A solo project to fulfill the DevOps Fundamentals Project Specifications
- Brief
- Solution
- Risk Assessment
- Kanban!
- Entity Relationship Diagrams
- Testing
- Deployment
- User Web Interface
- Relevant Information and Links
The client ( My Step-Son ) wanted an application to help him keep track of certain premier league players and save some details that he will later on perform statistical calculations on to help him predict the player’s future performance,He wanted the freedom to add or remove players as needed and also the freedom to edit or remove certain stats on certain dates for certain players.
To workout the solution utilising the supporting tools, technologies and methodologies covered during the training in the academy, I’ve decided the following:
- Adopt Agile methods to manage my project
- Perform risk assessment using the outline covered during the training
- Implement kanban method using Trello
- Build database ERD using Draw.io
- Create Git repository to be my VCS
- Use MySQL server, HTML5, CSS3 , Python, Flask and a few more python modules to build and test the application (refer to the requirements.txt for more details )
- Host the Application ( MySQL Database Server, Linux Web Application Server and Linux server hosting the automated deployment server for continuous integration ‘Jenkins’ ) on Google Cloud Platform ( free tier )
- Test the application
- Deploy the application
Let's cover as much as possible from above
| Risks | Likelihood | Impact |
|---|---|---|
| Wrong interpretation of client needs | Low | High |
| Lack of Technical knowledge | Medium | Medium |
| Automation causing issues | Low | High |
| Exposure of sensitive information | Low | High |
| Running out of GCP credit | Low | High |
This risk was mitigated by having regular meetings with the client throughout the development cycle to ensure that the product is meeting the client expectations.
This risk was mitigated by constantly reviewing each step and referring to documentations and best practices provided in the training and from other sources
This risk was mitigated by testing the automation process with a simple ‘Hello World’ application to ensure that the CI Pipeline is functional, Also keeping the server on to avoid having the IP address change
This risk was mitigated by using the best practice of exporting sensitive information as variables for each administration session only. This way sensitive information will not be part of the application code on git hub ( although currently it shows on Jenkins console log, trying to figure out a way to hide it from there also )
This one was mitigated by using micro VM instances as they don’t cost much
To create a back-log for the application and to also keep track of my progress while developing the application, I used Trello Board. It contained all the user stories and my To Do list. It gets updated regularly.
The Testing for the application was done using pytest in conjunction with selenium module, the test only covered the pages that didn't require login using the backend test. but using selenium a test to the login protected was performed successfully although it doesn't show on the coverage report.
Application deployment is automated using a webhook from github to the CI server Jenkins. Whenever a commit is made to the master branch, the CI server is notified and it will grab a copy of the changes build and deploy it to the application server, it can be better visualised looking at the diagram below.
- Application link : http://35.189.64.220:8080/
- Trello Board Link: https://trello.com/b/MYYfqs6z/soccer-players-tracker
To get this app running locally on an ubuntu machine below are extra things to consider.
- Install Python 3.
- Install Python3-venv.
- A secret Key exported to your terminal session.
- MySQL Server 5.7 contains an empty database.
- A Database URI exported to your terminal session.
- Install Requirements.txt in a virtual environment and activate it.
- Run the Application.
you can now run it on docker using the image which is alreayd on docker hub 'elsahib/devops' use this command to have it running in development server:
-
- export your Database URI and secret key to your shell session
-
- run the commnad
docker run -d -p 5000:5000 -e SECRET_KEY -e DATABASE_URI --name devops elsahib/devops
- run the commnad
-
- visit in your browser visit 'http://you_docker_server_IP:5000/'
You can also deploy it on AKS unsing the devops.yaml file.
you will need to substitute the values for the DATABASE_URI and SECRET_KEY with your exported values using this command sed -e "s,{{DATABASE_URI}},$DATABASE_URI,g" -e "s,{{SECRET_KEY}},$SECRET_KEY,g" devops.yaml | kubectl apply -f -
Osman Elsahib
My deepest regards and appreciation goes to the QA Academy team, and even more special regards goes to my direct trainers for their wonderful delivery of the training materials and their non-stop support.