rabbit-hutch

RabbitMq Trace Logger - Listen to multiple RabbitMq instances and log them to a single location or MongoDb database.

Applications included

  • rabbithutch - a service to listen to the exchanges
  • rabbithutchweb - a web site to configure the service

Overview

A ruby service for monotoring the trace output on RabbitMq Nodes and writing the output to console, Log files or MongoDb.

Once tracing is started on the target rabbitMQ nodes, the service creates a queue "rabbithutch" that binds to the exchange "amq.rabbitmq.trace". All messages that arrive on the queue are then picked up by any registered consumers and written to standard output, a log file or a MongoDb Database.

Environment

This has been tested on 
* Ubuntu Linux
* Windows 8

Installation

##Activate tracing on rabbitmq server

run the following command on each server you want to trace
rabbitmqctl trace_on

##Install the rabbit-hutch service

gem install rabbithutch

RabbitMq Configuration

In order for tracing to work, the trace_on option must be enabled on all rabbitMq nodes.
Turn on tracing on the nodes using the following command:
rabbitmqctl trace_on

Usage

Start up the Web Interface

    rabbithutchweb

Run as a command line process with a config file

    rabbithutch run -- ~/config.yaml

Run as a service

    rabbithutch start -- ~/config.yaml

For help on the service

    rabbithutch

Troubleshooting

Web Manager 
If the web service has not been shut down properly the process my still be alive in the background 
ERROR:
     (/home/USERNAME/.rvm/gems/ruby-1.9.3-p327/gems/eventmachine-1.0.0/lib/eventmachine.rb:526:in 
    `start_tcp_server': no acceptor (port is in use or requires root privileges) (RuntimeError))
To fix this kill the process named shotgun
- ps aux | grep shot
- kill -9 (the process number)

Configuration

To add RabbitMq nodes to monitor and consumers to enable you will need to change the configuration.

The config below shows two RabbitMq nodes with each with console, mongoDb and log file consumers

./config.yaml

application:
exchangename: amq.rabbitmq.trace
queuename: rabbithutch

rabbitmq:
    hosts:
      - displayname: local1
        enabled: true
        hostname: 127.0.0.1
        username: guest
        password: guest
      - displayname: local2
        enabled: false
        hostname: 127.2.2.2
        username: guest
        password: guest

consumers_config:
    consumers:
      - name: console_consumer
        enabled: true
      - name: mongo_consumer
        enabled: true
        hostname: 127.0.0.1
        username: guest
        password: guest
        database_prefix: rhutch_
        port: 27017
      - name: log4r_consumer
        enabled: true
        log_location: /tmp
        log_prefix: rhutch_

Contributing

Fork it
Create your feature branch (git checkout -b my-new-feature)
Commit your changes (git commit -am 'Add some feature')
Push to the branch (git push origin my-new-feature)
Create new Pull Request

Create the gem

Build and Test Locally
-gem build rabbithutch.gemspec
-gem install rabbithutch-X.X.X # where X.X.X is the version of the compiled gemspec

Build and deploy to rubygems.org
-gem update --system
-gem build rabbithutch.gemspec
-gem push rabbithutch-0.0.1.gem # where X.X.X is the version of the compiled gemspec

To do

- [ ] Add Http Appender to post messages to url e.g. pastebin.org or mongodb REST interface
- [ ] Come up with better config mgt for gems like adding a management Command Line Interface to set env settings
- [x] Add a Web UI to manage the service