5

Trying to install Java (JDK 6) on my new Ubuntu system and getting some bizarro errors. This is my first time ever using any flavor of Linux and so I'm sure it's a user issue (permissions or otherwise).

I downloaded the BIN file directly off Oracle's site (Java SE 6u23 for 64-bit Linux). This defaulted to downloading to /home/myUserName/Downloads.

From there I moved the file to /opt/java, which was a directory I created, because (as a Linux novice) that made sense to be the directory where Java should go.

I then ran the following 2 commands, per instruction I found online for running BINs:

chmod +x jdk-6u23-linux-x64.bin
sudo ./jdk-6u23-linux-x64.bin

Now, in my /opt/java directory I see both the BIN file and the jdk1.6.0_23 directory that seems to be intact upon inspection.

But, when I open a new terminal and run java -version, I get:

The program 'java' can be found in the following packages:
- gcj-4.4-jre-headless
- gcj-4.5-jre-headless
- openjdk-6-jre-headless
Try: sudo apt-get install

What is going on here?!?

(1) Was I wrong to try and make /opt/java my Java directory?
(2) Did I run the wrong commands?
(3) Is Java 1.6.0_23 even installed on my machine?
(4) What are all those gcj-xxx-headless targets?!?!

Thanks for any input!

Improve this question

7 Answers 7

Reset to default
6

Was I wrong to try and make /opt/java my Java directory?

Not really. Many Java developers install multiple JDK installations and always use /opt/jdk1.6.0_23 or similar paths. The bin file you downloaded is not an installer, but merely an extractor. It does not install the java binaries into system folders like /bin.

I usually download the JDK and execute it from within my home folder and afterwards move it to /opt and performing an chown.

Did I run the wrong commands?

Not really. In case you wanted to install a separate JDK, you did it correctly. In case you wanted system integration, you would be better off to use the distribution-specific packages, such as the one installed via aptitude install sun-java6-jdk or alike.

The bin you downloaded is imho more flexible, since I can use it to install multiple verisons of Java on the same system. I know this is something you don't often do on Linux machines.

If you want to use the java binary on command line, you'd have to manually set up the PATH and JAVA_HOME environment variables. I think on Ubuntu that's /etc/environment or /etc/profile or something like that.

Is Java 1.6.0_23 even installed on my machine?

Not really. See above answers.

What are all those gcj-xxx-headless targets?!?

The GCJ is the Gnu Compiler for Java. Obviously, it includes a Java Development Kit and a Java Runtime Environment.

Improve this answer
Sign up to request clarification or add additional context in comments.

1 Comment

Excellent answer, but a small comment: the bin file actually does a little more than just extracting as it also actually generates some stuff.
5

Why downloading a bin, when you can simply:

sudo apt-get install sun-java6-jdk

If there isn't any special reason why you'd want that specific version from the site, you should use apt-get because it will take care of all the stuff like PATH variable, etc.

Improve this answer

6 Comments

For a 'normal' user you are right, but as a Java developer you sometimes need a very specific version to test your software with, like Java 6 update 20, or 23 etc.
@arjan: As a Java developer, I never needed anything more specific than the capital version number. Although, I should say that I mostly developed webapps, and always using the lowest common set of language features.
Typically in web apps that you develop for your own company, the target server contains a very specific Java version. To exclude silly bugs from minor changes between JDK versions we typically use at our development stations exactly the same version as on our server. Of course we very frequently update the server's Java version, so this convention doesn't causes us to be stuck on older versions of Java.
@arjan: Well, what I was trying to say is exactly opposite. For in-house development of webapps you know the Java version on production (and ideally may even change it) while in case of desktop apps each client out there may have a different version of their own.
@arjan: Also, while your approach to have exact same versions between production and development is OK, I must point out that you most likely don't have same OS on all of them. The approach used in my team is to have different versions and even operating systems between different development machines, but at least one of them similar to production (preferably a separate test server). That way you detect and incompatibilities during code integration, rather than getting surprised when deploying it to production.
|
2

Please follow below steps to install oracle java:

Download the latest Java SE SDK version.

http://www.oracle.com/technetwork/java/javase/downloads/index.html

Untar the Archive:

tar -xzvf jdk-8-linux-x64.tar.gz
mv jdk1.8.0 /opt 
cd /opt/jdk1.8.0

This step registers the downloaded version of Java as an alternative, and switches it to be used as the default:

update-alternatives --install /usr/bin/java java /opt/jdk1.8.0/bin/java 1
update-alternatives --install /usr/bin/javac javac /opt/jdk1.8.0/bin/javac 1
update-alternatives --install /usr/lib/mozilla/plugins/libjavaplugin.so mozilla-javaplugin.so /opt/jdk1.8.0/jre/lib/amd64/libnpjp2.so 1
update-alternatives --set java /opt/jdk1.8.0/bin/java
update-alternatives --set javac /opt/jdk1.8.0/bin/javac
update-alternatives --set mozilla-javaplugin.so /opt/jdk1.8.0/jre/lib/amd64/libnpjp2.so

Test

To check the version of Java you are now running

java -version

Output

java version "1.8.0"
Java(TM) SE Runtime Environment (build 1.8.0-b132)
Java HotSpot(TM) 64-Bit Server VM (build 25.0-b70, mixed mode)

To check the browser plugin browse to http://www.java.com/ and click “Do I have Java?”

Ref: https://askubuntu.com/questions/437776/ubuntu-13-04-unable-to-install-jdk7

Improve this answer

Comments

1

You simply have put the JDK binaries in a directory. Although by convention /opt/java or /opt/jdk is often used, these are not directories that are automatically recognized by the system.

You can however update your PATH environment variable to include the /opt/java/bin dir, or symlink (ln -s) /opt/java/bin/java in one of the directories on your system that are included in your path like /usr/bin/

Improve this answer

2 Comments

I appreciate everyone's input - I am going to try and stray away from using apt-get because I like to know how things work, even if I have to go 3 sides around the barn and it takes more time/effort. Ater uncompressing the tarball I am left with that jdk1.6.0_23 directory in /opt/java. Now I am trying to manually set the PATH variable by opening .bashrc and adding PATH=/opt/java/jdk1.6.0_23/bin to the last line. Doing so results with java -version outputting correctly, but then I lose all my terminal tools like ls, cd, dir, clear, etc; error states usr/bin is not on PATH. What do?
This is a really basic Unix question (that's actually not much different from how it works on Windows), but you need to include your old path: path = new path + old path. Fur further details you'd best open a new question if needed.
1

The JDK you installed from Sun/Oracle is the original JDK. The "headless" JDK is the open source alternative. When you run the JDK BIN file, it simply extracts the archive. When you entered the java -version command, it found the FOSS Java, not the Java you had extracted in /opt. As somebody else had mentioned, developers keep multiple versions of the JDK. If you wish to use the Oracle's Java, then you need link /usr/bin/java to /opt/jdk1.6.0_23/bin/java.

sudo ln -s /usr/bin/java /opt/jdk1.6.0_23/bin/java

For this to work, the existing java command should be first delinked from the "headless" JDK. (Do the following before the previous command.)

sudo mv /usr/bin/java /usr/bin/java_old

This assumes that there is a link or executable named java in /usr/bin. Use the which command to be sure.

which java
Improve this answer

Comments

0

To add a new pathname to the existing PATH variable, you need to type this in Terminal: 

PATH=`echo $path`:/your/new/path
export PATH

If you had lost your original PATH variable, you could restore by entering this: 

PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games
export PATH
Improve this answer

Comments

0

Try: 

rm -rf /usr/bin/javac
rm -rf /usr/bin/jar 

ln -s /home/jdk1.6.0_13/bin/javac /usr/bin/javac 
ln -s /home/jdk1.6.0_13/bin/jar /usr/bin/jar

This way, your linux can find java && javac in /usr/bin

Improve this answer

1 Comment

linkfengbin2005.iteye.com/blog/1222849

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.