#BuildingTheFuture: Keeping our Dutch Feet Dry: Setup and Operation of a...
summary
Rijkswaterstaat runs a high‑available dual on‑premise active‑standby Cloud Foundry setup across two data centers, serving mission‑critical apps for a 450‑site national water sensor network and ~600 apps (up from ~400). Upgrades use automated PCF/CF pipelines with smoke tests, backups and a one‑week canary delay; tooling includes Ops Manager, Concourse, Terraform, Harbor, Kubernetes/CfCR and Greenplum for DB replication. The setup proves resilient but exposes issues: buildpack breakage, CIDR limits in S672, and tenant network sizing constraints.
AI-written summary · about
chapters (8)
- 0:00Introduction and welcome
- 0:20About Rijkswaterstaat and our journey
- 6:50Need for site redundancy
- 10:47Jump boxes and developer tooling
- 13:18Platform stability practices
- 16:52Upgrade policy for Cloud Foundry
- 21:27Proactive monitoring lessons
- 25:47Q&A on production platforms
related talks
31mPut Your Money Where Your CF Is
50mCracks in the Foundation: Operating Cloud Foundry...
38mWhat Does It Take to Run Cloud Foundry at Enterprise Scale?
19mKeynote: 5 Years of Cloud Foundry at Rakuten
17mDeployments We Can Believe In
16mDeployments We Can Believe In
28mHighly Available Cloud Foundry Deployment for Allstate
46mWhat's Cloud Foundry and why you want to use it