping letters a.

cut -b 2-10 oldboy.txt#<==Note: Using the option -c will use characters as the unit and the output will be normal. The option -b will only calculate in bytes (8 binary bits), and the output will be garbled. When encountering multi-byte characters, you can use the -n option. -n is used to tell cut not to split multi-byte characters.

cut -d : -f 1 /etc/passwd #<==The option -d specifies ":" as the separator, and the option -f specifies the first area to be displayed.

split [options] [input file] [output file name prefix]

split -l 10 inittab new_#<== splits every 10 lines, and the split file name starts with new_.

split -l 10 -a 3 inittab new2_#<==The parameter -a specifies the suffix length.

split -l 10 -d inittab num_#<== Parameter -d uses numeric suffix.

split -b 500K -d lvm lvm_#<==Split the file every 500KB.

paste [options] [file]

paste test1 test2 #<==2 files are merged line by line.

paste -d: test1 test2 #<==Use ":" as the separator.

paste -s test1 #<==Use the -s option to convert the content of 1 column into 1 row.

paste -s test1 test2 #<==Each file occupies one line.

sort [options] [file]

sort oldboy.txt #<== does not receive any parameters, will convert the file content into ASCII code, and then compare. Because in ASCII code, the ordering of numbers is the same as our perception, the result is as follows.

sort -n oldboy.txt #<==Use the -n option to sort numbers directly from small to large.

sort -nr oldboy.txt #<== Similar to this function, we have already learned it in the ls command. The -r option means reversal. The sort command sorts by default in ascending order (from small to large). If you use the -r option, it changes to descending order (from large to small).

sort -u oldboy.txt #<==Use the -u option to remove duplicate lines from the file. You will learn a uniq command later, which can also remove duplicate lines.

sort -t " " -k2 oldboy1.txt #<== The -t option specifies a space as the separator, and the -k option followed by 2 means sorting according to the second column.

join [options] [file1] [file2]

join a.txt b.txt#<==Note: The requirement for using join to merge files is that the two files must be sorted by sort.

uniq [options] [file or standard input]

uniq oldboy.txt #<==Remove duplicate lines without taking any parameters.

uniq -c oldboy.txt #<==Parameter -c displays the number of occurrences of the corresponding line.

sort -n oldboy.txt|uniq -c#<==uniq can only deduplicate adjacent duplicate lines, so you should use sort to process the file first and then deduplicate.

wc [options] [file]

wc /etc/inittab 26 149 884 /etc/inittab #<==If you don’t accept any parameters, what does the number displayed mean?

wc -l /etc/inittab #<==Number of lines.

wc -m /etc/inittab #<==Number of characters.

wc -w /etc/inittab #<==Number of words.

wc -L /etc/inittab #<==The length of the longest line.

iconv [options] [original encoding] [new encoding] [input file]

iconv -f gb2312 -t utf-8 GB2312.txt #<== Use the -f parameter to specify the original encoding of the file as gb2312, and use the -t parameter to specify the encoding to be converted to utf-8.

dos2unix [file]

diff [options] [file1] [file2]

diff test1 test2 1,3d0 #<==Delete line 1 to line 3 of file 1, delete line 0 of file 2, that is, do not delete.

6a4,5 #<==Add the following 2 lines of text to line 6 of file 1, namely lines 4 and 5 of text 2.

The default display format of diff has the following three prompts. ·a-add ·c-change ·d-delete

diff -y test1 test2 #<==Use the -y parameter to output side by side.

diff -y -W 30 test1 test2 #<==If you think the above is too wide, you can use the -W parameter to specify the width.

diff /etc/rc3.d/ /etc/rc6.d/ #<==diff can not only compare the differences in file contents, but also compare the differences in files in the directory.

vimdiff [options] [file1] [file2]

Vimdiff calls vim to open files. It can open 2, 3 or 4 files at the same time, up to 4 files, and will use different colors to distinguish the differences between the files.

vimdiff test1 test2 #<==To exit the vimdiff interface, you need to execute the operation of exiting vim twice in succession (:q). The vim command will be explained in detail later. Because the vimdiff command calls the vim function, the exit operation is the same as vim.

rev [file]

echo {1..10}|rev #<==The characters above are written backwards.

rev oldboy.txt #<==You can compare it with the tac command you learned earlier.

tr [options] [char1] [char2]

The tr command replaces, reduces, or deletes characters from standard input and writes the results to standard output.

tr 'abc' 'xyz' < oldboy.txt #<==The tr command to connect files is special and requires the redirection symbol "<".

tr '[a-z]' '[A-Z]' <oldboy.txt #<==Convert lowercase to uppercase.

tr '[0-9]' '[a-j]' < oldboy.txt #<==Replace the number 0 with a, replace 1 with b...one-to-one correspondence.

tr -d 'oldboy'<oldboy.txt #<==Use parameter -d to delete characters.

tr -d ' \t' < oldboy.txt #<== Use the -d parameter to remove all newline characters and tab characters. All lines become one line and the letters are connected together.

echo 'oooolllddbbboyyyyy' |tr -s oldboy #<==Use the -s parameter to compress consecutive characters into one.

tr '0-9' '*' < oldboy.txt #<==Replace all numbers with *.

tr -c '0-9' '*' < oldboy.txt #<==Use parameter -c, except for numbers, other characters including newlines will be replaced with *.

od [options] [file]

tee [options] [file]

ls|tee ls.txt #<==ls command takes over the channel and tee command, outputs the result of ls on the screen, and writes the result to ls.txt.

ls|tee -a ls.txt #<==Use parameter -a to append content to the file without clearing the existing content in the file.

vim [options] [file]

Three modes of vim

Summary of how to open files in vim

helpView built-in command help

Placeholder":"

"." and source

Conditional test "[" and test

Command aliases alias and unalias

Background task related commands bg/fg/jobs

break out of the loop

continue into the next loop

eval executes parameters as commands

exit exits the Shell command line

export View or set global variables

history View command history

read interactive assignment variable

type determines the command type

ulimit modifies system resource usage limits

unset clears variables

lsof [options]

lsof /var/log/messages shows the processes using the file.

lsof -c rsyslog #<==Use the -c option to display files opened by the specified process.

lsof -p 1277 #<==Use the -p option to display files opened by the specified process number.

lsof -i #<==View all processes.

lsof -i tcp #<==Displays process information of all tcp network connections.

lsof -i :22 #<==Display the process with port 22. This command is very commonly used.

lsof -i tcp:22 #<==Display processes that meet both TCP and port 22.

lsof -u oldboy #<==Use the -u option to display files used by the oldboy user.

lsof -U #<==Use the -U option to display all socket files.

uptime

free [option]

free #<==Without parameters, the default display is the number of bytes, which is difficult to understand.

free -m #<==Use the -m option to display memory usage in MB.

free -h #<==Use the -h option to automatically convert into KB, MB, and GB units based on the actual size to display memory usage.

free -h -s 10 #<==Use the -s option to refresh memory usage regularly, in seconds.

wget -O /etc/yum.repos.d/epel.repo http://mirrors.aliyun.com/repo/epel-6.repo yum -y install iftop

iftop [options]

iftop interface: eth0 #<==The first network card of the monitoring system is used by default. You can use the -i option to specify the monitoring network card.

iftop -nNBP ·-n: Do not perform DNS resolution and display the IP numeric address. ·-N: Display the port number in numeric form. ·-P: Display port number. ·-B: The default is to display the traffic in bit units, which needs to be calculated to conform to our understanding, but using the -B option will directly display the traffic in bytes.

vmstat [options] [interval [number of times]]

1) In the vmstat command and subsequent options, there must be at least one space between each element. 2) delay represents the interval between two outputs. 3) count represents the number of statistics according to the time interval specified by delay.

vmstat #<==If the "interval" and "number of times" parameters are omitted, the report will only be displayed once and then exit.

vmstat 5 #<== means that the output information is updated every 5 seconds, the output is looped, and the Ctrl C key combination is pressed to stop the output.

vmstat 5 6 #<== means that the output information is updated every 5 seconds, and the output stops after counting 6 times.

vmstat -a 2 5 displays active and inactive memory.

vmstat -s View memory usage details.

vmstat -d View disk reads/writes.

vmstat -p /dev/sda1 View the read and write statistics of the /dev/sda1 disk.

mpstat [options] [time interval [number of times]]

1) In the mpstat command and subsequent options, there must be at least one space between each element. 2) delay represents the time interval between two outputs. 3) count represents the number of statistics according to the time interval specified by delay.

mpstat #<==If the "time interval" and "number of times" parameters are omitted, the report will only be displayed once and then exit.

mpstat 5 6 #<== means that the output information is updated every 5 seconds, and the output stops after counting 6 times.

mpstat -P 0 #<==Display the information of the first CPU.

iostat [options] [interval [number of times]]

1) In the iostat command and subsequent options, there must be at least one space between each element. 2) interval represents the interval between two outputs. 3) count represents the number of statistics according to the time interval specified by delay.

iostat #<==If the "time interval" and "number of times" parameters are omitted, the report will only be displayed once and then exit.

iostat 2 3 #<==Refresh the display every 2 seconds and display it 3 times in total.

iostat -d #<==Option -d only displays disk statistics.

iostat -d -k #<==Option -k displays data in kB.

iostat -d -m #<==Option -m displays data in MB.

iostat -d -x -k #<==Option -x displays extended information.

iostat -c #<==The -c option is used to only display system CPU statistics.

iotop [options]

iotop

sar [options] [interval [number of times]]

1) In the sar command and subsequent options, there must be at least one space between each element. 2) interval represents the interval between two outputs. 3) count represents the number of statistics according to the time interval specified by interval.

sar -u 2 3 #<==Use the -u option to display the load status of all CPUs in the system during the sampling time, followed by 2 3 indicating statistics once every 2 seconds and 3 times.

sar -q 2 3 #<==Use the -q option to display the size of the run queue.

sar -r 2 3 #<==Use the -r option to display the usage of system memory during the sampling time.

sar -b 2 3 #<==Use the -b option to display the buffer usage during the sampling time.

sar -n DEV 2 3 #<==Use -n DEV to display network interface information.

sar -n EDEV 2 3 #<==Use -n EDEV to display network error statistics.

sar -n SOCK 2 3 #<==Use -n SOCK to display socket information.

sar -d 2 3 #<==Use the -d option to display the usage status of all hard disk devices in the system during the sampling time.

chkconfig [options]

chkconfig --list #<==Use the --list option directly to view the status of all services.

chkconfig --list sshd #<==Specify the system service name to display the startup status of this service.

chkconfig sshd off #<==Use off to turn off the sshd service and start it automatically at level 2, 3, 4, and 5.

chkconfig sshd on #<==Use on to enable the sshd service to start automatically at level 2, 3, 4, and 5.

chkconfig --level 3 sshd off #<==Use --level to specify to turn off the sshd service and start it automatically at level 3.

chkconfig --level 3 sshd on #<==Use --level to specify that the sshd service will start automatically at level 3.

ntsysv [options]

ntsysv

The setup command is a text interface-based system management tool that integrates user authentication management, firewall management, network management and system service management.

setup

ethtool [network card device]

ethtool eth0 #<==Connect the specified network card to display the parameters of the network card.

mii-tool [options] [network interface]

mii-tool eth0 #<==Do not accept any options and display simplified information.

mii-tool -v eth0 #<==Use the -v option to display detailed information.

dmidecode [options]

dmidecode -s system-product-name #<==The author's server model is Dell 2950. PowerEdge 2950

dmidecode -s system-serial-number #<==View the serial number keyword system-serial-number.

dmidecode -t memory #<==-Use the -t option followed by the keyword memory to only view memory information. More keywords can be viewed through the dmidecode -t command.

lspci [options]

lspci -s 02:04.0 #<==02:04.0 From Example 1, we can know the number of the network card device.

lspci -s 02:04.0 -v #<==View detailed information.

ipcs [options]

ipcrm [options]

ipcrm -s 0 #<==Remove the signal set with a specified semid of 0.

rpm [options]

rpm -qpi lrzsz-0.12.20-27.1.el6.x86_64.rpm #<==Displays the rpm package version, creation date and other information.

rpm -qpl lrzsz-0.12.20-27.1.el6.x86_64.rpm #<==View the files in the rpm package.

rpm -qpR lrzsz-0.12.20-27.1.el6.x86_64.rpm #<==View the files that are required to install this rpm package.

rpm -ivh lrzsz-0.12.20-27.1.el6.x86_64.rpm #<==Install the rpm package and use the -h parameter to display the progress bar. #<==rpm also supports online installation, directly connect to a URL address

rpm -qa lrzsz #<==The -p parameter is not used here because lrzsz is the software name and does not end with ".rpm".

rpm -e lrzsz #<== Use the -e parameter to uninstall the software package. This parameter is more dangerous. Generally, try not to uninstall the software package if it is not necessary, because it is very likely that some necessary files for the system will be deleted by mistake. Finally causing system damage.

rpm -qf $(which ifconfig) #<== Sometimes you will find that the system does not have certain files or commands, but you do not know which software package the file or command belongs to. In this case, you can use the -f parameter to query (in Query on the system that has this file). For example, in this example, the ifconfig command belongs to the net-tools software package.

yum [options] [command] [package]

yum install httpd

yum list httpd #<==Check the httpd installation list.

yum search httpd #<==Search for packages containing the httpd string.

yum grouplist #<==View installed and uninstalled package groups.

yum groupinstall "SNMP Support" -y #<==Install package group, search from yum grouplist.

ifconfig [network interface] [options]

interface is the name of the network interface. The network interface names under Linux are similar to eth0, eth1, and lo, etc., which respectively represent the first network card, the second network card, and the loopback interface. This is an optional option. If you do not add this option, all network card information in the system will be displayed. If you add this option, the specified network card information will be displayed.

Using the ifconfig command to configure network card information will only take effect temporarily. Restarting the network or server configuration will invalidate it.

The ifconfig command must be executed as the root user when configuring network card information.

ifconfig #<==Displays all network card information in the system.

Ifconfig eth0 #<== command is connected to the network card name to specify the information of the network card to be displayed.

ifconfig -a #<==Use the -a option to view all network card information.

ifconfig eth1 up #<==Add up after the network card name to start the network card.

ifconfig eth1 down #<==Add down after the network card name to shut down the network card.

ifconfig eth0 192.168.120.56 #<==Connect the IP address directly to the network card that needs to be configured.

ifconfig eth0:0 10.0.0.8 netmask 255.255.255.0 up#<== Syntax format: Alias ​​IP address Subnet mask Activate network card

ifconfig eth0:1 10.0.0.9/24 up #<==Add a second IP alias, 10.0.0.9/24 has the same effect as 10.0.0.9 netmask 255.255.255.0, 24 is the subnet mask 255.255.255.0 Another form of expression.

ifconfig eth0 hw ether 00:AA:BB:CC:DD:EE#<==Keywords to modify the MAC address hw (set MAC address) ether (network device type).

ifup [network interface]

ifup eth0 #<==Activate network interface eth0. Because eth0 has already started running, the prompt is as follows. RTNETLINK answers: File exists

ifup eth1 #<==Activate the network interface eth1. Under normal circumstances, there is no output.

fup eth2 #<==Activate the network interface eth2 and report an error. The configuration file of eth2 cannot be found.

ifdown [network interface]

ifdown eth1 #<==Close the eth1 network card.

route [options]

route #<==By default, the route command will perform DNS resolution on the IP address to generate a host name.

route -n #<==Use the -n option not to perform DNS resolution, which will speed up the display.

route del default #<==Delete gateway method 1.

route add default gw 10.0.0.2 #<==Add gateway method 1, you need to specify the gateway address 10.0.0.2 or other correct address.

route del default gw 10.0.0.2 #<==Delete gateway method 2.

route add default gw 10.0.0.2 dev eth0#<== Add gateway method 2, use dev to specify the network device, suitable for hosts with multiple network devices.

arp [options]

arp #<==Display all entries in the arp cache.

arp -n #<==Use the -n option to display all entries in the arp cache numerically.

arp -n 10.0.0.1 #<==Specify to query the arp information of 10.0.0.1.

arp -s 10.0.0.100 00:0c:29:c0:5a:ef #<==Bind IP address and MAC address.

arp -d 10.0.0.100 #<==Delete static ARP binding.

ip [options] [network object] [operation command]

ip link show dev eth1 #<==Display eth1 network card properties.

ip -s link show dev eth1 #<==Display detailed attributes.

ip -s -s link show dev eth1 #<==Use two -s to display more detailed attributes.

ip link set eth1 up #<==Activate the eth1 network card.

ip link set eth1 down #<==Turn off the eth1 network card.

ip link set eth1 address 0:0c:29:13:10:11 #<==Modify the MAC address.

ip a #<==The effect is the same as ip address. The displayed results include activated and inactive network cards.

ip a add 172.16.1.12/24 dev eth1 #<==Use the add option to add an IP address 172.16.1.12, subnet mask 255.255.255.0, abbreviated as 172.16.1.12/24, and use the dev option to specify the network device as eth1.

ip a add 172.16.1.13/24 dev eth1 #<==You can add multiple IP addresses, which are called auxiliary IPs. The alias created by the previous ifconfig command is IP. Nowadays, commonly used high-availability software such as heartbeat and keepalive all use auxiliary IP.

ip a del 172.16.1.12/24 dev eth1 #<==Delete the main IP, delete the IP address configured by the ip command, and directly change the add option in the previous add command to del.

·Deleting the main IP address of the network card will also delete all IP addresses of the network card. ·Deleting the auxiliary IP address of the network card will not affect other IP addresses of the network card.

ip a add 10.0.0.20/32 dev eth1 label eth1:1 #<== Use the label option to create an alias IP.

ip route|column -t #<==Use the column command to format, option -t, by default, the number of columns in the input row is determined based on space separation to create a table.

ip route add 10.1.0.0/24 via 10.0.0.253 dev eth0 #<==Add static route.

ip route del 10.1.0.0/24 #<==Delete static route.

ip neighbor #<==Use the neighbor command to view the arp cache.

ip neighbor add 192.168.1.100 lladdr 00:0c:29:c0:5a:ef dev eth0 #<==Add static ARP.

ip neighbor del 192.168.1.100 dev eth0 #<==Delete static ARP.

netstat [options]

netstat -an #<== Commonly used combinations -a and -n display all connection information.

netstat -lntup The function of the above command statement is to display all TCP and UDP listening connection information. ·-l: Display all network connections in LISTEN status. ·-n: Display the IP address without DNS resolution into host name and domain name. ·-t: Display all TCP connections. ·-u: Display all UDP connections. ·-p: Display process number and process name.

netstat -rn #<==Use the -r option to display routing table information. The -n option does not perform DNS resolution to speed up command execution.

The ss command is a tool similar to and will replace netstat. It can be used to view network status information, including TCP, UDP connections, ports, etc. Its advantage is that it can display more and more detailed information about the network connection status, and it is faster and more efficient than netstat.

If the system does not have the ss command, you need to install it. The ss command belongs to the iproute package, so the installation command is yum-y install iproute.

ss [options] [filter]

ss -an #<==Display all socket connections.

ss -an|column -t #<==The above output will be a bit messy when written in the document. Let's format it with column.

ss -lntup|column -t #<==Display all listening TCP and UDP connections.

ss -s #<== Count the current number of established, closed, orphaned and waiting TCP sockets.

ping [options] [target host]

ping www.oldboyedu.com #<==The ping command is directly connected to the domain name or IP, and the ping result will always be displayed.

ping -c 3 -i 3 -s 1024 -t 255 www.oldboyedu.com ·-c 3: Send ICMP packets 3 times. ·-i 3: The time interval between each packet sending is 3s. ·-s 1024: Set the sent packet size to 1024 bytes. ·-t 255: Set the ttl value of the sent data packet to 255.

traceroute [options] [hostname or IP] [packet size]

arping [options]

arping -f 10.0.0.1 #<== Use the -f option to exit when the first response is received to detect whether the destination host is alive.

arping -f 10.0.0.3 #<==A host that is not running will not respond, and the command waits until Ctrl C terminates.

telnet [options] [hostname or IP] [port]

telnet 10.0.0.12 22 #<==Replace 10.0.0.12 here with the reader’s own IP, and 22 is the default port number of the SSH service.

nc is a simple, reliable, and powerful network tool that can establish TCP connections, send UDP packets, listen to any TCP and UDP ports, perform port scanning, and process IPv4 and IPv6 packets. If the system does not have the nc command, you can install it manually. The installation command is yum-y install nc.

nc [options]

nc -l 12345 >oldboy.nc #<==Listen to port 12345 and write data to oldboy.nc. #<==After executing the above command, the current window hangs. #<==Open a new window to execute the command.

nc 10.0.0.12 12345 <oldboy.log #<==Use the nc command to transfer the oldboy.log file to the 12345 port of the 10.0.0.12 host.

nc -z 10.0.0.12 20-30 #<==Scan ports 20 to 30 of the 10.0.0.12 host.

nc -z 10.0.0.12 22 #<==A single address or address range can be followed by the host.

nc -z -v 10.0.0.12 20-30 #<==Use the -v option to display the scanning process in detail.

ssh [options] [user@][hostname or IP address] [command executed remotely]

ssh 10.0.0.29 #<==This is the abbreviation command for remote login to the server, which is equivalent to ssh -p 22 root@10.0.0.29 #<==The following four lines will only be prompted when connecting to the remote server for the first time, and will not be prompted when connecting again.

ssh -p 22 oldboy@10.0.0.29 #<== Use the oldboy user to log in to the remote server. This user must be an existing user on the remote server. -p specifies port 22.

ssh 10.0.0.29 "free -m" #<== Just put the command to be executed remotely to the end. Use quotation marks to make it more standardized. Here is the memory information of the server where you are viewing it.

ssh -v root@10.0.0.19 #<==Use the -v option to enter debugging mode.

·Support breakpoint download function. ·Support FTP and HTTP download methods. ·Support proxy server. ·Very stable, it has strong adaptability in situations with very narrow bandwidth or unstable network. If the download fails due to network reasons, wget will continue to try until the entire file is downloaded. If the server interrupts the download process, it will connect to the server again and continue downloading from where it stopped. This is useful for downloading large files from servers with limited connection times.

wget [options] [download address]

wget http://www.oldboyedu.com/favicon.ico #<==wget just connect to the download address.

wget -O /etc/yum.repos.d/epel.repo http://mirrors.aliyun.com/repo/epel-6.repo #<==This is a command to update the epel source, will epel-6. Download the repo and put it in the /etc/yum.repos.d/ directory and rename it epel.repo.

wget --limit-rate=3k http://www.oldboyedu.com/favicon.ico #<==Use the --limit-rate parameter to set the maximum download speed to 3K/s.

wget -c Use the -c parameter to resume the upload from a breakpoint.

wget -b http://www.oldboyedu.com/favicon.ico Use wget-b to download files in the background.

wget --user-agent="Mozilla/5.0 (Windows; U; Windows NT 6.1; en-US) AppleWebKit/534.16 (KHTML, like Gecko) Chrome/10.0.648.204 Safari/534.16" http://www.oldboyedu. com/favicon.ico #<==Use the --user-agent parameter to specify the client type.

wget -q -T 3 --tries=1 --spider www.baidu.com #<== Use silent access, timeout in 3 seconds, retry once, simulate crawler access.

mailq [options]

mailq #<==View the mail queue.

mail [options]

mail -s "Hello from oldboyedu" zhangyao@oldboyedu.com #<==The recipient is requested to change it to his or her own email address. hello,this is the content of mail. #<==Manually enter two lines of mail content. welcome to www.oldboyedu.com. EOT #<== Enter Ctrl D in a new blank line to end the input and send the email.

echo -e "hello,this is the content of mail. welcome to www.oldboyedu.com."|mail -s "Hello from oldboyedu" zhangyao@oldboyedu.com Command description: echo is followed by the email body.

#<==Send a single attachment. [root@oldboy ~]# echo "test"|mail -s "test" -a favicon.jpg zhangyao@oldboyedu.com

#<==Send multiple attachments. [root@oldboy ~]# echo "test"|mail -s "test" -a favicon.jpg -a web.sh zhangyao@oldboyedu.com

nslookup [options] [domain name/IP] [DNS server]

·Interactive mode: Users can query the domain name server for information about various hosts and domain names, or output a list of hosts in the domain name. ·Non-interactive mode: Only obtain specific names or required information for a host or domain name.

nslookup - 223.5.5.5 #<==You can directly specify the server address for domain name resolution on the command line, but pay attention to the writing and do not miss the "-".

nslookup www.oldboyedu.com 223.5.5.5 #<==Non-interactive query www.oldboyedu.com domain name.

dig [option]

dig www.oldboyedu.com

dig @223.5.5.5 www.oldboyedu.com #<==Use @ to specify the DNS server for query.

dig -x 101.200.195.98 #<==Use the -x option for domain name reverse resolution.

dig -t MX oldboyedu.com #<==Use the -t option to select the type of query.

dig @223.5.5.5 www.oldboyedu.com trace #<==Displays the complete process of domain name resolution into IP.

dig nocmd nocomment nostat www.oldboyedu.com #<== Simplify some description information.

dig short www.oldboyedu.com

host [options]

host www.oldboyedu.com #<==host command can be directly connected to the domain name.

host -a www.oldboyedu.com #<==Use the -a option to query detailed information.

host -a www.oldboyedu.com 223.5.5.5 #<== Directly connect to the specified DNS server IP.

host -t MX oldboyedu.com #<==Use the -t option to select the query type.

nmap [scan type] [general options] {scan target}

nmap 10.0.0.12 #<==nmap connects directly to the target host and scans the first 1~1000 ports by default.

nmap -p 1024-65535 10.0.0.12 #<==-p option specifies the scan range.

nmap 10.0.0.0/24 #<==Scan the LAN using the network segment format.

nmap -sn 10.0.0.0/24 #<==Use the -sn option not to scan the port.

nmap -sn 10.0.0.1-10 #<== This address range can be used for scanning.

nmap -O -sV 10.0.0.12 #<==-sV displays the service version number. #<==-O displays the system version, but the nmap command compares the detected TCP/IP fingerprint with its own fingerprint database. If the system is not in the fingerprint database, it will not be recognized.

tcpdump [options] [expression]

tcpdump #<==By default, starting tcpdump directly will monitor all packets flowing on the first network interface.

tcpdump -q #<==By default, the tcpdump command outputs more information. In order to display streamlined information, you can use the -q option.

tcpdump -c 5 #<==Use the -c option to specify the number of packets to monitor, so there is no need to use Ctrl C.

tcpdump -i eth0 #<==Use the -i option to specify the network card to monitor

tcpdump -n host 10.0.0.1 #<==Use the -n option to not perform DNS resolution and speed up the display. The keyword for monitoring the specified host is host, followed by the host name or IP address. The function of this command is to monitor all data packets received and sent by the host 10.0.0.1.

tcpdump -n src host 10.0.0.1 #<==Only listen to data packets sent from 10.0.0.1, that is, the source address is 10.0.0.1, and the keyword is src (source, source address).

tcpdump -n dst host 10.0.0.1 #<==Only listen to the data packets received by 10.0.0.1, that is, the target address is 10.0.0.1, and the keyword is dst (destination, destination).

tcpdump -nn port 22 #<==Using the -n option does not perform DNS resolution, but it will convert some protocols and ports, such as port 22 to ssh. Readers can compare the output results of Example 10-4. Therefore this example uses the -nn option. The keyword for monitoring the specified port is port, followed by the port number.

tcpdump -n arp #<==Listen to ARP packets, so the expression can be written directly into arp.

tcpdump -n icmp #<==Listen to icmp data packets (if you want to view the monitoring data below, you can use other machines to ping this machine)

tcpdump -n ip host 10.0.0.12 and ! 10.0.0.1 #<==Get the ip packets that host 10.0.0.12 communicates with all hosts (except host 10.0.0.1).

ps [options]

ps -ef #<== UNIX format parameters, use the -e parameter to display all processes, use the -f parameter to additionally display the UID, PPID, C and STIME fields.

ps -ef|grep ssh #<==Use grep to find the keyword ssh.

ps aux #<== BSD format parameters, use the a option and x option to display all processes, and use the u option to display the user information of the process.

ps -u root #<==UNIX format parameter, use parameter -u to display process information related to the specified user.

ps -l #<== UNIX format parameter, use the parameter -l to display the status of the process in a detailed format.

ps -eH #<==UNIX format parameters, use the -e parameter to display all processes, and use the -H parameter to display the process tree.

ps axf #<==BSD format parameters, use the a and x parameters to display all processes, and use the f parameter to display the process tree.

ps -o pid,ppid,pgrp,session,tpgid,comm #<==-o is followed by the fields to be displayed, which can be compared with the first line of the command result.

pstree [options] [process ID/user]

pstree #<==If the PID number of the process is not specified, or the user name is not specified, the init process will be used as the root process and all process information of the system will be displayed.

pstree mysql #<==mysql is the user name of the system.

pstree -c -p mysql #<==Use the -c option to display all processes, including child processes and parent processes, and use the -p option to display the process number of the process.

pstree -u #<==Use the -u option to display the user name corresponding to the process.

pgrep [options] [matching criteria]

pgrep crond #<==The pgrep command can be seen as a combination of the ps command and the grep command. The pgrep command specifies filtering the crond field and obtains the process number of the crond process.

pgrep -u root #<==Use the -u option to display all process numbers of the specified root user.

kill [options] [process number]

kill -l #<==Parameter -l displays all signals of the system.

kill -l SIGKILL #<== You can use the -l parameter to interchange signal names and digital signals.

kill -s 15 2203 #<==This format uses the -s parameter to explicitly specify to send a signal with a value of 15, and the effect is the same as kill 2203.

kill -9 2203 #<==Signal 9 will forcibly terminate the process, which will bring some side effects, such as data loss, or the terminal cannot be restored to normal state, so it should be avoided as much as possible unless the process cannot be terminated using other signals.

killall [options] [process name]

killall crond #<==Using killall to terminate the process can be executed several times. crond: no process killed #<==When you see this result, it proves that the process is dead, provided the name is correct.

killall -w crond #<==Use the -w parameter and you will see that the command operation ends after waiting for a few seconds.

killall -u oldboy nginx #<==This method can terminate all nginx processes belonging to the oldboy user

pkill [options] [process name]

pkill crond #<== Terminate the scheduled task process.

pkill -t tty1 #<==Use the -t option to kill the process of the specified terminal.

pkill -u oldboy #<==Use the -u option to kill all processes of the specified user. It is best to specify the process name to kill at the same time to avoid accidentally killing the service.

top [options]

top #<==Using the top command usually does not take any parameters. If you need other more powerful functions, you need to cooperate with interactive commands.

top -a #<==Use parameter -a to sort processes according to memory usage.

top -b #<==Use parameter -b to see that the command execution results are continuously refreshed downwards.

top -c #<==Use the parameter -c to display the entire command path of the process instead of just the command name.

top-d 3#<==Use parameter -d to specify the update period as 3 seconds, which means that the command results are refreshed every 3 seconds.

top -n 2 #<==Use the parameter -n to specify the number of updates to 2 times, which means that the command result will be refreshed twice before exiting. The -n parameter can be used in conjunction with the -d parameter.

top -p 15456 #<==Use the -p option followed by the specified process number to display only the information of this process.

nice [option] [command statement]

When the nice #<== command does not receive any content, it displays that the current system default program running priority is 0.

nice nice #<== Among them, the first nice command adjusts the priority of the second nice command with the default value of 10, that is, adding 10 to the system's default program running priority of 0 to get a new program Run priority 10, and then run the second nice command with priority 10. Finally, the second nice command shows that the current program running priority is 10.

nice -n -10 vim test2 & #<==Use nice to adjust to -10.

renice [options]

renice -n 5 -p 2711 #<==Use the -p parameter of renice to specify the process with a value of 2711, and adjust its NI value to 5.

nohup [options]

nohup ping www.oldboyedu.com nohup: ignoring input and appending output to `nohup.out'#<==The current terminal has been hung. If you force close the current terminal (such as closing the label or SSH client tool), the ping command will still run in the background.

nohup ping www.oldboyedu.com & #<==At work, we usually run the nohup command with the ampersand to let the program run directly in the background.

strace [options]

strace -tt -f /application/nginx/sbin/nginx #<== The -f parameter tracks the target process and all child processes created by the target process. The -tt parameter adds time information before each line in the output, accurate to microseconds, and finally the command statement to be detected. /application/nginx/sbin/nginx is the command to start the Nginx service.

strace -tt -f -e trace=file /application/nginx/sbin/nginx #<==-e trace=file is used to only trace system calls related to file operations.

strace -tt -f -e trace=file -p 1909 #<==Use the -p parameter to only trace the worker process, and the result will be more streamlined.

strace -c /application/nginx/sbin/nginx #<==Use the -c parameter to do a statistical analysis of all system calls of the process.

strace -c -o tongji.log /application/nginx/sbin/nginx #<==Use the -o option to output the results of strace to a file.

strace -T /application/nginx/sbin/nginx #<==Use option -T to print out the time spent on each system call. The time spent on each call is in the rightmost angle bracket of the call line.

ltrace [options]

ltrace /application/nginx/sbin/nginx #<==ltrace is directly followed by the command statement to be detected.

ltrace -p 3892 #<==Use -p to specify the process number.

runlevel [options]

runlevel N 3

·0: shutdown ·1: Single user mode ·2: Multi-user mode without network ·3: Multi-user mode ·4: Not used ·5: Graphical interface multi-user mode ·6: Restart

init [options]

init 0 #<==Shut down

init 6 #<==Restart.

service [service name] [execution command]

Optional values ​​for command include start, stop, status, restart, etc.

service --status-all #<==Display all service status.

service crond #<==Help information will be displayed when the command statement is not finished. crond is the name of the scheduled task service.

service crond stop #<==Stop the service.

service crond start #<==Start the service.

service crond restart #<==Restart the service.

service crond status #<==View service status.

This command was replaced by systemctl in CentOS 7.

fdisk [options] [device name]

fdisk -l #<==View the partition information of all disks in the current system.

fdisk /dev/sdb #<==Without parameters, partitioning can be done by directly connecting the device name.

partprobe [options]

partprobe /dev/sdb #<==It is best to add a specific disk, otherwise an error may be reported. Many people execute this directly and end up reporting an error, so they have to restart the system.

tune2fs [options]

tune2fs -l /dev/sda1|grep -i Mount#<== Check the number of mounts of the sda1 device, that is, the /boot partition.

tune2fs -C 30 /dev/sda1 #<==The parameter -C sets the number of times the file system has been mounted.

tune2fs -c 40 /dev/sda1 #<==The parameter -c sets the number of mounts for forced self-test.

tune2fs -c -1 /dev/sda1 #<==Turn off automatic checking and other functions.

tune2fs -l /dev/sda1|grep -i check #<==View the check period.

tune2fs -i 10 /dev/sda1 #<==The parameter -i setting is checked every 10 days.

tune2fs -i 0 /dev/sda1 #<==Restore to normal state.

parted [options] [device name]

parted -l #<==Display information about all disk partitions.

mkfs [options] [device name]

mkfs -t ext4 -v /dev/sdb #<==Use the -v parameter to display detailed information.

mkfs.ext4 /dev/sdb #<==This way of writing is simpler and the effect is the same.

dumpe2fs [options] [device name]

resize2fs [options] [device name]

fsck [options] [filesystem]

The file system must be unmounted before it can be checked, otherwise errors may occur. There is no need to use this command to check the disk at ordinary times. It only needs to be executed when a disk error is displayed when the system is turned on.

1) In addition to following the boot prompts to repair, you can also use the system disk to enter rescue mode or single-user mode to repair system faults. 2) Never execute fsck to check the disk when the computer is booted and working normally, because this may cause the normal disk to fail. 3) When the last column number in /etc/fstab is 1 or 2, fsck will be read to perform self-test on these system disks when the system is powered on. 4) Do not execute disk repair commands such as fsck on an already mounted file system, as this may cause failure.

dd [options]

dd if=/dev/sda1 of=dev_sda1.img #<==Use if to read data from /dev/sda1, and use of to specify the output to dev_sda1.img in the current directory.

dd if=/dev/zero of=test.data bs=1M count=2 #<==Read data from /dev/zero and write it to test.data. The size of the generated file test.data is bs*count=1M *2=2M.

dd if=test.txt conv=ucase of=test.txt_u#<==Use the conv parameter to set ucase to convert lowercase to uppercase.

mount [options] [device] [directory]

The mounted directory must exist in advance and is preferably empty. If the directory is not empty, the previous directory contents will be covered after mounting the device, but the contents in the original directory will not be damaged. Therefore, if the corresponding device is uninstalled , then the previous directory contents can be accessed again.

mount #<== Enter the mount command directly and press Enter to view the system's mounting information.

mount -l #<==The parameter -l can also view mount information.

mount /dev/cdrom /mnt #<==-t iso9660 is not specified here, but the mount command will automatically recognize it. mount: block device /dev/sr0 is write-protected, mounting read-only#<== Prompts that the device is write-protected and mounted read-only.

mount -o remount,rw / #<==remount attempts to remount "/" as rw (readable and writable).

umount [options] [directory|device]

umount uninstallation can connect to the mount point directory or device files.

umount /mnt #<== can be uninstalled by connecting to the mount point, and umount /dev/cdrom can also be uninstalled.

umount /mnt/ #<== Because it is currently in the mnt directory, it cannot be uninstalled. Here, perform method one to exit the current directory and uninstall.

umount -lf /mnt/ #<==Method 2: Use the -lf parameter to force uninstall.

df [options] [<file or directory>]

If the file parameter after the command is not specified, the usage of all disk partitions will be displayed. If a file is given, the usage of the disk partition where the file is located will be displayed.

df #<==If you do not specify the file parameters after the command, the usage of all disk partitions will be displayed.

mkswap [options] [device file]

mkswap /dev/sdb #<==By default, the entire disk cannot be used as a swap partition.

mkswap -f /dev/sdb #<==Use the -f parameter to force the entire disk to be used as a swap partition.

swapon [option]

swapon /dev/sdb #<==Activate swap partition.

swapon -s #<==Use the -s option to see that there are two swap partitions.

swapoff [option]

When closing the swap partition, make sure that the swap partition is not in use. Otherwise, the system will prompt the error message "device is busy".

swapoff /dev/sdb #<==Close the /dev/sdb swap partition.

swapoff -a #<==Close all swap partitions.

sync [options]

sync #<== Execute the sync command multiple times without any output.

useradd [options] [username] useradd -D [options]

When useradd ett creates a user, it also creates a user group with the same user name. In this example, we added a system user named ett. When viewing the /home/ directory, we will find that the system automatically created an ett directory, which is the user's starting directory after logging in, that is, the home directory.

useradd -g sa -u 901 oldgirl #<==The created user oldgirl belongs to the sa group and the uid is 901.

useradd -M -s /sbin/nologin tingting #<==-M does not create a home directory, -s specifies the shell after the user logs in, here is /sbin/nologin, which means login is prohibited. This example is often used when deploying services such as Nginx and MySQL in production scenarios.

useradd -u 806 -s /bin/sh -c SysUser -G root,sa -e "2017/07/12" -f 2 -d /tmp/inca inca #<==Add user inca and set user comment information For "SysUser", the UID is specified as 806, It belongs to the user group root and sa members, its shell type is /bin/sh, the home directory is set to /tmp/inca, the user expiration time is 2017/07/12, and the user rights will be suspended two days after expiration.

useradd -D -s /bin/sh #<==Modify the default login shell.

useradd -D -e "2018/07/12" #<==Modify the user's default validity period.

The function of useradd-D can be completely replaced by editing and modifying vim/etc/default/useradd.

usermod [options] [username]

usermod -u 888 -s /sbin/nologin -c TmpUser -G root,sa,tech -e "2018/07/12" -f 30 -d /home/inca inca #<==inca’s user comment information is modified to "TmpUser", the UID is changed to 999, the ownership is changed to the user group root, sa, and tech members, its Shell type is /sbin/nologin, the home directory is set to /home/inca, the user expiration time is 2018/07/12, expired Suspension after 30 days.

userdel [options] [username]

userdel zuma #<==Delete zuma user.

userdel -r oldgirl #<== Delete the oldgirl user with the -r parameter.

groupadd [options] [user group]

groupadd -g 123 test1 #<==Add the test1 user group with GID 123.

groupdel [user group]

groupdel root #<==Deleting the root user group failed because the root user still exists.

groupdel cannot delete the main user group to which the user belongs.

passwd [options] [username]

·The root user can change the password of any user, and ordinary users can only change their own password. ·When the root user changes the password, if it does not comply with the system password rules, a warning message will be given, but the password setting will still take effect. When ordinary users change their passwords, if a weak password is used, a warning message will be given and the change will be invalid.

passwd -S oldgirl Only root can do that. #<==Prompt that this parameter can only be executed under root.

echo "123456"|passwd --stdin oldgirl #<==--the stdin parameter can obtain the password from standard input.

passwd -n 7 -x 60 -w 10 -i 30 oldgirl #<==oldgirl The user cannot change the password within 7 days, and must change the password after 60 days. The user is notified 10 days before expiration, and the user is prohibited from logging in after 30 days.

chage [options] [username]

chage -m 7 -M 60 -W 10 -I 30 oldboy #<==oldboy users cannot change their passwords within 7 days, and must change their passwords after 60 days. Oldboy users will be notified 10 days before expiration, and users will be prohibited from logging in after 30 days. .

chage -m7 -M60 -W10 -I30 oldboy #<==The second way of writing.

chpasswd [options]

chpasswd #<==Enter chpasswd on the command line and press Enter. root:123456 #<==Format Username: Password, the user must exist. oldboy:123456 #<==One per line #<==Enter Ctrl D on a new blank line to end the input.

su [options] [username]

1) To switch from an ordinary user to a root user, you can use su- or su-root, but the root password must be entered to complete the switch. 2) To switch the root user to an ordinary user, you can use the writing method "su-ordinary username". There is no need to enter any password to complete the switch. In the CentOS 5. /sbin and other following commands), then you need to use the full path to execute or adjust the content of the PATH variable for ordinary users. CentOS 6 and CentOS 7 do not have this problem. 3) If you only want to execute commands under a certain user without directly switching to that user, you can use su-username-c "command".

visudo [options]

visudo -c #<==Use the -c option for syntax checking.

sudo [options]

id [options] [username]

id #<== does not accept user parameters, the default is the current logged in user.

id oldboy #<==Specifies to display the information of oldboy user.

id -g #<==Display user group GID.

id -u #<==Display user ID.

id -un #<==Display user name (-n parameter means not to display numbers, but to display name).

w [options] [user]

w #<== can generally be used without any parameters.

w -h #<==Use the -h parameter to not display the first two lines of header information.

who [option]

who #<== generally can be used without any parameters.

who -b #<==Display startup time.

who -d #<==Display logged out users.

who -l #<==Display the login process.

who -H #<==Display title.

ho -H -a #<==Use the -H parameter to display the title, and use the -a parameter to display all information.

users #<==If the same user logs in multiple times, the user name will be displayed several times.

The whoami command is used to display the currently logged in user name. This command can be regarded as the abbreviation of the English phrase who am i.

last [option]

last #<== will display many lines.

last -10 #<==Specifies the number of lines to display, and can also be used with the less command through pipes.

last oldboy #<==Displays the login status of the oldboy user, but the oldboy user has not logged in, so the display is empty.

lastb [options]

lastb #<==You need to pay more attention to the results of this command execution. If you find unknown login failure information, you must consider whether the system has been violently cracked to log in.

tar [options] [file or directory]

tar zcvf www.tar.gz ./html/ #<==Option v will display the packaging process. You need to remember the commonly used packaging command combination zcvf. If you do not want to display the packaging process, you can omit the v option, that is, the option combination is zcf.

tar ztvf www.tar.gz #<==Use option t to view the contents of the compressed package without decompression, and option v to display the file attributes.

tar tf www.tar.gz #<==If the z option is not specified, the tar command will also automatically determine the type of compressed package and automatically call the gzip command.

tar zxvf www.tar.gz -C /tmp/ #<==Option C specifies the decompression path. If C is not added, decompression will be to the current directory.

tar xf www.tar.gz -C /tmp/ #<==If you don’t want to see too much output, you can remove the v option and the function will not be affected. At the same time, the z option can also be omitted. As long as the decompression operation is involved, the tar command can automatically identify the compression type of the compressed package, but the z option must be added when compressing.

tar zcvf www.tar.gz ./html/ --exclude=html/oldboy/test #<==Do not add / at the end of the test directory, otherwise it will not succeed.

tar zcvf www.tar.gz ./html/ --exclude=html/oldboy/test --exclude=html/oldboy #<== Method to exclude more than 2 directories: use multiple --exclude in parallel.

tar zcvfX paichu.tar.gz list.txt ./html/ #<==Use parameter X to connect the list of files to be excluded.

When using tar's general option zcf to package a file, if the file is a link file such as /etc/rc.local, then tar will only package the link file itself, not the real file pointed to by the link file, so additional use - The h option packages the entity file corresponding to the soft link file

If the directory to be packaged is a relative path, only the relative path can be followed after --exclude. ·If the directory to be packaged is an absolute path, --exclude can be followed by either an absolute path or a relative path. ·For convenience, the path followed by --exclude and the packaging path should be in the same form, either relative paths or absolute paths.

tar zcvf etc.tar.gz 'find etc/ -type f' #<== Use find to find all ordinary files, and nest a find command statement enclosed in backticks in the tar command statement.

gzip [options] [file]

gzip *.html #<==Use the gzip command to compress all files ending with ".html" in the current directory.

gzip -l *.gz #<== Use the -l parameter to display the compression information of the file without decompression. Because the source files are all empty files, the compression rate is 0.0%.

gzip -dv *.gz #<==Use the -d parameter to decompress the file, and use the -v parameter to display the decompression process.

gzip -c services >services.gz #<== Use the -c option with the output redirection symbol to direct output to services.gz.

zcat services.gz|head #<==zcat command can directly connect the compressed file to read the compressed package.

zcat services.gz >services #<== can also be directly decompressed and redirected to the file.

zip [options] [file or directory]

zip services.zip ./services #<==Format: zip compressed package name compressed file.

zip tmp.zip ./tmp/ #<==This only compresses the file in the directory, and the files in the directory are not compressed.

zip -r tmp.zip ./tmp/ #<==Use the -r option to compress recursively.

zip -r tmp1.zip ./tmp/ -x tmp/services.zip #<==-x option specifies uncompressed files.

unzip [options] [compressed file]

unzip -l tmp.zip #<==Use the -l option to view the file list in the compressed package.

unzip tmp.zip #<==Extract the file directly under the root. Because the source file still exists, the following prompt will appear. #<==Whether to replace files, y yes nno AReplace all files NDo not replace all files rRename

unzip -v tmp.zip #<==Display some information when decompressing.

unzip -o tmp.zip #<==Does not prompt whether to overwrite when decompressing.

unzip -d /tmp tmp.zip #<==You can use the -d option to connect the directory to specify the decompression directory.

scp [options] [user@host1:file1] [user@host2:file2]

scp /etc/services 10.0.0.9:/tmp#<==scp Transferred file name Target host IP address: The directory you want to transfer to.

scp -p /etc/services 10.0.0.9:/tmp #<== Use the -p option to keep file attributes transferred.

scp -rp /tmp 10.0.0.9:/tmp #<==You need to use the -r option to copy the directory

scp 10.0.0.9:/etc/services . #<== Just reverse the order of the pushed commands and download the /etc/services file from the 10.0.0.9 host to the current directory.

scp -rp 10.0.0.9:/tmp . #<==Pull the tmp directory of the 10.0.0.9 host to the current directory.

1) Local mode: rsync [options] [source file] [destination file]

2) Via remote shell access mode: Pull: rsync [options] user@host:source file [destination file] Push: rsync [options] [source file] user@host:destination file

3) rsync daemon mode Pull: rsync [options] user@host::source file [destination file] rsync [options] rsync://user@host:port/source file [destination file] Push: rsync [options] [source file] user@host::destination file rsync [options] [source file] rsync://user@host:port/destination file

rsync -av /data1/ /data2 #<==If there is a slash at the end of the source directory, the contents of the directory will be copied instead of the directory itself.

rsync -av /data1 /data2 #<==If the source directory does not have a slash, the directory itself and the contents under the directory will be copied.

rsync -av /etc/hosts /tmp #<==The source file /etc/hosts and the target directory /tmp are both on the same host.

rsync -av --delete /null/ /tmp/ #<== option --delete makes the contents of the tmp directory consistent with the empty directory. Different files and directories will be deleted, that is, what is in null will be in tmp. If there is any content that is not in null but is in tmp, it must be deleted. Because the null directory is empty, this command will delete all the contents in the /tmp directory.

rsync -av 10.0.0.9:/tmp/ /tmp #<== Pull.

rsync -av /tmp/ 10.0.0.9:/tmp/ #<==Push.

rsync -av -e 'ssh -p 22' /tmp 10.0.0.9:/tmp/ #<== In the previous case, the data synchronized using rsync is transmitted in clear text. In scenarios where data security is required, -e can be used The option uses the SSH tunnel to encrypt and transmit data. -p is the option of the SSH command and specifies the port number for SSH transmission as 22. The result of this command is to push the data in the local /tmp directory to 10.0.0.9 through the SSH encrypted tunnel. The host's /tmp directory. In the same way, execute "rsync -av -e 'ssh -p 22' 10.0.0.9:/tmp/ /tmp" to pull data from the /tmp directory of the 10.0.0.9 host to the local /tmp directory through the SSH encrypted tunnel. .

uname [option]

uname -a #<==Display all relevant information about the system.

uname -m#<==64-bit hardware architecture.

uname -n#<==The host name is oldboy.

uname -r#<==Kernel release version number.

uname -s #<==Kernel name.

uname -v#<==kernel version number.

uname -p#<==The processor type is 64-bit CPU.

uname -o#<==Operating system name.

uname -i#<==hardware platform.

hostname [options]

hostname #<==If no parameters are taken, the host name will be displayed.

The hostname A #<==hostname command can be used to temporarily modify the hostname by adding a hostname.

vi /etc/sysconfig/network #<==Only by modifying the configuration file can the modified host name remain valid after the system is restarted.

It should be noted that in the CentOS7 system, the host name configuration file is changed to /etc/hostname. Modifying /etc/sysconfig/network will not take effect. You need to modify /etc/hostname.

hostname -s #<==Display short format hostname.

hostname -a #<==Display the alias of the host.

hostname -i #<==When the network is not very good, the results will be very slow.

hostname -I #<== It is recommended to use -I to obtain the IP address of the system. The number of IP addresses will be displayed as many network cards (with IP addresses) as there are.

dmesg [options]

dmesg|less #<==View the buffer contents in paging through the less command, which is used to view information such as hardware faults.

stat [options] [file or directory]

du [options] [file or directory]

du -a #<==Displays the size of all files in the current directory (including hidden files and all files in subdirectories).

du -s #<==Display the total size of the current directory.

du -h #<==-h parameters will be converted into easy-to-read and understandable results such as K, M, and G.

du -sh #<==-sh is a commonly used command combination and a recommended method.

du -sh /usr/local/ #<==Display the total size of the specified directory.

du -h --max-depth=1 /usr/local/ #<==Only displays the size of the first level directory.

du -h --max-depth=2 /usr/local/ #<==Only displays the size of the first and second-level directories.

du -h --max-depth=2 /usr/local/ --exclude=/usr/local/share #<==Do not display the size of the /usr/local/share directory.

date [options] [date format]

date %y #<==Display year (short format).

date %Y #<==Display year (long format).

date %m #<==Display month.

date %d #<==Display the day.

date %H #<==Display the hour.

date %M #<==Display minutes.

date %S #<==Display seconds.

date %F #<==Display special format date (year-month-day).

date %T #<==Display special format time (hour: minute: second).

date %F -d "-1day" #<==Show yesterday (concise writing).

date %F -d "yesterday" #<==Show yesterday (English writing).

date %F -d "-2day" #<==Display the day before yesterday.

date %F -d "1day" #<==Display tomorrow.

date %F -d "tomorrow" #<==Display tomorrow (English writing).

date %F -d "2day" #<==Display 2 days later.

date %F -d "1month" #<==Displays 1 month later.

date %F -d "1year" #<==Displays 1 year later.

date -s 20170706 #<== is set to 20170706, and the specific time is empty, which is 00:00:00.

date -s 00:00:03 #<==Set the specific time and the date will not be changed.

date -s "00:00:03 20170706" #<==This can set the entire time.

echo [options] [text]

echo Hello world! #<==echo directly receives the text you want to output.

echo "hello world" >>hello.txt #<== Use the append redirection symbol ">>" to write text to the file.

echo -n "oldboy";echo "oldboy" #<==Use the -n option to output the output in one line without line breaks.

watch [options] [command]

watch -n 1 -d netstat -ant #<==netstat will be explained in detail later, -n specifies the execution of commands every second, and -d highlights.

watch cat oldboy.log<#==Monitor changes in the oldboy.log file in the current directory.

watch -t cat oldboy.log -t parameter does not display the title

which [option] [command name]

which date #<==View the full path of the date command.

which which #<==If an alias is set for the specified command, using the which function will also display the alias.

which cd #<== Bash built-in command cannot use which.

which -a mysql #<==All paths containing mysql commands are displayed.

whereis [options] [filename]

whereissvn

whereis -b svn #<==Find only executable files.

whereis -m svn #<==Only search the man help file.

whereis -s svn #<==Find only source code files.

locate [options] [filename]

locate pwd #<== is directly followed by the file name you want to find. As long as it contains the pwd string, it can be found.

locate -c pwd #<==Only display the number of matching lines.

locate /etc/sh #<==Output as long as part of it matches.

locate /etc/sh* #<== You can also use wildcards.

locate -c /etc/*sh*

updatedb [options]

updatedb -vU /root/ #<==-v displays the update process, -U specifies the update path.

The grep command is one of the most important commands in the Linux system. Its function is to filter matching lines and data from text files or pipe data streams.

grep -v "oldboy" test1.txt #<== Filter lines that do not contain the oldboy string. Pay attention to the filtered string and use double quotes as much as possible.

grep -n "oldboy" test2.txt #<== Output the lines containing the oldboy string and display the line number.

grep -n "." test2.txt #<== displays the line numbers of all lines (similar to cat -n test2.txt). The "." here means matching any single character, that is, matching all content, so, display line numbers of all lines.

grep -i "alex" test2.txt #<==Use the -i parameter to filter alex case-insensitively.

grep -Ei "oldboy|alex" test2.txt #<== It is not case sensitive and filters out strings containing oldboy and alex.

grep -Ei --color=auto "oldboy|alex" test2.txt #<==Add the --color parameter. #<== Matched strings will be displayed in red color.

grep -c "oldboy" test2.txt counts the number of matching strings

sed [options] [sed built-in command characters] [input file]

sed '2a 106,dandan,CSO' persons.txt #<==The sed built-in command a append function is used here.

sed '2i 106,dandan,CSO' persons.txt #<==The sed built-in command i insertion function is used here.

sed '2a 106,dandan,CSO 107,bingbing,CCO' person.txt appends multiple lines of text after the specified line in the file.

sed '2d' person.txt #<==The sed built-in command d is used here to implement the deletion function, specifying the deletion of the text in line 2

sed '2,5d' person.txt #<== "2,5" is a combination of numeric addresses, separated by commas. Its function is to delete the second to fifth lines of the file (delete multiple lines) of text, including Rows 2 and 5, so only row 1 remains.

sed 's#zhangyao#dandan#g' person.txt #<==The sed built-in command s is used here to implement the replacement function, and the global replacement flag g is used to replace all strings matching zhangyao in the file. You need to pay attention to the grammatical format. Place the text "zhangyao" that needs to be replaced between the first and second "#", and place the replaced text "dandan" between the second and third "#" between. The result is that "zhangyao" in the second line is replaced with "dandan".

sed '2p' person.txt #<== The sed built-in command p is used here to implement the query function, and the content of line 2 is specified in combination with the numerical address. However, we will find that the result is not only the output of line 2, but also the rest of the file. The content is also displayed because the sed command has a default output function.

sed -n '2p' person.txt #<== In order to solve the problem of the above command displaying redundant content, use the option -n to cancel the default output and only output the text of the matching line, so you only need to remember to use the command p with the required option - n.

sed -n '2,3p' person.txt #<==Of course, you can use the address range "2,3" to view the contents of lines 2 to 3.

awk [parameter] 'condition {action}' file...

awk 'NR==5' oldboy.txt #<==Compared with the above content, it is indeed line 5.

awk 'NR==2,NR==6' oldboy.txt

awk '{print NR,$0}' oldboy.txt adds a line number before each line of the file.

awk 'NR==2,NR==6 {print NR,$0}' oldboy.txt #<==Display lines 2 to 6 of the oldboy.txt file and print the line number.

awk -F ":" '{print $1,$3,$NF}' oldboy.txt displays the first, third and last columns of the oldboy.txt file.

awk '{gsub("/sbin/nologin","/bin/bash",$0);print $0}' oldboy.txt Replace /sbin/nologin in the file with /bin/bash (awk function function practice).