RabbitMQ

Overview

This check monitors RabbitMQ through the Datadog Agent. It allows you to:

• Track queue-based stats: queue size, consumer count, unacknowledged messages, redelivered messages, etc.
• Track node-based stats: waiting processes, used sockets, used file descriptors, etc.
• Monitor vhosts for aliveness and number of connections

And more.

Setup

Installation

The RabbitMQ check is included in the Datadog Agent package. No additional installation is needed on your server.

Configuration

RabbitMQ exposes metrics in two ways: the RabbitMQ Management Plugin and the Rabbitmq Prometheus Plugin. The Datadog integration supports both versions. Please follow the configuration instruction in this file that pertain to the version you intend to use. The metrics accessible via the Prometheus plugin version of the integration are marked with [OpenMetricsV2] in their metadata.csv descriptions. The Datadog integration also comes with an out-of-the-box dashboard and monitors for each version, as labelled by the Dashboard and Monitor titles.

Prepare RabbitMQ

RabbitMQ Prometheus Plugin.

Starting with RabbitMQ v3.8, the RabbitMQ Prometheus Plugin is enabled by default and the integration communicates with it over HTTP API using OpenMetricsV2.

The Prometheus plugin version of RabbitMQ requires Python 3 support by the Datadog agent, and so can only be supported by Agent V6 onwards. Please ensure your agent is updated before configuring the Prometheus plugin version of the integration.

Configure the prometheus_plugin section in your instance configuration. When using the prometheus_plugin option, settings related to the Management Plugin are ignored.

instances:
  - prometheus_plugin:
      url: http://<HOST>:15692

This enables scraping of the /metrics endpoint on one RabbitMQ node. Datadog can also collect data from the /metrics/detailed endpoint.

instances:
  - prometheus_plugin:
      url: http://<HOST>:15692
      unaggregated_endpoint: detailed?family=queue_coarse_metrics

This enables scraping of the /metrics/detailed endpoint to collect queue coarse metrics.

RabbitMQ Management Plugin.

Enable the plugin. The Agent user then needs at least the monitoring tag and these required permissions:

Create an Agent user for your default vhost with the following commands:

rabbitmqctl add_user datadog <SECRET>
rabbitmqctl set_permissions  -p / datadog "^aliveness-test$" "^amq\.default$" ".*"
rabbitmqctl set_user_tags datadog monitoring

Here, / refers to the default host. Set this to your specified virtual host name. See the RabbitMQ documentation for more information.

Validation

Run the Agent’s status subcommand and look for rabbitmq under the Checks section.

Data Collected

Metrics

Events

Service Checks

rabbitmq.aliveness
Only available with the RabbitMQ Management Plugin. Returns the status of a vhost based on the RabbitMQ Aliveness API. The Aliveness API will create a test queue, then publish and consume a message from that queue. Returns OK if the status code from the API is 200 and CRITICAL otherwise.
Statuses: ok, critical

rabbitmq.status
Only available with the RabbitMQ Management Plugin. Returns the status of the RabbitMQ server. Returns OK if the agent was able to contact the API and CRITICAL otherwise.
Statuses: ok, critical

rabbitmq.openmetrics.health
Only available with the RabbitMQ Prometheus Plugin. Returns CRITICAL if the Agent is unable to connect to the OpenMetrics endpoint, otherwise returns OK.
Statuses: ok, critical

Troubleshooting

Need help? Contact Datadog support.

Further Reading

Additional helpful documentation, links, and articles:

• Key metrics for RabbitMQ monitoring
• Collecting metrics with RabbitMQ monitoring tools
• Monitoring RabbitMQ performance with Datadog

Migrating to Prometheus Plugin

The Prometheus Plugin exposes a different set of metrics from the Management Plugin. Here is what to be aware of as you migrate from the Management to the Prometheus Plugin.

• Look up your metrics in this table. If a metric’s description contains an [OpenMetricsV2] tag, then it is available in the Prometheus Plugin. Metrics available only in the Management Plugin do not have any tags in their descriptions.
• Any dashboards and monitors using Management Plugin metrics do not function. Switch to the dashboards and monitors marked as OpenMetrics Version.
• The default configuration collects aggregated metrics. This means, for example, that there are no metrics tagged by queue. Configure the option prometheus_plugin.unaggregated_endpoint to get metrics without aggregation.
• The rabbitmq.status service check is replaced by rabbitmq.openmetrics.health. The service check rabbitmq.aliveness has no equivalent in the Prometheus Plugin.

The Prometheus Plugin changes some tags. The table below describes the changes to the more common tags.

FAQ

• Tagging RabbitMQ queues by tag family