Shulou(Shulou.com)06/02 Report--

The following is the package management front-end tools yum, package manager compilation and installation, sed commands, find file lookup.

1. Linux package management: [front-end tools] for package management

1. Brief introduction of yum software package repository

The management tools of CentOS department are yum and dnf. Yum is a distributed version of Linux that uses rpm as the default package management tool.

URL is the main identification method for yum to locate software package repositories and program files in the repository, as shown below:

For example: schema:// [username @] hostname/PATH/TO/ document [paramers] [methord]

(the Web site in the example can be represented as http/https, that is, the specific example can be: http://www.163.com)

Yum is the [CMARS] architecture, where the servers are called software repositories, or yum repositories. Yum needs to write URL on the client side to specify the server and package path; on the server, the software repository is a directory that contains package files and metadata files. The metadata files are stored in the repodate directory, which can also be said that the path where the repodate directory is located is the command of the software repository.

The commonly used third-party yum repositories are Fedora and EPEL. You can download the epel-release package from the official site or mirror site, and use the rpm command to install to the corresponding Linux distribution. In the / etc/yum.repos.d directory, there will be an extra epel.repo file, and update repolist can be used. Among them, Linux virtual machine is required to connect to the Internet.

2. Create a local yum repository: createrepo command

Createrepo: (createrepo-Create repomd (xml-rpm-metadata) repository)

Format: createrepo [options]

The directory in the format where the package is stored

For example: createrepo / PATH/TO/PACKAGES

# yum-config-manager-add-repo=URL

Display: loaded plug-ins: fastestmirror, refresh-packagekit

Adding repo from: URL

[URL]

Name=added from: URL

Baseurl=URL

Enabled=1

3. The use of [client] in yum's Cramp S architecture

The pointing information of the user-defined repository can be written in the main configuration file, or the pointing information of the user-defined repository can be suffixed with .repo and stored in the / etc/yum.repos.d directory.

(1), main configuration file

The main configuration file: / etc/yum.conf;, which only provides the public configuration content of the information pointed to by the yum repository.

Note: in general, no changes are made to this file.

The main configuration file for yum is in the following directory file: / etc/yum.conf. The information is the ini-style configuration information mode, that is, segmented configuration content, as shown below:

[repositoryid]

Statements

Note: the effective range of all configuration statements starts with the first [] and ends with the next []. The specific examples are as follows:

[repo1]

Statements

[repo2]

Statements

Note: a.repository must be globally unique

B.repository cannot be included. Any special character other than _ can contain letters and numbers, and letters are strictly case-sensitive.

The specific code of the yum main configuration file is shown as follows:

# yum / etc/yum.conf

[main]

Cachedir=/var/cache/yum/$basearch/$releasever

Keepcache=0

Debuglevel=2

Logfile=/var/log/yum.log

Exactarch=1

Obsoletes=1

Gpgcheck=1

Plugins=1

Installonly_limit=5

Bugtracker_url= http://bugs.centos.org/set_project.php?project_id=23&ref=http://bugs.centos.org/bug_report_page.php?category=yum

Distroverpkg=centos-release

In the example, $releasever and $basearch are the variables of the yum configuration file, as follows:

Releasever: the major version number of the current distribution of the operating system

For example, CentOS 7.2-1511, the major version number of its distribution is 7, that is, $releasever is 7

CentOS 6.7. the major version number of its distribution is 6, that is, $releasever is 6

$basearch: the type of infrastructure architecture used by the current computer

For example, there are two kinds of Intel compatible machines $basearch:

I686

X86_64

There are two types of apple computer devices $basearch:

Ppc

Ppc64

(2), fragment configuration file: / etc/yum.repos.d/*.repo

In the fragment configuration file, each file with the suffix .repo is a repository file; so each file can be configured to point to one or more yum repositories.

The yum program looks according to these configuration files when [looking for yum repositories]; each time it may find multiple yum repositories, find the latest version of the package from all available repositories and cache it locally for installation, upgrade and other operations.

Multiple repositories of the same kind are called images, as long as one is available; multiple repositories of different types can exist at the same time, and yum can use these repositories at the same time.

4. Definition of .repo file

The format of the .repo definition is as follows:

[repositoryid]

Name=Full name of repository

Baseurl=schema1://hostname/PATH/TO/REPODATA

Schema2://hostname/PATH/TO/REPODATA

Schema3://hostname/PATH/TO/REPODATA

Schema4://hostname/PATH/TO/REPODATA

# mirrorlist=mirror_url

Enabled= {0 | 1}

Gpgcheck= {0 | 1}

Gpgkey=schema://hostname/PATH/TO/PUBLIC_KEY_FILE

Failovermethod= {roundrobin | priority}

Enablegroups= {1 | 0}

Cost=num

In the definition format, the paragraphs are explained as follows:

Schema can be divided into the following two categories: http [s], ftp, nfs: network server pointing to

File: the local server points to.

Enabled= {0 | 1}: 0 means to disable this repository pointing to

1 means enabled; if there is no enabled statement, it is enabled by default.

Gpgcheck= {0 | 1}: 0 means no package authentication

1 indicates authentication; if it is 1, you need to specify the public key file required for authentication and use the gpgkey statement.

Gpgkey=schema://hostname/PATH/TO/PUBLIC_KEY_FILE

For example: gpgkey= file:///etc/pki/rpm-gpg/RPM-GPG-KEY-CentOS-$releasever

Failovermethod= {roundrobin | priority} failure switching method. The two modes are as follows:

Roundrobin: random search, default method

Priority: select in order

Enablegroups= {1 | 0}: 0 means disabled

1 means enabled.

Cost=num, the default is 1000 for all warehouses.

Example: baseurl= https://mirrors.aliyun.com/epel/7/x86_64

General repo points to content. Specific examples are as follows:

[repoid]

Name=

Baseurl=

# mirrorlist=

Enbaled=1

Gpgcheck=0

# gpgkey=

5. Yum command (CentOS6)

Yum: (yum-Yellowdog Updater Modified)

Format: yum [options] [command] [package...]

(1), [options]

-y,-- assumeyes: all questions that interact with users during yum execution are answered in yes.

-Q,-- quiet: quiet mode, during a certain yum execution, no information is output to stdout.

-- enablerepo=repoidglob enabled

-- disablerepo=repoidglob disabled

It is explicitly specified during a yum execution that a repository point is enabled or disabled; after using such an option, the role of the warehouse pointing to the enabled statement in the file will be invalidated.

-- noplugins: during a yum execution, make it clear that no plug-ins are used.

-- nogpgcheck: during a yum execution, it is clearly stated that if the package integrity is not verified, the gpgcheck=1 statement in the repository pointing to the file will be invalidated.

(2), [command]

* install package1 [package2] [...] Install the package

Note: a. If you just give the package name, yum searches all repositories and finds the latest version of the cache installation

b. If the name and version of the package are given, look for the package that installs the specified version

c. If the package given is the full package name, you are only looking for a package

d. The package name can be globbing, (*? [] [^])

* update [package1] [package2] [...] Package upgrade

* update-to [package1] [package2] [...]

* check-update

* upgrade [package1] [package2] [...]

* upgrade-to [package1] [package2] [...]

* downgrade package1 [package2] [...] Package degradation

* remove | erase package1 [package2] [...] Uninstall the package

* list [...]

For example: yum list LIST OPTIONS

The LIST OPTIONS is:

All: displays all packages, including those that have been installed and those that are not yet installed but available. Note: anaconda: the installer of its CentOS system, the package installed when installing the operating system; # yum list all

Installed: shows all installed packages; # yum list install

Available: shows all packages that have not been installed but can be installed; # yum list available

Updates: shows all packages that can be upgraded and updated; # yum list updates

* info [...] detailed information

For example: yum info LIST OPTIONS

The LIST OPTIONS is:

All: displays all packages, including those that have been installed and those that are not yet installed but available. Note: anaconda: the package that is installed when installing the operating system for the installer of the CentOS system; # yum info all

Installed: shows all installed packages; # yum info install

Available: shows all packages that have not been installed but can be installed; # yum info available

Updates: shows all packages that can be upgraded and updated; # yum info updates

Name [globbing]

* provides | whatprovides feature1 [feature2] [...]

[view specified features] which package is provided

* clean [packages | metadata | expire-cache | rpmdb | plugins | all]

Clean up local cache information

* makecache build cache

* groupinstall group1 [group2] [...] Package group installation

* groupupdate group1 [group2] [...] Package group update

* grouplist [hidden] [groupwildcard] [...] Package group display

* groupremove group1 [group2] [...] Package group deletion

* groupinfo group1 [...] Package group details

* search string1 [string2] [...]

Match based on keywords, and by default, search for package names and summary information; this is a fuzzy search.

* shell [filename]

Provide yum with a [interactive command interface]

* localinstall rpmfile1 [rpmfile2] [...] Local package installation

(maintained for legacy reasons only-use install)

* localupdate rpmfile1 [rpmfile2] [...]

(maintained for legacy reasons only-use update)

To achieve local package installation, that is, do not rely on the software repository, directly operate the rpm format package files. These two commands have been deactivated and replaced with install and update.

* reinstall package1 [package2] [...] Reinstall

* deplist package1 [package2] [...] Show all dependencies for each package

* repolist [all | enabled | disabled] metadata information

The list shows the metadata information of the yum repository; when the following parameters are not given, the default is enabled

* version [all | installed | available | group-* | nogroups* | grouplist | groupinfo]

* history [info | list | packages-list | packages-info | summary | addon-info | redo | undo | rollback | new | sync | stats]

View the history of yum transactions; yum will open transactions during installation, upgrade, uninstall, etc.

* load-transaction [txfile]

Restart a stopped transaction

* check certification

* help [command]

6. Yum command in CentOS7

The yum command in CentOS7 differs from the yum command in CentOS6 in that the administrative commands for package groups are different:

* groups [...]

Groups {install | remove | list | info | update}

Alias: groupinstall groupremove grouplist groupinfo groupudpate

* autoremove [package1] [...] Automatic unloading

* repoinfo [all | enabled | disabled]

7. [variable] in the repo configuration file of yum:

Releasever: the major version number of the current distribution of the operating system

For example, CentOS 7.2-1511, the major version number of its distribution is 7, that is, $releasever is 7

CentOS 6.7. the major version number of its distribution is 6, that is, $releasever is 6

$basearch: the type of infrastructure architecture used by the current computer

For example, there are two kinds of Intel compatible machines $basearch:

I686

X86_64

There are two types of apple computer devices $basearch:

Ppc

Ppc64

$arch: platform architecture type

$YUM0~$YUM9:yum built-in 10 user-defined variables

For example: baseurl= https://mirrors.aliyun.com/epel/7/x86_64 in .repo

8. Why do you use these variables

Because the server (repository) must be able to provide all the common distributions and platform-related packages, but the server cannot predict the specific version usage of any client, so variables are applied to replace it, allowing users to choose when using it.

II. Compilation and installation of CentOS package Manager

1. Why compile and install the source code?

a. The corresponding package cannot be obtained directly

b. The compiled binary does not contain the functions we need.

c. The compiled binary contains so many functions that we don't need

2. Two source code packages:

(1) Source code directly packaged by programmers: name-VERSION.tar.gz

(2) rpm source code package made by a third party using rpm tools: name-VERIONS-release.src.rpm; / usr/src/name-VERIONS

3. The organization of the source code:

(1) composition of multiple documents

(2) there may be dependencies between the code in the file, or even between the code segments of different files.

4. Project management tools for source code:

(1) the project management tool: make

(2) Java project management tool: maven

5. Source code compilation and installation process:

Source code-- > preprocessing (preprocessor, cpp)-- > compilation (compiler, gcc)-- > assembly (assembler)-- > link (linker)-- > binaries used to perform operations

6. Source code files

For each source code file that needs to be compiled, there may be one or more files that he depends on or he [dependent files], and this information is saved to a configuration file; this configuration file tells us:

(1) which preprocessor is called and how to preprocess

(2) which compiler to call and how to compile

(3), the compilation order of files, etc.

Source code files are generally [makefile files], and such makefile files do not exist in advance and do not treat the needs of all users equally. Programmers can use special tools to analyze and extract content from all source code files to build makefile files. All makefile files have a unified [template], namely [Makefile.in]. Users can build makefile files according to their own needs, so as to achieve the compilation process that can meet their needs. Among them, it is usually a script file called [configure] that helps users generate makefile files.

7. Two development tools:

(1), autoconf: help users generate configure scripts

(2), automake: help users generate template files for Makefile.in

For any source code package, INSTALL or README documentation files will be provided after unzipping. Before compiling, it is best to consult the contents of such files to determine the compilation process.

8. There are two ways to obtain the source code of open source programs:

(1) official self-built site:

Apache.org (ASF)

Mariadb.org

(2), code hosting site

SourceForge

Github

Code.google.com

The compiler commonly used to compile the source code is: gcc (GNU C Complier)

9. Two steps for the installation of [Development Environment]:

Step 1:

CentOS7:yum groups install "development tools" or: yum groups install "Development Tools"

CentOS6:yum groupinstall "Development Tools"Server Platform Development"Desktop Platform Development"

Step 2:

Yum install make automake autoconf gcc glibc qt ncurses

10. Three major steps of compiling installer in Linux system:

(1), configure [options...]

(2), make

(3), make install

Step 1:. / configure [options...] [VAR=VALUE]...

Common option [options...]:

(1) setting of installation path:

-- prefix=/PATH/TO/SOMEWHERE

Specify the default program installation path, if this option is not given, the default path / usr/local/

-- sysconfdir=DIR

Specify the installation path for the configuration file

(2) system type:

(3) optional features:

-- disable-FEATURE

Disable a feature or function during compilation

-- enable-FEATURE [= ARG]

Enable a feature or function during compilation

(4) optional package:

-- with-PACKAGE [= ARG]

During the compilation process, choose to use additional packages

-- without-PACKAGE

No additional packages are used during compilation

Step 2: make

Step 3: make install

11. The final compilation work:

After compilation and installation, / usr/local/apache2 will contain bin, sbin, etc, lib, lib64, include, man,. You need to finish the work and export respectively, as shown below.

(1) Export binaries:

Create a file: / etc/profile.d/httpd.sh

Export PATH=/usr/local/apache2/bin:/usr/local/apache2/sbin:$PATH

(2) Export the library file:

Create a file: / etc/ld.so.conf.d/httpd.conf

/ usr/local/apache2/lib64

Ldconfig [- v]

(3) Export header file:

In the / usr/include directory, create a symbolic link to the header file for compiling the installed software:

Ln-sv / usr/local/apache2/include / usr/include/httpd

(4) Export help file:

CentOS6: / etc/man.config

Add a new line: MANPATH=/usr/local/apache2/man

CentOS7: / etc/man_db.conf

Add a new line: MANPATH_MAP=/usr/local/apache2/bin / usr/local/apache2/man

Example: write a script that requires:

1. Provide users with a list of all disks

two。 When the user selects the disk, he prompts the user that the operation is risky and asks if he wants to continue

3. If the user chooses no, exit the script

4. If the user selects yes, [erase all partitions on this disk]

#! / bin/bash

#

Select I in fdisk-l | grep-o "/ dev/sd [a Murz]\ >" quit; do "

Case $I in

/ dev/sd [a Murz])

Read-p "the operation is risky, do you want to continue?" CHOICE

[- z $CHOICE] & & CHOICE=no

Case $CHOICE in

Yes)

Dd if=/dev/zero of=$I bs=512 count=1

Sync

Sleep 1

Sync

Sleep 1

Sync

No)

Break

Esac

Quit)

Break

*)

Continue

Esac

Done

12. Example installation of httpd-2.4.18

(1) first of all, the development tool environment should be solved.

Yum groupinstall "Development tools"

(2) use apr and apr-util tools suitable for httpd-2.4.18 version

Install apr tools

~] # tar xf apr-1.5.2.tar.bz2-C / usr/src

~] # cd / usr/src/apr-1.5.2

~] # / configure-- prefix=/usr/local/apr-1.5.2

~] # make

~] # make install

(3) install arp-util tools

~] # tar xf apr-util-1.5.4.tar.bz2-C / usr/src

~] # cd / usr/src/apr-util-1.5.4

] # / configure-- prefix=/usr/local/apr-util-1.5.4-- with-apr=/usr/local/apr-1.5.2

~] # make

~] # make install

(4) pcre-devel needs to be installed to meet the regular expression requirements of httpd.

Yum-y install pcre-devel

(5) compile and install httpd:

# tar xf httpd-2.4.18.tar.gz-C / usr/src

# cd / usr/src/httpd-2.4.18

#. / configure-with-apr=/usr/local/apr-1.5.2-with-apr-util=/usr/local/apr-util-1.5.4

# make

# make install

(6) closing:

a. Export binaries:

Vim / etc/profile.d/httpd.sh

Export PATH=/usr/local/apache2/bin:$PATH

Source / etc/profile.d/httpd.sh

b. Export the library file:

Vim / etc/ld.so.conf.d/httpd-x86_64.conf

/ usr/local/apache2/modules

Ldconfig-v

c. Export the header file:

Ln-sv / usr/local/apache2/include / usr/include/httpd

3. Sed command

1. Introduction to sed

Sed:Stream EDitor, the stream editor, edits one or more files in behavioral units; each time sed processes [one line] in a given file.

When sed processes text, the current line being processed is stored in a temporary buffer called "pattern space"; the current line is used to match the given PATTERN, and if it can be matched, command editing is used; if it does not match, the default output is output to standard output; and then proceed to the next line until the end of the file.

By default, all processing behavior of sed does not affect the contents of the source file. We usually use the sed command to automatically edit one or more files to simplify the [loop operation] of the file.

2. Sed command

(sed-stream editor for filtering and transforming text)

Format: sed [OPTION]... 'script' [input-file]...

(1). Common options [OPTION]:

-n,-- quiet,-- silent: lines that do not match patterns are not output to the screen by default

-e script,-- expression=script: multi-conditional editing

-f script-file,-- file=script-file: read the script from the specified script-file

Note: script-file is a path that can locate the script file, preferably an absolute path

-I [SUFFIX],-- in-place [= SUFFIX]: direct editing of the original file

-r,-- regexp-extended: enable sed to support extended regular expressions

(2), script part: 'AddressesCommand'

Addresses (address delimiting):

1. Empty address: all lines of all files are processed

two。 Single address: sed processes the only line that can match that address

Num: indicates the line number

/ pattern/: can match all lines of the pattern

$: represents the last line of the document

3. Address range:

Addr1,addr2: all lines in the middle from addr1 to the end of addr2

For example: 2Jing 8

First~step: start with the line number marked by first and take the number represented by step as the step

For example: 1-2, 2-2

Addr1,+N: start with addr1, including the line of addr1, and continue to calculate N lines backwards

For example, 2 is equivalent to 2, 2, 2, 2, 7

Addr1,~N: start with addr1, including the line of addr1, and calculate the addr1*N line backward

For example, 2 ~ 5 is equivalent to 2 ~ 2 ~ 2 ~ 5

/ pattern1/,/pattern2/: is calculated from the first line matched by pattern1 to the end of the first line matched by / pattern2/

Command (processing command):

=: displays the line number of the row matched by the pattern

For example: sed-n'/ ^ $/ ='/ etc/grub2.cfg

A\ text: appends the contents of text to the lines that are matched by the pattern, and supports the use of line breaks to achieve multiple line appends.

I\ text: inserts the contents of text in front of the rows that are matched by the pattern, and supports the use of line breaks to achieve multi-line insertion.

C\ text: change the line matched by the pattern to the content of text, and support the use of\ nnewline, so as to change one line into multiple lines.

D: delete rows that are matched by patterns in the pattern space; such rows can no longer be stdout

Note: the-n option should not be used when using the d command

P: displays the rows in the pattern space that are matched by patterns

Note: when using the p command, it is usually accompanied by the-n option

W filepath: save the lines in the pattern space that are matched by the pattern to the filepath file

R filepath: appends the contents of the filepath file to the lines in the pattern space that are matched by the pattern

! Command: lines that are matched by patterns in pattern space do not execute the Command command; on the contrary, lines that are not matched by patterns execute the Command command

Find and replace. The delimiter can be changed at will, as long as it is the same.

S, s, s

S/pattern/text/ [control]

Swatches _ petchure _

Pattern: what you plan to find and replace

Text: the result to replace

Control: how to replace

G: replace all in the line

P: show the rows that have been replaced successfully

W: filepath: save the replaced line to the filepath file

SUniverse / support backward reference:

S /\ (string\) / & /

S /\ (string\) /\ 1 /

(3) Advanced editing commands:

H: overwrite the contents of the schema space to the holding space

H: append the contents of the schema space to the holding space

G: take the data from the hold space and overwrite it to the schema space

G: take out the content from the hold space and append it to the pattern space

X: interchange the content in the pattern space with the content in the hold space

N: read the next row of the matched row to overwrite the pattern space

N: append the next line of the matched row to the pattern space

D: delete rows in schema space

D: delete all rows in multiline mode space

Example: sed-n'nbot p 'FILE

Sed'1, Gentleman, FILE.

Sed'$! n 'FILE

Sed'$! d 'FILE

Sed 'G'FILE

Sed 'g' FILE

Sed'/ ^ $/ dscape G'FILE

Sed 'ntrandd'FILE

Sed-n'1g / n / 1g / h / n / n / 1 / g / h / n / 1 / n / n / 1 / 2 / 2 / 2 / 3 / 3 / 3 / 3 / 3 / 3 / 3 / 3 / 3 / 3 / 3 / 3 / 3 / 3 / 3 / 3 / 3 / 3 / 3 / 2 / 3 / 3 / 3 / 3 / 3 / 3 / 1 / 2 / 1 / 2 / 1 / 2 / 2 / 1 / 2 / 1 / 2 / 1 / 2 / 1 / 2 / 1 / 2 / 1 / 2 / 1 / 2 / 1 / 2 / 1 / 2 / 1 / 3 / 1 / 2 / 1 / 3 / 1 / 2 / 1 / 2 / 1 / 2 / 1 / 2 / 1 / 2 / n'1 / n'1 / 2 /

IV. File lookup: search for files by file name or file attribute

1. Locate command

Locate is a fuzzy lookup, which does not scan the system, but is based on a dedicated database that should be created in advance and updated regularly. The search speed of locate is very fast, and the search accuracy is very limited.

You can update the locate database manually using the updatedb command.

2. Find command

Find is "precise search", which has high precision and slow speed. Traversing the file name or file attributes at the specified location is strongly discouraged from find the operation root directory.

Find is a real-time lookup. Using the find command, you can only search for commands that the current user has read and execute permissions.

Find command:

(find-search for files in a directory hierarchy)

Format: find [OPTIONS...] [search path] [search condition] [processing action]

[find path]: the default is the current working directory, and you can specify a specific command path.

[search criteria]: the criteria for this search can be file name, file size, file type, file permissions, etc.: specified by default

All files in the directory.

[processing action]: perform a processing operation on files that meet the criteria; the search results are output to the display by default.

[search criteria]

(1) look up according to the file name:

-name file name. Globbing is supported. (*,?, [], [^])

-iname file name, ignore the case of letters, support the use of Globbing, (*,?, [], [^])

(2) search according to the inode number of the document:

-inum inode number: find the corresponding file name and path through the given inode number

-samefile name: find the corresponding inode number through the given file name, and then determine all the file names and paths with the inode number

-links n: find all files with n links

# find /-inum 802210

# find /-samefile test

# find /-link 3 find all files linked to 3

(3) look up according to the regular expression:

-regex pattern: matches the entire file path string with pattern, not just the name of the given file

(4) search according to the owner and subordinate group of the document:

-user uname: search based on the user name of the specified user by the owner

-uid UID: search for a UID based on the owner

-group gname: search by genus group name

-gid GID: search according to the group GID

-nogroup: there is no group name on the file group; there is no group name on the file group, but there are still digital UID and GID.

-nouser: there is no corresponding user name on the owner of the file

(5) look up according to the file type:

Format 1:-type file type

The specific file types are as follows:

B: block equipment

C: character device

D: catalog file

F: ordinary files

L: symbolic link file

P: pipe fil

S: socket file

Format 2:-xtype file type: the matching to the symbolic link file needs to be matched with other options, and the rest is similar to type.

(6) search according to the timestamp:

In days:

-atime [+ | -] n: search based on access time

-ctime [+ | -] n: search according to the change time

-mtime [+ | -] n: search based on modification time

N: that is, [n], denoting the previous (nasty) day and the period within the previous n days-middle period

+ n: that is, [nasty 1, infinity], indicating what happened the day before-preceding paragraph

-n: that is, [now,n], indicating the period from the previous n days to the present-later period

In minutes:

-amin [+ | -] n: search based on access time

-cmin [+ | -] n: search according to the change time

-mmin [+ | -] n: search based on modification time

Example: 5-28-11-18

-mtime-3

5-25-11-18

-mtime 3

5-24-11-18

-mtime + 3

(7) search according to the size of the file:

-size [+ | -] n [cwbkMG]

N: (n Mei 1Jing n]

-n: [0rem nmur1], which indicates the part smaller than n

+ n: (NMagna + ∞)

Example: find-size + 2k

All files greater than 2KB in the current directory

Find-size 2k

Files between all 1KB-2KB in the current directory

Find-size-2k

All files smaller than 1KB in the current directory

(8) search according to permissions:

-perm [/ | -] mode

Mode: exactly matches the specified permissions

/ mode: the logical OR relationship is implied. The condition can be satisfied as long as there is a permission match in any permission bit.

-mode: implies the relationship between logic and. The permission of each permission bit must also contain the specified permission bit in order to meet the condition.

# find-perm 640

# find-perm / 444 matches as long as one has read permission

# find-perm / 540 that is # find-perm / 444-a-perm / 100

# find-perm-444 three read permissions must exist at the same time to match

# find-not-perm / 222 does not have write permission to match

Combination conditions:

-a: logic and, by default, can be omitted

-o: logical or

-not,!: logic not

# find-nouser-nogroup

# find-nouser-o-nogroup

[processing action]

(1),-print

-print: output to the display screen, default action

(2),-ls

-ls: execute the ls-li command to display the results found

(3),-exec COMMAND {}\; execute the COMMAND command for the found result

Difference:-exec is non-interactive

-ok is interactive

{}: placeholder to refer to the path information of all files found by the find command

Instead of performing operations,-exec and-ok:

Chmod Amurr $(find-perm-444-type f)

Find-perm-444-type f | xargs chmod Amurr

Note: the pipeline carries pure string information, so if the command after the pipe is not a command for processing strings, you need to use the xargs command to convert it into parameters that can be processed by subsequent commands.

# find-perm-444-type f-exec chmod Amurr {}\

Namely: # chmod Amurr $(find-perm-444-type f)

# find-not-perm / 444-type f-exec chmod aquir\

Namely: # chmod aqr $(find-not-perm / 444-type f)

# find-perm-444-type f | chmod Amurr runs [failed] because chmod is not dealing with strings

# find-perm-444-type f | tr'amurz''AmurZ' runs [correctly], because tr is already dealing with strings. Grep also handles strings, so it can also run successfully.

Note: the pipeline carries pure string information, so if the command after the pipe is not a command for processing strings, you need to use the xargs command to convert it into parameters that can be processed by subsequent commands.

(4),-ok

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.