This repository includes some infrastructure automation definitions written by Heat for Openstack.

• ELK Stack
• Cassandra Stack
• Wordpress Stack
• Sample Stacks

First example is about creating and extending elastic search cluster with following features;

It creates ELK cluster with

• 3 elastic masters,
• 3 elastic data,
• 2 elastic ingest nodes,
• 1 logstash and 1 kibana servers,
• Openstack LB for EL Data Pool,
• Openstack LB for EL Index Pool.

Before creating this stack, make sure that project has,

• It's own router, internal and external network,
• Key-pair,
• OS image (tested on RHEL7, CentOS7 should be OK too),
• Security group (ingress port tcp 5601 for kibana, desired port for logstash, tcp 9200 and 9300 for elastic)
• Accessible NFS server that holds most recent elastic search, kibana and logstash tarballs.
• OS should have jdk for Elastic and Logstash, so it can be added into tarballs.
• logstash.conf and license.json file should be placed in NFS server.

No need to allocate Floating IP, yaml allocates while creating stack.

Creates new elastic data node and adds it behind Openstack elastic-data LB that previously created with "provision_ha_elk_cluster.yaml", so Elastic data cluster scales out.

Creates new elastic ingest node and adds it behind Openstack elastic-ingest LB that previously created with "provision_ha_elk_cluster.yaml", so Elastic ingest cluster scales out.

Creates new Kibana node. Cloud-init service configures Kibana, Openstack elastic-ingest LB IP is added to Kibana configuration. It visualize by getting data from Openstack elastic-ingest LB, not reaching directly to ingest node(s), so probablity of losing connection to Elastic Search cluster due to the failure of ingest node(s) that feeds kibana is eleminated.

Creates new Logstash node. Cloud-init service configures Logstash, Openstack elastic-data LB IP is added to Logstash configuration. It sends data to Openstack elastic-data LB, not directly to data node(s), so probablity of losing connection to Elastic Search cluster due to the failure of data node(s) that configured in Logstash is eleminated.

Provisions master node in case of master failures. Replaces new master instead of failed one.

This example is about creating and extending Cassandra cluster with following features;

It creates Cassandra cluster with

• 2 cassandra seeds,
• 2 cassandra nodes,
• Openstack LB for Cassandra instances.

Before creating this stack, make sure that project has,

• It's own router, internal and external network,
• Key-pair,
• OS image (tested on RHEL7, CentOS7 should be OK too),
• Security group (ingress port tcp 9042 for cql , 7000 for internode communication)
• Accessible NFS server that holds most recent Cassandra tarballs.
• OS should have jdk for Cassandra, so it can be added into tarballs.

No need to allocate Floating IP, yaml allocates while creating stack.

Creates new cassandra node and adds it behind Openstack Cassandra LB that previously created with "provision_cassandra..yaml".

This example is about creating wordpress stack with following features;

It creates Wordpress stack with

• 1 Application server instance,
• 1 Database server instance,

Before creating this stack, make sure that project has,

• It's own router, internal and external network,
• 2 Internal networks can be used for app & db network isolation,
• Key-pair,
• 2 prepared OS templates , Mysql installed and Apache+PHP+Wordpress installed, (tested on RHEL7, CentOS7 should be OK too),
• Security group (ingress port tcp 80,443 and 3306)
• Also, wp_first_install.sh curl script allows to configure Wordpress first initialization.

No need to allocate Floating IP, yaml allocates while creating stack.

This example is about creating some sample stacks with following features;

It creates RHEL7 instance with

• key pair,
• floating ip,
• security group,
• time server, name server configuration

Before creating this stack, make sure that project has,

• It's own router, internal and external network,
• Key-pair,
• OS image (tested on RHEL7, CentOS7 should be OK too),
• Security group (ingress port ssh for communication)

No need to allocate Floating IP, yaml allocates while creating stack.

It creates Openstack Load Balancer with following specs,

• Front Port
• Redirect Port
• Floating IP attached to Virtual IP

It creates RHEL7 instance (same specs with provision_rhel7.yaml ) and adds it as a member of previously created Load Balancer with provision_LB.yaml.