To address these challenges and improve the overall QA process, the following solutions were proposed and implemented:
Unified Testing Environment:
A unified environment was established, providing all QA team members access to a shared database and a single URL. This allowed testing to begin from the initial stages, ensuring consistency and collaboration among team members. The shared environment streamlined the testing process and enabled comprehensive validation.
Comprehensive Documentation:
Detailed test data and test cases were created, prioritizing comprehensive documentation. Each change and bug fix required a well-formatted change document, facilitating effective test planning and ensuring adequate test coverage. The focus on documentation enhanced visibility and traceability, enabling better coordination between the QA team and developers.
Enhanced Server Access for QA:
QA team members were granted access to log files through WinSCP, enabling them to review and analyze logs independently. This eliminated the need for constant developer assistance, leading to faster issue resolution and increased testing efficiency. With improved access to logs, the QA team could identify and address potential problems without relying solely on developers.
Test Automation:
API testing automation was implemented using the Python programming language and the pytest framework. Allure was utilized for reporting purposes. Automated tests covered various API scenarios, reducing manual effort and ensuring consistent and reliable API testing. Test automation improved efficiency and test coverage, particularly for frequent API changes.
Every project has its own challenges and we always welcome every challenge because it helps us in building our strength and make us better. Here are few challenges that we dealt with:
- Coordinating with globally distributed teams from different time-zones across Europe, America and Asia.
- Implementing and maintaining functional automation test and automated integration test with facebook and twitter.
- Most of the on-premise components are desktop application. Biggest challenge was to identify an open source reliable automation tool & framework which can deliver the desired results. After doing multiple POC and evaluations, we have successfully implemented automation using Robot framework, Sikuli and Jython.
- Facebook does not allow automated tests to be executed on their portal and their intelligent spybots could catch automated test scripts instantly resulting in blocked accounts. We overcame this challenge by firing up our own facebook mock which could generate fake facebook interactions and helped us testing our system.
- Having very little knowledge about the system architecture initially and lack of proper documentation, we faced a challenge to understand and test the tasks properly. But we managed to work this out with our dedication and analytical abilities.
- Managing client issues and their testing which included at least 10 different versions of different desktop application components along with backward compatibility tests for all component versions.
