Test Infrastructure on Steroids
summary
Creates on-demand feature-branch test stacks using Rancher, deploys browser clusters with Selenoid, and records video for failed tests. Infrastructure is managed by Ansible playbooks so broken nodes are reprovisioned automatically; fast system-level feedback is achieved by running a short heartbeat test per branch before longer suites. Notes Windows requires extra hardening and monitoring because the Docker Hyper-V checkbox can flip and stop containers.
AI-written summary · about
chapters (12)
- 0:00Introduction
- 3:57Testing problems
- 8:55Avoiding team conflict
- 11:27Test infrastructure importance
- 14:40Local dev with Selenoid & Docker
- 20:32Windows and Docker issues
- 24:28Remote test infrastructure
- 27:41Selenoid on Windows setup
- 30:11CI/CD for feature branches
- 33:03Cluster monitoring & stability
- 34:48Key lessons
- 36:54Recommendations & conclusion
related talks
47mTest infrastructure on steroids
51mSetting up a scalable CI platform with jenkins, docker and rancher
26mOn-demand Test Infrastructure on ECS as code
32mUsing Docker in test automation
29mTest-driven infrastructure development
45mHighway to Continuous Integration
42mAutomating Infrastructure Deployment for Kubernetes
28mFrom Monitoring to Automated Testing of Your Infrastructure Code