Operationalizing Kubernetes Sidecars in Production at Salesforce
summary
Salesforce describes running mutating admission webhooks to inject sidecars across Hyperforce, with an open source framework that standardizes hooks, cuts development and testing time, and enforces safety checks before rollout. A new sidecar injection operator performs rolling-update deployments so services stay available while proxies land. Talk covers failure modes observed in production, how many clusters and webhooks are managed, testing harnesses used, and tradeoffs versus alternatives for large-scale sidecar rollout.
AI-written summary · about
chapters (5)
- 0:00Introduction to talk
- 5:22Salesforce Kubernetes context
- 11:24Sidecar operational challenges
- 17:52Production deployment strategies
- 23:22Key takeaways summary
related talks
33mMagic of Kubernetes Sidecar Injection: Present and Future - Anton Ovchinnikov
38mBringing Kubernetes into Salesforce [B]
27mProductionise any Application in Kubernetes Using Sidecar Containers
32mA Peek Inside the Enterprise Cloud at Salesforce
25mRelease the Kraken: Bring Sidecar Containers to Next Level
27mPragmatic Pod Patterns: Leveraging sidecar containers in Kubernetes
32mWhy Sidecars? Moving the Stack out of the Monolith
29mThe Great Sidecar Debate