and upgraded over time. Patching is critical to security Anything that isn’t automated is slowing you down. Operators are automated software managers for Kubernetes clusters: Install and Lifecycle. Why you care about Operators
project’s requirements You don’t want to become an expert killernewDB DBA Operators make it easier to run foundation services You can use Operators today
and manage instances of complex stateful applications on behalf of a Kubernetes user 1. Extend the Kubernetes API through the Custom Resources (CR(D)) mechanism 2. Custom controller watches CRs, acts in response What are Operators?
and rights - (local technique, done, smaller font for space) $ kubectl create -f deployment.yaml # Deploy the etcd Operator $ kubectl create -f etcd-cluster.yaml # Create an etcd cluster $ kubectl {delete,apply,get} # etcd failure recovery, version upgrade, Operator failure recovery $> etcdctl put foo bar # While using etcd API
app. The SDK makes it easier to build Operators: • High level APIs and abstractions to write operational logic • Scaffolding and code generation to bootstrap new projects • Extensions to cover common Operator use cases Build your Operator
https://github.com/operator-framework/operator-sdk • You need: Docker local (to build into), Kubernetes or OpenShift (deploy and test) • Local dev clusters: Mini^(kube shift) • sdk-generated skeleton • change for your handler • `Member` name change to v0.0.2 • Rebuild, push, deploy • CRD • Service Account, Role, Binding • Operator • Operator’s managed app: a memcached cluster • $ kubectl describe memcached/example-memcached