Multi-node Kafka cluster

Fixed MirrorMaker systemd service and some minor improvements on Kafka Connect.

Multi-node Kafka cluster

Major rework on use of variables to work with one single inventory file.
Several updates in playbook and genererated kafka-demo.sh script.

Multi-node Kafka cluster

Updated Kafka cluster with:

• 3 broker nodes with ZK ensemble
• 2 worker nodes with Connect, REST API
• 1 Kafka control center node
• 1 Elasticsearch single node cluster
• 1 MySQL node
• MySQL source connector
• Elasticsearch sink connector

Multi-node Kafka server

Initial release

Single-node Kafka server - updated

1.1

Added option "get-topics"

Single-node Kafka server

This release features a single-node Kafka server based on:

• Oracle Linux 7.4
• Confluent Platform 4.1.x

The following Confluent Platform components have been enabled and configured:

• Confluent ZooKeeper cluster manager
• Confluent Kafka broker instance
• Confluent Schema Registry
• Confluent Kafka REST service
• Confluent Kafka Connector (distributed mode)

These components have been integrated into the systemd startup sequence they
will start automatically at boot.

All services will listen by default on the IP address of eth1.
This IP address will appear on-screen during the initial Ansible provision phase.

NOTE:
In this release the Vagrant provisioner "ansible_local" is used.

Future releases will migrate to the regular ansible provisioner to be able
to provision multiple hosts in a single playbook.

Windows hosts can not function as Ansible control nodes so this means that
future releases will only be able to run on Linux and/or Mac OS hosts.