Operations From Scratch to Production at Orange
summary
Orange builds an on‑prem, multi‑DC Cloud Foundry platform with BOSH, contributes a CloudStack CPI & stemcell, and spends four months to get a first cf push working. Production runs apps like musique.orange.fr on three sites using three separate CC DBs (no DB stretching), syncs config with Terraform + Concourse pipelines, and aggregates metrics via Prometheus federation. It handles ~100M pageviews/day, passes security audits with WAF and credential rotation work remaining, automates repave and explores ML for outage detection.
AI-written summary · about
chapters (9)
- 0:00Introduction and context
- 4:24Starting from scratch with Cloud Foundry
- 7:40Multi-site Cloud Foundry deployment
- 10:36Operational challenges and training
- 13:47Developer experience improvements
- 14:30Moving to full-time production
- 16:17Automating deployments with Concourse
- 17:00Monitoring and Federation with Prometheus
- 21:59Future plans and automation goals
related talks
33mCase Study: Orange Labs & Cloud Foundry
29mThe Challenges of Taking Open Source Cloud Foundry to Production
27mCultivate DevOps
17mDeployments We Can Believe In
48mCloud Foundry - your PaaS on OpenStack
29mCloud Foundry Growing Pains: Lessons Learned Keeping the Largest CF Environment Alive and Kicking
31mWinning in Production
29mLessons Learnt Operating Multiple Cloud Foundry on OpenStack Deployments