Automatic testing of a heavily integrated system - a case-study
summary
A government case system runs 5,400 tests overnight (about seven hours) against ~25 external integrations and ~100 SOAP/REST calls; the team builds a Spring Boot service mocker and an Elasticsearch-backed test data repository on OpenShift to serve recorded responses, support XPath response modifications triggered by write requests, and let tests request dynamic test persons. The setup drastically reduces flaky failures and maintenance from hardcoded mocks, though Elasticsearch nondeterminism forces returning many candidates and re-running to isolate bad data.
AI-written summary · about
chapters (17)
- 0:00Introduction
- 1:08Show Hands
- 2:23Background
- 8:10System under test
- 13:05Service maka
- 16:15Test Data Repository
- 18:35Test Requests
- 19:50Search Index
- 21:05Service Marker
- 22:50Elastic Search
- 28:35Testing Data Repository
- 30:55Reading data
- 33:25Response modification
- 36:05Service virtualization
- 37:25Summary
- 39:35Example
- 43:45QA
related talks
19mAutomatic Testing for Services at Scale
40mStrategic Business Functionality Testing: Accelerating Development Through Mocking
57mEnd-to-End Automated Testing in a Microservices Architecture
33mThe one with the automation
05mTaming the Testing Beast
25mHow to ensure Testing Robustness in Continuous Testing Environments
44mTesting: Integration Tests - The good, bad and ugly
01h 20mIntegration Testing from the Trenches