Installation

Prerequisites

Fritz Speed is working on a standard Python 2.7 environment and relies on the following packages for querying the Fritz!Box.

For storing the data in a round robin database it uses the python bindings to RRDtool.

To get the scripts you need Git to clone the repository using the URL https://github.com/westfeld/fritz-speed.git

Installation on a RaspberryPi Running Raspian

1. Downloading the Scripts

To install the Fritz Speed scripts clone the GitHub repository:

git clone https://github.com/westfeld/fritz-speed.git

2. Installing the Dependencies

There are many ways to install the dependencies but for the common packages we rely on aptitude:

aptitude install python-rrdtool python-requests python-lxml python-pip

The fritzconnection package is not in the raspbian repository so it will be installed using pip by running:

pip install fritzconnection

3. Configuration

Configuration basically means setting the location of the round robin archive (rra) file in which the data is stored. In addition the properties and location of the individual graphs have to be defined.

The first step is to copy the example configuration file fritz-speed.ini.example to fritz-speed.ini to enable further updates easily.

Starting from the default fritz-speed.ini file typically only two variables have to be edited: rra_filename which defines the round robin archive filename in which the traffic data is stored and graph_basedir which is the output directory under which all graph files will be stored.

Note

The user under which both the monitor-traffic.py and graph-traffic.py scripts are executed has to have read/write permissions to the rra_filename and the graph_basedir directory.

An example of a typical fritz-speed.ini file is shown here:

[DEFAULT]
rra_filename: /var/rrdtool/wan_traffic.rra
graph_height: 300
graph_width: 500
graph_basedir: /var/www/htdocs/pics
# graph type for up/downstream: LINEx (where x = line width), AREA
# more information about plotting
# https://oss.oetiker.ch/rrdtool/doc/rrdgraph_graph.en.html
graph_type_up: LINE1
graph_type_down: LINE1
# graph color for up/downstream in RGB from 00 to FF in 000000 format
graph_color_up: ff0000
graph_color_down: 0000ff

# definitions of graphs to be plotted
# each section corresponds to one graph
[day]
# interval in seconds to be plotted
interval: 86400
# title of plot
title: Traffic of last day
# filename of plot
filename: %(graph_basedir)s/traffic_1d.png

[week]
# interval in seconds to be plotted
interval: 604800
# title of plot
title: Traffic of last week
# filename of plot
filename: %(graph_basedir)s/traffic_1w.png

[month]
# interval in seconds to be plotted
interval: 2592000
# title of plot
title: Traffic of last month
# filename of plot
filename: %(graph_basedir)s/traffic_1m.png

4. Create Round Robin Archive

To create the empty archive simply run the following script:

./create-rra.py

During this setup step the maximum up- and downlink speed is queried from the router and set as the maximum valid value in the archive (an error margin of 10% is added to the determined value). This setting is important because if the router’s transferred bytes counter is reset this would otherwise lead to a spike in the traffic graph.

5. Running as a Cronjob

To monitor the traffic on the Fritz!Box the monitor-traffic.py script is executed periodically, which then reads the traffic counter on the router and stores the value in the round robin archive.

To generate the graphs from the stored data the script graph-traffic.py is executed. It would also update already existing images.

Typically one would add the following line to the user’s crontab to record the data and update the graphs every minute. To do so the command crontab -e opens the user’s crontab and the following line is added

*/1 * * * * [absolute path to script]/monitor-traffic.py && [absolute path to script]/graph-traffic.py