HBase + Subversion + Eclipse + Windows
(it should be easy to adapt for Linux)This is a simple setup in order to play with the source code of HBase under Microsoft Windows.
Update : please note that since HBase-4336 / HBase 0.96 the source tree is split in more than one Maven module this post is no more relevant, i have created a new post on this subject : http://michaelmorello.blogspot.fr/2012/06/hbase-096-eclipse-maven.html
Since HBase use some Unix specific commands like chmod the only requirements here are Cygwin and a working Maven 3 environment. (It is obvious that you need Java and Eclipse, but you DON'T need anything else like the Eclipse Maven plugin or any SSH configuration)
1. Checkout the source code
The first step is to check out the source code from the Subversion repository. I did it under my cygwin home repository. In this example i want to play with the 0.90 branch :
svn co http://svn.apache.org/repos/asf/hbase/branches/0.90/ hbase-0.90
2. Generate the Eclipse configuration files
Go into the hbase-0.90 directory and run the following maven command :
Maven will download half of the internet the first time and at the end you sould have .classpath and .project files at the root of the project directory.
3. Prepare your Eclipse workspace
Start Eclipse and select a new workspace :
Declare the M2_REPO variable :
Menu -> Window -> Preferences
Java -> Build Path -> Classpath Variable
4. Import the source code into your Eclipse workspace
Menu -> File -> Import ...
General -> Existing Projects into WorkspaceSelect root directory" select the directory where you have checked-out and executed the "mvn eclipse:eclipse" command
Now the project HBase should appear in your workspace :
5. Create a simple hbase-site.xml configuration file
In the "conf" directory edit the hbase-site.xml file (it should just contains the root xml element "configuration" for the moment) and add the following lines :
The value of the hbase.rootdir should be an empty writable directory. It is used by HBase in replacement of the HDFS file system in order to store stuff like filestores, wal files, etc....
6. Prepare for ignition
Open the "Run Configurations ..." window from the menu Run -> Run configuration ...
6.1 Main tab
- At the top replace "New_configuration" with "HBase (start)"
- Select the hbase project and in the Main class field enter : org.apache.hadoop.hbase.master.HMaster
6.2 Arguments tab
Add "start" as program argument
6.3 Classpath tab
Add the "conf" directory in the bootstrap entries
The conf directory must appear in the bootstraps entries6.3 Environment tab
As i said earlier HBase will use some Unix commands, so you must add the /bin directory of your Cygwin directory in your PATH environment variable :
7. Ignition !
Press the Run button, and if everything is ok you should see something like that :
You can also check the web ui with your favorite browser : http://localhost:60010/master.jsp
8. Play with the shell
In order to play with the shell create a new "Run configuration"
Menu -> Run -> Run Configurations
Create a new Run Configuration called HBase (Shell) :
Add org.jruby.Main as the main class
In the arguments tab add the absolute path to the hbase-0.90/bin/hirb.rb file
As in the 6.3 part add the conf directory to the bootstarp entries of the classpath
Press the Run button, and if everything is ok you should see something like that in the console :
In green are some commands that you can type in order to check that everything is ok.
9. Stop HBase
It is not recommended to press the red square in Eclipse and kill abruptly HBase. You can duplicate the "HBase (Start)" Run configuration into a "HBase (Stop)" one :
Just replace "start" by "stop"
Now if you select the "HBase (STOP)" run configuration and press the Run button HBase should be gently stopped.
I hope that this post will be useful to anybody that want to discover and play with HBase. It should be very easy to adapt for a Unix/Linux environment since you don't have to care about all the cygwin stuff.