Cooja Log Parser

Another Log Parser for Contiki-NG Experiments. Based on examples/benchmarks/result-visualization, the program uses the same node.c

What it does:

• Controls Cooja's experiments.
• Runs Cooja in non-GUI mode.
• Retains the generated data.
• Supports bulk runs (multiple runs with a newly generated random seed).
• Utilises the base example node.c as a mote (node.c create).
• Requires authentication for specific tasks (admin:letmein).
• After the run, the log is parsed into an ORM.
• More than 40 metrics and network behaviour can be extracted from the ORM.

Live Demonstration

You can access a live demo: http://coimbra.ifrn.edu.br/nalpios/ or

You can access a run example

Classes

Runner.py

It's responsible for calling Cooja, and running the experiments could easily used for calling the Cooja by other projects. It is an encapsulated version of run-cooja.py

api.py

It's responsible for presenting experiments, their runs and generated data. It was made in Flask, and the user can create new simulations and runs. The generated information is presented by layer.

Model.py

It's responsible for processing all the things. It's my attempt to develop using the Object-oriented paradigm. TODO: Split apart the data from the presentation

Install and Run

1. Inside the Contiki-NG examples folder clone this repository

   mkdir deploy
   cd deploy/
   git clone https://github.com/contiki-ng/contiki-ng.git
   cd contiki-ng/
   git submodule update --init --recursive
   cd examples/
   git clone https://github.com/ivanilsonjunior/pythonLogParser.git
   cd pythonLogParser/
   . runMe.sh
   

2. Finnaly, just run 'python3 api.py'

Command-Line

1. You can use ipython:

load Model.py
exp = Experiment(name = "Test Experiment", experimentFile = "Metrics25m2Node.csc")
db.add(exp)
db.commit() # After that you have a saved experiment, and you can run
exp.runs # List of runs linked to that experiment
exp.run() # Run the Experiment
r = exp.runs[0] # Gets the first run
r = db.query(Run).filter_by(id=1).first() # Returns the id 1 run
r.records # Shows the run record list
len(db.query(Record).filter_by(run = r).filter_by(node = 2).filter_by(recordType = "TSCH").all()) # Shows the quantity of TSCH records from node 2

Due to my programming skills limitation, I've decided to write the "FrontEnd" using Jinja Templates over Flask