Building an On-Premise Kubernetes...
summary
Shopify migrates its Rails monolith and microservices from a homegrown system to on‑prem Kubernetes, building Etcd quorum (3/5/7), anycast master VIP for sub‑second failover, custom Ingress (NGINX) and persistent volumes backed by distributed storage. Challenges include an NGINX reload hotspot bug, hard upgrade matrix between master/plugins, and deciding replication versus rack maintenance risk. The cluster runs production traffic but Shopify ultimately runs production then reverts to cloud because operational drag outweighs business differentiation.
AI-written summary · about
chapters (8)
- 0:00Start
- 0:33Outline
- 6:37On Premise work
- 7:01Master Components
- 7:55(High) Availability Strategies
- 8:51Multi-Master
- 15:04Persistent Storage (Volumes)
- 16:59Successful Failure
related talks
46mKeep Building Fresh: Shopify's Journey to Kubernetes
33mBuilding Shopify's PaaS on Kubernetes
41mShopify's Move from the Data Centre to the Cloud
32mOperating Rails in Kubernetes
42mKubernetes on-premises
32mScaling Shopify's Search: Enhancing Elasticsearch Resilience With Kubernetes and KE...
41mKubernetes on-premises, why you don't want to do it?
44mMigrating a Monolith to Kubernetes