This ansible project is a deployment of Apache Flink as a cluster.
Edit the host.ini file:
nano playbooks/inventory/host.ini
Then, remove and replace:
- jobmanager_servers by the server name that will be used as a master
- taskmanager_servers by the list of servers that will be used as slaves (one server name by line)
A default configuration file is provided with variables that are used for the project:
nano playbooks/group_vars/all/config.yml
The playbook downloads Java8 and Apache Flink archive, sets the configuration, creates an alias and a flink systemd service.
cd playbooks
ansible-playbook -i inventory/host.ini deploy.yml
After the deployment, you can access the Apache Flink web interface in your browser:
http:<ip_addr_jobmanager>:8081
During the deployment, a service is set for flink. you can use systemctl to manage the cluster:
systemctl ( start | stop | restart | ... ) flink
A "flink" alias is created inside the .rc file of the shell you informed in config.yml