Preparing Java, Python environments for big data software

Environment: SUSE 11 64-bit

Installing the Java JDK

1, determine the version. It is common to install the latest JDK (Java SE development Kit). Individual software and systems require a specific version of the JDK, which is downloaded according to actual needs.

2,: http://www.oracle.com/technetwork/java/javase/downloads/jdk8-downloads-2133151.html In fact, Baidu a bit of JDK came out.

3, this is the Java 8 version of the JDK, the small version is 131. This installation will need to be verified later.

4, confirm the model.

X86 Server Download

Itanium minicomputer (IA) needs to be downloaded

5, download. tar.gz can be, direct decompression than with RPM installation is convenient, you can also make the Java directory, more flexible.

6. Create a new Java group. With the root user, using the YaST tool, create a new Java group (the YaST tool is handy and can be used with a statement). Groups can all use the default configuration without configuration. Add the root user to the Java group as well.

7, use root user, establish/java directory. Command: Mkdir/java

8. Upload the downloaded tar.gz pack ftp or other tools to the server/java directory. If you have SSH login, you can upload it with sftp. Unzip the compressed package. Command: TAR-ZXVF xxxx.tar.gz

9. Modify the permissions of the Java directory. With the root user. Command: Chown-r Root:java/java

10, modify the root user profile file. Use the root user. Command: Vi/etc/profile

At the end of the profile file is added (jdk1.8.0_91 is the version I downloaded, which needs to be replaced with your download):

Export java_home=/java/jdk1.8.0_91

Export JRE_HOME=/JAVA/JDK1.8.0_91/JRE

Export path= $JAVA _home/bin: $JRE _home/bin: $PATH

Export classpath=.: $JAVA _home/lib/jt.jar: $JAVA _home/lib/tools.jar: $JAVA _home/lib: $CLASSPATH

If the original java_home, jre_home, need to delete the original, otherwise there may be conflict.

If the original path, classpath, need to add this new additions to the original path, classpath, or it will cause the previous installation of software can not be used.

11, other users need to use Java, to add content in the user's profile file. For example, the system has a user Hadoop and needs to add profile content in the home directory of Hadoop. Command: Log in with Hadoop user

VI ~/.profile

Add the following at the end of the file:

Export java_home=/java/jdk1.8.0_91

Export JRE_HOME=/JAVA/JDK1.8.0_91/JRE

If the original java_home, jre_home, need to delete the original, otherwise there may be conflict.

12. Verify the Java installation. Using the root user, enter Java-version. If the version number (build 1.8.0_91-b14, this version is JDK 8u91) can be displayed with the downloaded version number, the installation is successful.

Installing Python 2.7

1. Confirm the Python version. The Python version is available in version 2 and 3. Do not covet the high version, if the document requires the use of 2 version (such as 2.6, 2.7 and so on), do not download 3 version, or it is likely not use! But always upgrade to a higher version, so you can experiment more. In short, must be tested and then put into production.

2. Python is on the general server. Python versions are generally 2.6, so upgrades are required.

3. Download Python. : https://www.python.org/getit/Download the compressed package back.

4, install the system software. Python software requires some system software support. Otherwise the installation, use will be error. The following software needs to be installed on the SUSE 11 system:

Install gcc gcc-c++ with yast and use the system tray to install

Install the Ncurses-devel with YaST and use the system tray to install it.

install the tack with yast and use the system tray to install it.

GCC and tack are required to be installed, otherwise the Python installation will error. Ncurses-devel is the software required to install tack.

The following two devel packages are optional and only affect Python's use of some additional features, without affecting big data software such as Kafka, Storm, etc.

install readline-devel need to download the installation package online, use RPM installation, csdn have download.

install zlib-devel need to download the installation package online, use RPM installation, csdn have download.

5. Unzip the downloaded compressed package to the/USR/LOCAL/SRC directory

6. Compile the new version of Python. To enter the/USR/LOCAL/SRC directory, the upgrade command is as follows:

./configure--prefix=/usr/local/python2.7 (here is the 2.7 version, according to the actual revision of the version number)

Makes && make install (the compilation process will output a long list of the error)

7, upgrade Python. Replace Python in/usr/bin with a new version of Python, with the following command:

Mv/usr/bin/python/usr/bin/python_old (Make a backup of the existing old Python version)

Ln-s/usr/local/python27/bin/python/usr/bin/(/usr/local/python27 is the directory of the new version of Python, modified according to the actual situation)

8, verify the upgrade situation, using the root user, input Python, display the following interface is the success:

Python 2.4.3 (#1, Sep 3 2009, 15:37:37)

[GCC 4.1.2 20080704 (Red Hat 4.1.2-46)] on linux2

Type "Help", "copyright", "credits" or "license" for more information.

>>>

Preparing Java, Python environments for big data software