line_monitor

Scripts to ping a host, record the round-trip-time, and graph the resulting report.

Overview

I coded this project to record the uptime of my home internet. I'm running a Netgear R7000 router, which is supported by the DD-WRT project.

This firmware is capable of answering ping requests, as well as being able to run a dynamic DNS client. This project resolves the dynamic DNC hostname to a real IP address, pings that IP, and then records the RTT to a CSV file. I run this container on a cheap DigitalOcean droplet, and get an uninterupted report of my home's internet connection.

Project Structure

This project is dockerized, and is intended to be run in one of two modes:

1. Dev mode, where the ./docker.sh run command maps the project's app directory into the container to facilitate live coding.
2. Production mode, where the contents of the app directory is actually copied into the image such that the resulting artifact is suitable to be deployed in production, wholly self contained.

Running In Dev Mode

To use this project in dev mode, use docker.sh to drive docker:

1. Ensure LINE_MONITOR_REPO in docker.sh is pointing to this project directory
2. Ensure your target hostname is exported: export TARGET_HOST=foo.example.com (dynamic dns services work nicely here)
3. Build the docker image with: $ docker.sh build
4. Create and run the container with: $ docker.sh run (the container will self-delete when stopped)

You should now have a container named line_monitor running in daemon mode. The output log/report are placed in the app/logs and app/reports directories, repsectively.

Running In Production Mode

To use this project in production mode:

1. Uncomment the second COPY command in the Dockerfile so that the app directory contents will physically be copied into the image, not just mapped

2. Execute $docker.sh build to pick up and build the above change

3. Push the new image (of course changing to your repository name): $ docker push sgskinner/line_monitor:latest

4. Pull the image on your production machine: $ docker pull sgskinner/line_monitor

5. Create a persistnant place for logs/reports on the host to map into the container:

    $ mkdir logs
    $ mkdir reports
   

6. And finally, start a new container with something like:

    #!/bin/bash
   
    TARGET_HOST="foo.example.com"
    export TARGET_HOST
    
    function main() {
        docker run \
            -d \
            -v "$HOME/logs:/usr/src/app/logs" \
            -v "$HOME/reports:/usr/src/app/reports" \
            -e TARGET_HOST="$TARGET_HOST" \
            --rm \
            --name line_monitor \
            sgskinner/line_monitor:latest
    }
    
    main
   

Graphing the Ping Report

The app/scripts/plot_line_report.py script is currently meant for offline use. Meaning, this script is meant to be used locally with a given reports file. The generated graph will be sent to your browser for viewing.

To use, scp the reports/line-monitor.csv to your local machine, ensure the plot_line_report.py's input is pointed at this file, and then execute.

This might be an opportunity for a new feature: Perhaps a django instance that will display the same directly on the remote host?