Description
Introduction
In this project, you will refactor MiniTwit to use the Web Service API that you created in Project
1, then run multiple instances of both the front- and back-end servers behind a load balancer.
Test Environment
You may use any platform for development and testing, and NGINX is partially supported on
Windows, but note that per the Syllabus the test environment for projects in this course is the
Ubuntu MATE VM for available from http://michael.shafae.com/#resources.
Refactoring MiniTwit
Remove all references to sqlite3 and all database queries from minitwit.py, replacing them
with requests to mt_api.py via the Requests library. Replace the DATABASE configuration
setting with a new API_BASE_URL setting.
In order to run minitwit.py and mt_api.py at the same time, you’ll need to run them on
different ports (e.g., 5000 and 5001).
(Tip: When installing Requests, you can safely ignore Python 3. The library is officially
supported on Python 2.7. Also note that you can use pip without switching to pipenv.)
It’s not impossible that you will run into a design mistake in your API from Project 1, resulting an
an action in the MiniTwit front-end that you cannot accomplish directly using the back-end API.
If so, don’t panic: modify the API as necessary and document the required changes.
Running Multiple Application Servers
To simulate running MiniTwit in production, you will need to run three instances of minitwit.py
and three instances of mt_api.py. To do this, you can create a Procfile and use the foreman
command-line utility.
To install foreman on the Ubuntu MATE VM, use the following shell commands:
$ sudo apt update
$ sudo apt install -y ruby-foreman
(Tip: in recent versions of Foreman, the -c option for concurrency has been replaced by the -m
or –formation option. The version of foreman installed with Ubuntu 16.04 continues to use -c.
Note also that bundle exec is a command for running Ruby applications. You’ll want to use
flask run instead.)
Load Balancing
Install NGINX and verify that you can see the Welcome to nginx! page on http://localhost/.
Now configure two different NGINX server blocks:
● One block, listening on port 80, using the three minitwit.py instances as upstream
servers.
● Another block, listening on port 8080, using the three mt_api.py instances as upstream
servers.
Set the API_BASE_URL for minitwit.py to the load-balanced URL http://localhost:8080.
Testing
You will know that the application is working when you can log into and use MinitTwit on
http://localhost/ and see activity logged by foreman for all three front-end and all three back-end
processes.
Submission
Turn in the code for your project by placing minitwit.py, Procfile, modified configuration
files from /etc/nginx, and any other relevant files in the project2/ subdirectory of the folder
that was you on Dropbox. If you needed to modify mt_api.py, include the updated code and
documentation. You may work alone, or make a single submission for a team of 2-3 students. If
you work in a team, make only one submission.
To complete your submission, print the following sheet, fill out the spaces below, and submit it
to the professor in class by the deadline. Failure to follow the instructions exactly will incur a
10% penalty on the grade for this project for all students on the team.
Project Submission
CPSC 476, Section 1
Project Number ____________________
Names of up to three students for this submission
1. _____________________________________________________________________
2. _____________________________________________________________________
3. _____________________________________________________________________
CSUF email of the Dropbox account containing the project files for this submission
_____________________________________
Comments on your submission
____________________________________________________________________________
____________________________________________________________________________
____________________________________________________________________________
____________________________________________________________________________
____________________________________________________________________________
____________________________________________________________________________
____________________________________________________________________________
____________________________________________________________________________
____________________________________________________________________________
____________________________________________________________________________
____________________________________________________________________________
____________________________________________________________________________
____________________________________________________________________________
____________________________________________________________________________
