Ansible: Setting up Hadoop cluster using ansible playbook
Introduction:
Ever thought setting-up Hadoop cluster will be so simple? Like you just need to run an ansible play-book, that’s it!. Yes, today we are going to learn how we right from installation and configuring hdfs-site.xml and core-site.xml file in both master and slave nodes, create hadoop cluster within a few seconds of time.
So lets get started….!
Setting-up inventory file: Add IP addresses of Master node and slave node:
Below is the image where we have created .txt file in which host-name [Master] and [Slave] is created.
Now in this file we need to add ip address, username of system where you are configuring master node and password of that system and as we are configuring a linux system, connection protocol used is ssh.
To get more details about how we can configure ansible, read my blog, Ansible: Configuring httpd inside docker container. In this blog I have explained the pre-requisites.
Let us check whether JDK and Hadoop are present on the manage nodes or not:
Now let us design ansible playbook:
Master code: Explanation..
Slave Node : Explanation…
***Note that above two pictures are just for easy illustration, both master and slave are included in one single play book***
Now let us run the playbook, and see what happens….
Command: ansible-playbook hadoop.yml
Here, the master has configured successfully…..
Here the code execution has completed successfully….!
Now let us check JDK and Hadoop are installed or not:
As you can see both JDK and Hadoop has installed successfully…!
Now let us check whether Master node/Name node and Slave node/ Data node has started or not:
First we will check whether Master node/Name node has started or not:
Lets check Data node as well:
Great..! So now next step is to check connection:
So here we have successfully connected Data node with Name node..!
Thank-you..!