Bunny is a RabbitMQ client that focuses on ease of use. It is feature complete, supports all recent RabbitMQ features and does not have any heavyweight dependencies.
One can use Bunny to make Ruby applications interoperate with other applications (both built in Ruby and not). Complexity and size may vary from simple work queues to complex multi-stage data processing workflows that involve many applications built with all kinds of technologies.
Specific examples:
-
Events collectors, metrics & analytics applications can aggregate events produced by various applications (Web and not) in the company network.
-
A Web application may route messages to a Java app that works with SMS delivery gateways.
-
MMO games can use flexible routing RabbitMQ provides to propagate event notifications to players and locations.
-
Price updates from public markets or other sources can be distributed between interested parties, from trading systems to points of sale in a specific geographic region.
-
Content aggregators may update full-text search and geospatial search indexes by delegating actual indexing work to other applications over RabbitMQ.
-
Companies may provide streaming/push APIs to their customers, partners or just general public.
-
Continuous integration systems can distribute builds between multiple machines with various hardware and software configurations using advanced routing features of RabbitMQ.
-
An application that watches updates from a real-time stream (be it markets data or Twitter stream) can propagate updates to interested parties, including Web applications that display that information in the real time.
Modern Bunny versions support
- CRuby 2.7 through 3.3 (inclusive)
- TruffleRuby
For environments that use TLS, Bunny expects Ruby installations to use a recent enough OpenSSL version that includes support for TLS 1.3.
Bunny works sufficiently well on JRuby but there are known JRuby bugs in versions prior to JRuby 9500 that cause high CPU burn. JRuby users should use March Hare.
Bunny 1.7.x
was the last version to support CRuby 1.9.3 and 1.8.7
Modern Bunny releases target currently supported RabbitMQ release series.
To use Bunny in a project managed with Bundler:
gem "bunny", ">= 2.23.0"
To install Bunny with RubyGems:
gem install bunny
Below is a small snippet that demonstrates how to publish and synchronously consume ("pull API") messages with Bunny.
For a 15 minute tutorial using more practical examples, see Getting Started with RabbitMQ and Ruby using Bunny.
require "bunny"
# Start a communication session with RabbitMQ
conn = Bunny.new
conn.start
# open a channel
ch = conn.create_channel
ch.confirm_select
# declare a queue
q = ch.queue("test1")
q.subscribe(manual_ack: true) do |delivery_info, metadata, payload|
puts "This is the message: #{payload}"
# acknowledge the delivery so that RabbitMQ can mark it for deletion
ch.ack(delivery_info.delivery_tag)
end
# publish a message to the default exchange which then gets routed to this queue
q.publish("Hello, everybody!")
# await confirmations from RabbitMQ, see
# https://www.rabbitmq.com/publishers.html#data-safety for details
ch.wait_for_confirms
# give the above consumer some time consume the delivery and print out the message
sleep 1
puts "Done"
ch.close
# close the connection
conn.close
For a 15 minute tutorial using more practical examples, see Getting Started with RabbitMQ and Ruby using Bunny.
Bunny documentation guides are under docs/guides
in this repository:
- Queues and Consumers
- Exchanges and Publishers
- AMQP 0.9.1 Model Explained
- Connecting to RabbitMQ
- Error Handling and Recovery
- TLS/SSL Support
- Bindings
- Using RabbitMQ Extensions with Bunny
- Durability and Related Matters
Some highly relevant RabbitMQ documentation guides:
- Connections
- Channels
- Queues
- Quorum queues
- Streams (Bunny can perform basic operations on streams even though it does not implement the RabbitMQ Stream protocol)
- Publishers
- Consumers
- Data safety: publisher and consumer Confirmations
- Production Checklist
Bunny has a mailing list. Please use it for all questions, investigations, and discussions. GitHub issues should be used for specific, well understood, actionable maintainers and contributors can work on.
We encourage you to also join the RabbitMQ mailing list mailing list. Feel free to ask any questions that you may have.
If you find a bug you understand well, poor default, incorrect or unclear piece of documentation, or missing feature, please file an issue on GitHub.
Please use Bunny's mailing list for questions, investigations, and discussions. GitHub issues should be used for specific, well understood, actionable maintainers and contributors can work on.
When filing an issue, please specify which Bunny and RabbitMQ versions you are using, provide recent RabbitMQ log file contents, full exception stack traces, and steps to reproduce (or failing test cases).
The other widely used Ruby RabbitMQ client is March Hare (JRuby-only). It's a mature library that require RabbitMQ 3.3.x or later.
See CONTRIBUTING.md for more information about running various test suites.
Released under the MIT license.