Skip to content

Latest commit

 

History

History
121 lines (84 loc) · 5.17 KB

README.md

File metadata and controls

121 lines (84 loc) · 5.17 KB
Raw

Kubernetes API Server Operator

The Kubernetes API Server operator manages and updates the Kubernetes API server deployed on top of OpenShift. The operator is based on OpenShift library-go framework and it is installed via Cluster Version Operator (CVO).

It contains the following components:

  • Operator
  • Bootstrap manifest renderer
  • Installer based on static pods
  • Configuration observer

By default, the operator exposes Prometheus metrics via metrics service. The metrics are collected from following components:

  • Kubernetes API Server Operator

Configuration

The configuration observer component is responsible for reacting on external configuration changes. For example, this allows external components (registry, etcd, etc..) to interact with the Kubernetes API server configuration (KubeAPIServerConfig custom resource).

Currently changes in following external components are being observed:

  • host-etcd endpoints in kube-system namespace
    • The observed endpoint addresses are used to configure the storageConfig.urls in Kubernetes API server configuration.
  • cluster image.config.openshift.io custom resource
    • The observed CR resource is used to configure the imagePolicyConfig.internalRegistryHostname in Kubernetes API server configuration
  • cluster-config-v1 configmap in kube-system namespace
    • The observed configmap install-config is decoded and the networking.podCIDR and networking.serviceCIDR is extracted and used as input for admissionPluginConfig.openshift.io/RestrictedEndpointsAdmission.configuration.restrictedCIDRs and servicesSubnet

The configuration for the Kubernetes API server is the result of merging:

  • a default config
  • observed config (compare observed values above) spec.spec.unsupportedConfigOverrides from the kubeapiserveroperatorconfig.

All of these are sparse configurations, i.e. unvalidated json snippets which are merged in order to form a valid configuration at the end.

Debugging

Operator also expose events that can help debugging issues. To get operator events, run following command:

$ oc get events -n  openshift-cluster-kube-apiserver-operator

This operator is configured via KubeAPIServer custom resource:

$ oc describe kubeapiserver

apiVersion: operator.openshift.io/v1
kind: KubeAPIServer
metadata:
  name: cluster
spec:
  managementState: Managed

The current operator status is reported using the ClusterOperator resource. To get the current status you can run follow command:

$ oc get clusteroperator/kube-apiserver

Developing and debugging the operator

In the running cluster cluster-version-operator is responsible for maintaining functioning and non-altered elements. In that case to be able to use custom operator image one has to perform one of these operations:

  1. Set your operator in umanaged state, see here for details, in short:
oc patch clusterversion/version --type='merge' -p "$(cat <<- EOF
spec:
  overrides:
  - group: apps
    kind: Deployment
    name: kube-apiserver-operator
    namespace: openshift-kube-apiserver-operator
    unmanaged: true
EOF
)"
  1. Scale down cluster-version-operator:
oc scale --replicas=0 deploy/cluster-version-operator -n openshift-cluster-version

IMPORTANT: This approach disables cluster-version-operator completely, whereas the previous patch only tells it to not manage a kube-apiserver-operator!

After doing this you can now change the image of the operator to the desired one:

oc patch pod/kube-apiserver-operator-<rand_digits> -n openshift-kube-apiserver-operator -p '{"spec":{"containers":[{"name":"kube-apiserver-operator","image":"<user>/cluster-kube-apiserver-operator"}]}}'

Developing and debugging the bootkube bootstrap phase

The operator image version used by the https://github.com/openshift/installer/blob/master/pkg/asset/ignition/bootstrap/bootstrap.go#L178 bootstrap phase can be overridden by creating a custom origin-release image pointing to the developer's operator :latest image:

$ IMAGE_ORG=sttts make images
$ docker push sttts/origin-cluster-kube-apiserver-operator

$ cd ../cluster-kube-apiserver-operator
$ oc adm release new --from-release=registry.svc.ci.openshift.org/openshift/origin-release:v4.0 cluster-kube-apiserver-operator=docker.io/sttts/origin-cluster-kube-apiserver-operator:latest --to-image=sttts/origin-release:latest

$ cd ../installer
$ OPENSHIFT_INSTALL_RELEASE_IMAGE_OVERRIDE=docker.io/sttts/origin-release:latest bin/openshift-install cluster ...