ansible_hadoop

Setup Hadoop HDFS Cluster with Ansible

A role to setup a Hadoop HDFS Cluster. It configures specified systems(hosts) as NameNode and DataNodes and connect them to form a HDFS Cluster.

Requirements

For this role, you have to make gather_facts true as behind the scene it uses some of the Facts of the host otherwise no requirements.

Role Variables

This role has 1 main variable hadoop_type_of_node that needs to be specified. So if you want to configure a system as Hadoop Master then the value of this variable can be namenode or master and similarly for DataNodes(or slaves) the value can be datanode or slave.

Dependencies

This role does not depends on any other Galaxy roles.

Example Playbook

- hosts: namenode
  gather_facts: true
  roles:
    - { role: ajaypathak372.ansible-hadoop ,vars: {hadoop_type_of_node: "namenode"}}

- hosts: datanodes
  gather_facts: true
  roles:
    - { role: ajaypathak372.ansible-hadoop ,vars: {hadoop_type_of_node: "datanode"}}

License

BSD

Author Information

This role is created by Ajay Pathak. For any suggestions, corrections or any other queries you can contact me at LinkedIn.

About

A role to setup a Hadoop HDFS Cluster. It configures specified systems(hosts) as NameNode and DataNodes and connect them to form a HDFS Cluster

Install
ansible-galaxy install Ajaypathak372/ansible-hadoop
GitHub repository
License
Unknown
Downloads
25
Owner
AWS Community Builders | RHCE | Aviatrix Certified Engineer | DevOps Assembly Lines | Hybrid Multi-Cloud | MLOps | Terraform | Ansible | AWS EKS | Kubernetes