servermonitor is a small project to provide a few simple web graphs showing server CPU and bandwidth usage

servermonitor is a small project utilising Python, PHP and javascript to provide a few simple graphs on CPU and bandwidth usage. It works by using Python as a webserver to serve json encoded information, and then the javascript to product animated graphs based on that data.

For my original purposes, I had the webserver serving the HTTP and the server serving the json on separate machines, so that is why I chose to separate the two. This project was made to monitor the CPU and bandwidth usage of a game server in real time, and I didn't want the game server running a web server, so the lightweight Tornado Python module was chosen.

Live example: https://www.sheen.id.au/monitor/

Example image:

It depends on following:

• The Python module Tornado (http://www.tornadoweb.org/en/stable/)
• The Python module psutil (https://github.com/giampaolo/psutil)
• The javascript library smoothie charts (http://smoothiecharts.org/)

We also assume you have a web server, such as Apache which can serve up HTML and PHP - note that whilst I am using PHP in this example, it can also be simply HTML - there is no actual PHP - the code running on my live site that I produced this Git from does have PHP which does various unrelated stuff, so me being lazy I just ripped out the irrelevant code and was left with no actual PHP.

Copy the www folder contents to your webserver folder.

Run the Python script servermonitor.py using the following:

/usr/bin/python servermonitor.py 82

Note the port number to listen to is passed as a parameter - in this example I'm using port 82. This is the port the Python webserver will serve the json up on.

You can check the json being served using a browser and visiting the :PORT/monitor addess - eg:

https://sheen.id.au:82/monitor

You should see something like this:

{"rx": 0.2744903564453125, "tx": 0.9374618530273438, "cpu03": 58.0, "cpu02": 26.0, "cpu01": 7.1, "cpu00": 23.2, "cpu07": 1.0, "cpu06": 0.0, "cpu05": 21.2, "cpu04": 4.1}

Now you'll need to edit the index.php to make it look at your Python service for the json - in the code look for the following line and replace it with your own url.

url: "https://www.sheen.id.au:82/monitor"

With that done, you can now visit your front-end web server to view the graphs.