in this project we have two way for deploying our CDH:

• FullStack - deploying CDH with HA and wide range of machines.
• Minimal - deploying CDH with core machines and without HA .this deployment is used for testing.

• A we should bring up the machines. We are deploying the machines using vagrant tool, so you should install the latest relase of virtualbox and vagrant.Then ,install vagrant-hostmanager plugin using the commands:

$ vagrant plugin install vagrant-hostmanagaer $ vagrant plugin install vagrant-hostsupdater

• B clone the repository to your local machine in path /home/vagrant/(minimal/fullstack)/ using the command git clone.

  $ git clone https://github.com/lolbunny385/Zootopia/ /home/vagrant/

• C , go to path /home/vagrant/(minimal/fullstack) and press the command $ vagrant up ,this processs will provision for you the needed machines.
• D, after the machines are up ,ssh to gw1 using the command $ vagrant ssh gw1.Then,move to /home/vagrant/(minimal/fullstack)/playbook directory and press the command $ansible-playbook site.yaml - this command will deploy for you the CDH cluster.

For fullstack deploying the machines will be :

• manager1 - this machine running cloudera manager server
• manager2 - this machine running CMS
• gw1 - used as haproxy
• gw2 - used as a haproxy.
• master1 -this machine running Hadoop Application ,for example : HDFS,Hbase.
• master2 - this machine ruunning Yarn and HDFS Secondary.
• edge - this machine used as a gateway to hadoop ecosystem ,inclued hadoop gw applications.for example: Hue.
• worker[1-3] - this machines responsible like its name to working .ruuning Datanode and impalad and nodemanager.

For minimal deploying the machines will be :

• manager1 - this machine running cloudera manager server and CMS.
• gw1 - used as haproxy.
• master1 -This machine running hadoop application ,for example : HDFS,Hbase,Yarn.
• edge - this machine used as a gateway to hadoop ecosystem ,inclued hadoop gw applications.for example: Hue.
• worker[1-2] - this machines responsible like it's name to be a worker .ruuning Datanode and impalad and nodemanager.

• site.yaml - This is the main playbook,it is responsible for running all the other playbooks.
• preinstall.yaml - This playbook is responsible for configure optimization to CDH cluster and install cloudera-scm-agent and JDK on the cluster machines.
• mariadb.yaml - this playbook is responsible for configure maria galera db cluster.
• ntp.yaml - this playbook is responsible for configure ntp to machines.
• CM.yaml - this playbook is responsible for installing cloudera-scms-server and provosiong Clouder manager wizard.
• cluster_conf - this playbook is deploying the hadoop cluster ecosystem using cloudera rest api.