Under the background of pan-agile ideological changes and the popularity of DevOps, the mode of small steps and fast running has greatly reduced the time for quality assurance activities. In the entire R&D efficiency system, testing capabilities have increasingly become a bottleneck that restricts our quality and efficiency improvement. Under the XOps system, Zhongan Technology has always been exploring DevOps quality management engineering , and implemented the construction practice from a single-point dilemma to an integrated quality center
Bottleneck of Quality Assurance Capability under XOps - Agile Testing Transformation
Due to multiple factors such as digital transformation, cloud-native architecture evolution, and the unification of K8s de facto standards, DevOps has exploded and landed rapidly. Many companies have done enough vertical and in-depth work in different subdivisions, and have also precipitated best practices that are in line with their own. In particular, the engineering practice on the CI/CD side can bring a relatively direct and even impactful improvement in R&D efficiency, while the continuous testing capability on the quality assurance side still poses great challenges to the quality team . The ZhongAn Technology team has experienced the transformation from traditional testing to agile testing, and responds to rapid changes through continuous and rapid feedback of test results, mainly reflected in:
- Functional testing: formulate a risk-based testing strategy, shift testing to the left, intervene in risk assessment at the stage of requirements analysis and technical design, control the approval of DOD , and prevent one-sentence or ten-thousand-word requirements from entering the development cycle to ensure the fulfillment of requirements Testability, while assessing the scope of impact of requirements changes and formulating and designing test strategies based on these CheckLists .
- Automated regression: Realize high-frequency and efficient full-volume automated regression test verification through normalized timed task regression, recording and playback, etc.
- Quality access control: Through deep docking with the DevOps release system, automatic use case execution is automatically triggered and execution results are collected to achieve highly credible quality access control
The practice of Zhongan's single-point dilemma to an integrated quality center
Agile testing puts forward higher requirements for automated testing methods, and requires a professional tool system to support the seamless development of quality assurance activities . Fragmented single-point tools or platforms can no longer meet the requirements of continuous delivery on the R&D side of enterprises. The transformation of test agility requires an urgent need for a one-stop/one-click quality middle platform system. The original intention of the birth of ZhongAn Technology Magic Quality Center is to solve the isolated pain points of internal and external users in quality assurance activities.
Based on the one-stop integration of single-point tools in vertical fields such as the company's internal code scanning and automated testing, and the innovative introduction of traffic recording and playback technology, a one-stop, one-click, free-to- detachable DevCube Magic quality center is formed to maximize Reduce the verification cost of full regression to a certain extent.
Highlights of DevCube Magic Quality Center
Online use case assets: use case assets of the same data can be presented in both list and mind map; use case review activities are carried out based on the perspective of requirements, and use case reviews leave traces and can be traced.
·Redefining the test plan: The test plan starts from iteration, free-style series function test, API automation, UI automation and other behaviors, and produces a complete test report to form a closed test loop.
UI automation test: It has comprehensive element positioning, failure rerun and low-resource high-concurrency capabilities, and can display remote execution animations in real time, minimizing scripting costs to improve ROI.
API automated testing: realize the full life cycle management of interface testing, keyword customization, interface resource extraction, various task execution, use case visual arrangement management, pipeline integration and other capabilities.