What's in a name?
Before we go into more detail on how to execute plugins, manipulate and evaluate data with Moncli we first have to take a look how we are going to consume the data produced by Moncli from RabbitMQ. I found myself quite often in the situation where I had to write again and again a daemon or process which consumes data from RabbitMQ and do something with it. I already started to write Krolyk which initial goal was to consume Nagios passive check results from RabbitMQ and write them into the Nagios command pipe. I found that too specific, so I changed Krolyk into a more modular or plugin system, which allows you to write a simple Python class which consumes data from RabbitMQ. The nice thing about it is when you write a plugin for Krolyk, it will start one or more parallel consumers without you having to worry about how to organize all that. This makes writing consumers for RabbitMQ easy and accessible to more people. Krolyk's use is however notlimited to the monitoring and metrics collection scope. It's intended to be generic, allowing one to easily write parallel RabbitMQ consumers.
Consuming Moncli data from RabbitMQ
So let's setup Krolyk to consume and simply display data produced by Moncli.
To realize this we can use the skeleton.py plugin. The skeleton plugin can be used as a base for your new and more complex plugins. What skeleton does out of the box is just print the content it consumes from the broker to stdout and acknowledge it back to the broker so it's removed from the queue. Just have a look at the consume function. This should pretty much give you an idea of what you can do with it.
Configuration
Krolyk has a config file in which you can define parameters to connect to RabbitMQ and parameters which are available to the plugin you write. If we take a look at krolyk.cfg we can see under the ["plugins"] section a configuration section for each individual plugin. You have to make sure that the name of the plugin section is exactly the same (including case) as the name of your class.
[ccnbw_ini width="500" lines="-1" ]
[ "plugins" ]
[[ "Skeleton" ]]
"_enabled" = False
"_workers" = 5
"_broker" = "sandbox"
"_queue" = "moncli_reports"
"_user" = "guest"
"_password" = "guest"
"blah1" = "whatever"
"fu" = "bar"
[/ccnbw_ini]
Each plugin/class should have at minimal the parameters defined required to connect to the RabbitMQ broker. These parameters all start with an underscore. All other variables you define will be available to your class as a dictionary called "self.configuration"
A practical example
So let's start an instance of Krolyk which consumes data produced by Moncli and prints it to stdout. Keep in mind Krolyk that if the queue doesn't exist, Krolyk will create a durable one for you.
Start Krolyk in the foreground with a config file which works for your environment:
jelle@indigo:~$ /opt/krolyk/bin/krolyk debug --config /opt/krolyk/etc/krolyk.cfg
Submit a test string to the AMQP default exchange:
And see it appear on the command line:
This example isn't really doing anything exciting but, I at least it gives you an idea on what functionality Krolyk offers. Using the information from this post you should be able to display the data generated by Moncli, so you have an idea what comes out of it.
In the next post we will be creating Requests and Reports with Moncli.