Shulou(Shulou.com)06/01 Report--

Today, the editor will share with you the relevant knowledge points about the installation method of the software package under Linux, the content is detailed, and the logic is clear. I believe most people still know too much about this knowledge, so share this article for your reference. I hope you can get something after reading this article. Let's take a look at it.

Source package installation

A lot of software under Linux is released through the source package. Although it is more complicated to configure and compile than the binary package, its portability is much better. For different architectures, software developers often only need to release the same source package, and different end users can run correctly after compilation.

Specific steps for installation:

$tar zxvf XXXX.tar.gz (or tar jxvf XXXX.tar.bz2) $cd XXXX$. / configure$ make# make install # # install $make clean # # to clean up temporary files after installation $ake uninstall # # to uninstall the software

1. Decompress:

There are two most common source packages (XXXX.tar.gz or XXXX.tar.bz2). In fact, these source packages are compressed by two tools. The tools used in the files at the end of tar.gz are tar and gunzip, while those at the end of tar.bz2 are tar and bzip2. The reason for this is that tar can only package multiple files but does not have the function of compression, while gz and bz2 are just the opposite. Only a single file can be compressed, so our common source code package is actually packaged with different source files through tar, and then released after compression through gunzip or bzip2. Of course, these two steps can be implemented with one command:

The relevant commands for decompressing $tar zcvf XXXX.tar.gz XXXX (or tar jcvf XXXX.tar.bz2 XXXX) are as follows: $tar zxvf XXXX.tar.gz$ tar jxvf XXXX.tar.bz2

2 、. / configure:

After the decompression is completed, the installation will officially begin (add one point, before starting configure, please use the patch tool to patch the source code if necessary). First jump to the decompressed directory of the source code. As we all know, you should carefully read the README or INSTALL files in the source directory before starting configure. Many installation precautions are listed here. Configure is actually a script file. Type ". / configure" in the current directory and shell will run the configure script in the current directory. It must be noted that in the whole configure process, compilation has not yet been carried out, and configure is only doing some preparatory work related to compilation. It mainly does some dependency checks on your current working platform, such as whether the compiler is installed and whether the connector exists. If there are no errors in the detection process, you are lucky. The configure script will generate another file, Makefile, to be used for the next compilation link in the current directory. Of course, configure supports and its rich command line parameters. You can type ". / configre-help" to get specific information, probably the most common one is:

$. / configure-- prefix=/opt/XXX it is used to set the installation directory of the software.

3 、 make:

If the configure process is completed correctly, then the corresponding Makefile file will be generated in the source directory. The Makefile file simply includes a set of file dependencies and the relevant steps of compiling links. In fact, the real compilation link work is not done by make. Make is just a general tool. In general, make will call the appropriate compiler to compile all the source code dependent on the current software according to the rules in Makefile. Generate all relevant object files, and then use the linker to generate the final executable program:

$make

4 、 make install:

When the above two steps are completed correctly, it means that the compilation link process is completely completed. The last thing to do is to install the executable program to the correct location. In this step, ordinary users may not have the permission to operate the relevant directories. Temporarily switching to root is a good choice. "install" is only a label in the Makefile file. "make install" means that the make tool performs all the relevant operations under the "install" label in the Makefile file. If you do not use "- prefix=/opt/XXX" to specify the installation directory of the application during the configure phase, then the application will generally be installed to / usr/local/bin by default. If / usr/local/bin already exists in your PATH, then the installation is almost complete:

$make install

After doing the above, you can wait for the installation to complete.

$make clean$make uninstall

These two steps are only the follow-up operations of the installation. It must be noted that "clean" and "uninstall" are also two corresponding labels in the Makefile file. When performing these two steps, the Makefile file must be retained. "make clean" is used to clear some temporary files in the process of compiling the connection. "make uninstall" is to uninstall related applications. Similar to make install, make uninstall also needs to switch to root for execution. However, the "uninstall" label has been omitted in many Makefile.

RPM package installation

1. How to install the rpm package

The installation of the rmp package can be done using the program rpm. Execute the following command

Rpm-I your-package.rpm

Where your-package.rpm is the file name of the rpm package you want to install, which is usually placed in the current directory.

The following warnings or prompts may appear during installation:

... Conflict with... It is possible that some files in the package to be installed may overwrite existing files, which cannot be installed correctly by default. You can use rpm-- force-I to force the installation. Is needed by. Is not installed... You can use rpm-- nodeps-I to ignore this information if you do not install some of the software required for this package. That is to say, rpm-I-- force-- nodeps can ignore all dependencies and file problems, and any package can be installed, but this forcibly installed package does not guarantee full functionality.

2. How to install the .src.rpm software package

Some packages, which end with .src.rpm, are rpm packages that contain source code and are installed at the time of installation

Needs to be compiled. There are two ways to install this kind of software package

Method 1:

1. Execute rpm-I your-package.src.rpm2. Cd / usr/src/redhat/SPECS3. Rpmbuild-bp your-package.specs a specs file with the same name as your package 4. Cd / usr/src/redhat/BUILD/your-package/ A directory with the same name as your package 5. / configure this step is the same as compiling ordinary source software, you can add the parameter 6. Make7. Make install

Method 2:

1. Execute rpm-I you-package.src.rpm2. The first two steps of cd / usr/src/redhat/SPECS are the same as method one. Rpmbuild-bb your-package.specs a specs file with the same name as your software package

At this point, in the / usr/src/redhat/RPM/i386/ directory (depending on the package, it may be i686 rpm, etc.), there is a new rpm package, which is a compiled binary, which can be installed by executing rpm-I new-package.rpm.

3. How to uninstall the rpm package

Use the command rpm-e package name, which can contain information such as the version number, but cannot have the suffix. rpm

For example, to uninstall the package proftpd-1.2.8-1, you can use the following format:

Rpm-e proftpd-1.2.8-1rpm-e proftpd-1.2.8rpm-e proftpd-rpm-e proftpd

Cannot be in the following format:

Rpm-e proftpd-1.2.8-1.i386.rpmrpm-e proftpd-1.2.8-1.i386rpm-e proftpd-1.2rpm-e proftpd-1

Sometimes there are errors or warnings:

... Is needed by... This shows that this software is needed by other software and cannot be uninstalled casually.

You can use rpm-e-nodeps to force uninstall

4. How to get the files in the rpm package without installing

Use the tools rpm2cpio and cpio

Rpm2cpio xxx.rpm | cpio-virpm2cpio xxx.rpm | cpio-idmvrpm2cpio xxx.rpm | cpio-- extract-- make-directories

The parameter I is the same as extract, which indicates that the file is extracted. V indicates the execution process

D is the same as make-directory, which means that the directory is established according to the original path of the files in the package.

M indicates the time to keep the file updated.

5. How to view files and other information related to rpm packages

All of the following examples assume the use of the package mysql-3.23.54a-11

1. Those rpm packages are installed on my system

Rpm-qa lists all installed packages

If you want to find all installed packages that contain a string sql

Rpm-qa | grep sql

two。 How to get the full file name of a software package

Rpm-Q mysql can get the full name of the mysql package installed in the system, from which you can get information such as the version of the current package.

In this example, you can get the information mysql-3.23.54a-11.

3. Where are the files in a rpm package installed?

Rpm-ql package name Note here is the name of the package that does not include the .rpm suffix

In other words, you can only use mysql or mysql-3.23.54a-11 instead of mysql-3.23.54a-11.rpm. If you just want to know where the executable program is put, you can also use which, such as

Which mysql

4. A rpm package contains those files

A software package that has not been installed, using: rpm-qlp *. Rpm an already installed software package, you can also use: rpm-ql *. Rpm

5. How to get information about the version, purpose and so on of a software package?

A software package that has not been installed, using rpm-qip *. Rpm an already installed software package, you can also use rpm-qi *. Rpm

6. Which package is installed for a program, or which package contains the program

Rpm-qf `which program name` returns the full name of the software package rpm-qif `which program name` returns information about the software package rpm-qlf `which program name` returns the file list of the package

Note that this is not the quotation mark, but `, which is the key in the upper left corner of the keyboard. You can also use rpm-qilf to output package information and file list.

7. Which package installed a file, or which package contains this file

Note that the method in the previous question applies only to executable programs, while the following method can be used not only for executable programs, but also for any ordinary file, as long as you know the file name, first get the full path to the program, you can use whereis or which, and then use rpm-qf for example:

# whereis ftptopftptop: / usr/bin/ftptop / usr/share/man/man1/ftptop.1.gz# rpm-qf / usr/bin/ftptopproftpd-1.2.8-rpm-qf / usr/share/doc/proftpd-1.2.8/rfc/rfc0959.txtproftpd-1.2.8-1

To get the information related to the software package, use rpm-Q _ Q _ Q to indicate the query query, followed by other options, such as

I stands for info to get the information of the software package; l for list to get the list of files; a for all to execute queries in all packages; f for file to make relevant queries according to files; p for package. The query conditions needed to query according to the software package can be generated using grep or from the command line in "``"

Yum installation

Automatically search for the fastest image plug-in: yum install yum-fastestmirror

Install the yum graphics window plug-in: yum install yumex

1. Installation

Yum install install all yum install package1 install the specified installation package package1yum groupinsall group1 installer group group1

2. Update and upgrade

Yum update all updates yum update package1 updates specified packages package1yum check-update checks updatable programs yum upgrade package1 upgrades specified packages package1yum groupupdate group1 upgrades group group1

3. Find and display

Yum info package1 displays installation package information package1yum list displays all installed and installable packages yum list package1 displays specified packages installation package1yum groupinfo group1 displays program group group1 information yum search string finds installation packages based on the keyword string

4. Delete the program

Yum remove | erase package1 delete package package1yum groupremove group1 delete program group group1yum deplist package1 to view program package1 dependencies

5. Clear the cache

Yum clean packages clears the software packages in the cache directory yum clean headers clears the headersyum clean oldheaders in the cache directory clears the old headersyum clean in the cache directory, yum clean all (= yum clean packages; yum clean oldheaders) clears the software packages in the cache directory and the old headers

6 、 TIPS

1. Let the yum command download the package, but do not install it

Yum has a plugin called yum-downloadonly, which allows users to download only software packages.

We can install this plug-in using the following command:

Yum install yum-downloadonly

After the installation is completed, yum has two more command parameters, namely:

-- downloadonly--downloaddir=/path/to/dir

The meaning of these two command parameters is very clear, so there is no need to explain. In this way, users can use yum to download only the software package and not install it automatically.

two。 Do not use the yum-downloadonly plug-in, keep the installation package when installing or upgrading the software

Yum by default, downloaded rpm packages are deleted after upgrade or installation.

However, we can also set the following settings not to delete the downloaded rpm package after the upgrade

Vim / etc/ yum.conf[main] cachedir=/var/cache/yumkeepcache=0

Change keepcache=0 to keepcache=1, and after installation or upgrade, you will have the downloaded rpm package in the directory / var/cache/yum.

These are all the contents of the article "what are the installation methods of software packages under Linux". Thank you for reading! I believe you will gain a lot after reading this article. The editor will update different knowledge for you every day. If you want to learn more knowledge, please pay attention to the industry information channel.

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.