Where DevOps Meets QA: An Observability & Synthetic Testing Success Story
summary
GWI turns k6 browser end-to-end scripts into Kubernetes CronJobs that run every five minutes, export Google Core Web Vitals and custom metrics via k6 remote_write to Prometheus, push logs to Loki and traces to Tempo, and surface failures to Alertmanager -> on‑call; screenshots go to a PV synced to GCS for debugging. Tests retry three times before firing alerts to avoid noise and use SOPS/Helm for secrets plus Makefile wrappers for local dev. Result: constant synthetic monitoring with proactive alerts and actionable dashboards.
AI-written summary · about
chapters (6)
- 0:00Introductions
- 1:20Identifying problems
- 2:45GWI's stack
- 3:26Why reliability is hard
- 6:10Demo on GWI's project
- 23:45Challenges
related talks
01h 55mWorkshop: Continuous Performance with k6 and Grafana
48mPerformance Testing Workshop with k6
27mHow We Streamlined Our SDLC with Observability - from GitHUB via Jenkins, Harbor, Argo to K8S
34mPerformance testing: from zero to hero with k6 & Azure
19mUser-centered observability: load testing, real user monitoring & synthetics
33mMeasuring DevOps
35mIntro: Testing SIG
28mImprove Monitoring and Observability for Kubernetes with OSS Tools