RabbitMQ monitoring + alerting
It is now possible to set up alerts to monitor your RabbitMQ queues for any server that is using our server monitoring service, Server Density. In a case of eating our own dog food, this feature was necessitated by our recent conversion to RabbitMQ. More on that in a future post.
This feature is available to all trial and paid users. You can choose to monitor total RabbitMQ connections or consumers, messages, messages unacknowledged, messages uncommitted, or memory of your queues.
It requires installation of the rabbitmq-status plugin, which you can find out more about in our documentation. You will also have to install our latest monitoring agent, version 1.5.0. See the docs for details about updating.
Please let us know if you have any other questions.
![Reblog this post [with Zemanta]](http://img.zemanta.com/reblog_e.png?x-id=44117c8e-0ccc-4e99-b21e-1a1d2436053f)
Speaking of Dog food… is there a Mongo monitor in the works?
Yes. MongoDB v1.3 has new status functions so we’ll shortly be releasing an update to include those. We use Mongo heavily so want to get it into Server Density!
Great, after that don’t forget couchdb ;)
Already done. Will be in the next release: http://github.com/boxedice/sd-agent/commit/59ae724c76e1a597cef51c771d15cf40227e78bf
It is *sad* to see plugins that:
1.- Have no documentation other than how to install it.
2.- Have no tests
3.- Have no logging or error reporting.
How is someone supposed to debug this if the above is not met?
What extra documentation do you need?
Why do you want tests?
You can enable the debug log by editing the agent’s agent.py file and changing the debug mode parameter.
this is not a problem with agent.py, and nothing that agent.py can inform if the plugin itself is not working correctly.
If the plugin has some issue, it is not able to report at all what went wrong.
The logger object is passed to the plugin __init__ method. You can use this to log messages to the debug log.