Key metrics to monitor in RabbitMQ - Commands & Configuration

Introduction

Monitoring key metrics in RabbitMQ helps you keep your message broker healthy and efficient. It shows you if messages are flowing well or if there are problems like slow consumers or full queues.

When you want to check if your RabbitMQ server is handling messages without delays

When you need to find out if queues are filling up and might cause message loss

When you want to monitor how many messages are being published and consumed

When you suspect consumers are slow or stuck and want to confirm

When you want to track resource usage like memory and disk to avoid crashes

Commands

This command shows the overall status of the RabbitMQ server including memory, disk, and running nodes.

Terminal

rabbitmqctl status

Expected Output

Status of node rabbit@my-server ...

[{pid,12345},

 {running_applications,[{rabbit,"RabbitMQ","3.11.10"},{os_mon,"CPO  CXC 138 46"}]},

 {memory,[{total,104857600},{connection_readers,123456},{connection_writers,123456}]},

 {disk_free,5000000000},

 {alarms,[]}]

Lists all queues with their name, number of messages ready to be delivered, and number of consumers connected.

Terminal

rabbitmqctl list_queues name messages consumers

Expected Output

queue1	10	2

queue2	0	1

queue3	5	0

Shows all exchanges with their type and the number of messages published to them.

Terminal

rabbitmqctl list_exchanges name type messages

Expected Output

amq.direct	direct	100

amq.fanout	fanout	50

my-exchange	direct	200

Displays all current connections with their state and how many channels each connection has open.

Terminal

rabbitmqctl list_connections name state channels

Expected Output

127.0.0.1:5672	running	3

192.168.1.10:5672	running	1

Lists consumers for each queue, showing if they require acknowledgments which affects message reliability.

Terminal

rabbitmqctl list_consumers queue consumer_tag ack_required

Expected Output

queue1	consumer1	true

queue2	consumer2	false

Key Concept

If you remember nothing else from monitoring RabbitMQ, remember: watch queue message counts and consumer activity to catch bottlenecks early.

Common Mistakes

Ignoring the number of consumers on a queue

If no consumers are connected, messages pile up and cause delays or memory issues.

Always check consumers count with queues to ensure messages are being processed.

Not monitoring memory and disk usage

RabbitMQ can crash or slow down if memory or disk space runs out.

Use 'rabbitmqctl status' regularly to watch resource usage.

Only checking messages published but not messages ready in queues

High published messages with low ready messages might hide consumer problems.

Check both published messages on exchanges and ready messages in queues.

Summary

Use 'rabbitmqctl status' to check server health and resource usage.

List queues with message and consumer counts to find bottlenecks.

Check exchanges for message publishing rates.

Monitor connections and consumers to ensure smooth message flow.