opsworks_rabbitmq

This is a wrapper cookbook which utilizes the rabbitmq cookbook for installing RabbitMQ on AWS OpsWorks with support for automatic clustering from layer instances.

The recipes included in this cookbook is based on the recipes provided by the community rabbitmq cookbook with modifications which provide better recipe granularity for AWS OpsWorks lifecycle events. For all intents and purposes, it is very similar to the original cookbook.

Requirements

Platforms

• Amazon Linux
• Ubuntu 12.04 LTS
• Ubuntu 14.04 LTS

Cookbooks

• erlang (≥ 0.9)
• rabbitmq (≥ 3.7)

Attributes

Key	Type	Description	Default
['rabbitmq']['opsworks']['layer_name']	String	OpsWorks stack's layer shortname that contains RabbitMQ instances	"rabbitmq"

Recipes

Most of the recipes included by this cookbook used to be in rabbitmq::default on the original rabbitmq cookbook. They have been mostly broken up into two recipes to be executed during AWS OpsWorks instance lifecycle events.

opsworks_rabbitmq::install

Installs RabbitMQ.

Configure your stack layer to run opsworks_rabbitmq::install during Setup lifecycle event in your layer instances.

opsworks_rabbitmq::configure

Sets up directory permissions and generate configuration files from templates. If clustering is enabled, it will also run opsworks_rabbitmq::cluster recipe.

Configure your stack layer to run opsworks_rabbitmq::configure during Configure lifecycle event in your layer instances. See below for more information on why.

NOTE: When clustering is disabled, it should not matter whether this recipe is configured to run during Setup or Configure lifecycle event as long as it runs after rabbitmq::install recipe.

opsworks_rabbitmq::cluster

Automatically gathers all instances in your stack layer and automatically generates the node['rabbitmq']['cluster_disk_nodes'] list required for clustering setup in opsworks_rabbitmq::configure.

Because the list of nodes needs to be updated when a layer instance goes online or shuts down, this recipe must run during Configure lifecycle event in your layer instances.

opsworks_rabbitmq::policy_management

Enables any policies listed in the node['rabbitmq'][policies] and disables any listed in node['rabbitmq'][disabled_policies] attributes.

This recipe should be used to set up High Availability queues in RabbitMQ.

Example Custom JSON:

{
  "rabbitmq": {
    "policies": {
      "ha-all-queues": {
        "pattern": ".*",
        "params": { "ha-mode": "all" },
        "priority": 0
      }
    }
  }
}

This recipe may be executed any time after opsworks_rabbitmq::install recipe. It will automatically execute opsworks_rabbitmq::configure recipe if it has not run before.

opsworks_rabbitmq::user_management

Enables any users listed in the node['rabbitmq']['enabled_users] and disables any listed in node['rabbitmq'][disabled_users] attributes.

opsworks_rabbitmq::virtualhost_management

Enables any vhosts listed in the node['rabbitmq'][virtualhosts] and disables any listed in node['rabbitmq'][disabled_virtualhosts] attributes.

License

This cookbook is licensed and distributed under the Simplified BSD license. See LICENSE for more details.