Skip to content

Latest commit

 

History

History
85 lines (63 loc) · 4.84 KB

index.adoc

File metadata and controls

85 lines (63 loc) · 4.84 KB

Post-installation configuration overview

After installing {product-title}, a cluster administrator can configure and customize the following components:

  • Machine

  • Bare metal

  • Cluster

  • Node

  • Network

  • Storage

  • Users

  • Alerts and notifications

Performing post-installation configuration tasks

Cluster administrators can perform the following post-installation configuration tasks:

  • Configure operating system features: Machine Config Operator (MCO) manages MachineConfig objects. By using MCO, you can perform the following on an {product-title} cluster:

    • Configure nodes by using MachineConfig objects

    • Configure MCO-related custom resources

  • Configure bare metal nodes: The Bare Metal Operator (BMO) implements a Kubernetes API for managing bare metal hosts. It maintains an inventory of available bare metal hosts as instances of the BareMetalHost Custom Resource Definition (CRD). The Bare Metal Operator can:

    • Inspect the host’s hardware details and report them on the corresponding BareMetalHost. This includes information about CPUs, RAM, disks, NICs, and more.

    • Inspect the host’s firmware and configure BIOS settings.

    • Provision hosts with a desired image.

    • Clean a host’s disk contents before or after provisioning.

  • Configure cluster features: As a cluster administrator, you can modify the configuration resources of the major features of an {product-title} cluster. These features include:

    • Image registry

    • Networking configuration

    • Image build behavior

    • Identity provider

    • The etcd configuration

    • Machine set creation to handle the workloads

    • Cloud provider credential management

  • Configure cluster components to be private: By default, the installation program provisions {product-title} by using a publicly accessible DNS and endpoints. If you want your cluster to be accessible from within an internal network only, configure the following components to be private:

    • DNS

    • Ingress Controller

    • API server

  • Perform node operations: By default, {product-title} uses {op-system-first} compute machines. As a cluster administrator, you can perform the following operations with the machines in your {product-title} cluster:

    • Add and remove compute machines

    • Add and remove taints and tolerations to the nodes

    • Configure the maximum number of pods per node

    • Enable Device Manager

  • Configure network: After installing {product-title}, as a cluster administrator, you can configure the following:

    • Ingress cluster traffic

    • Node port service range

    • Network policy

    • Enabling the cluster-wide proxy

  • Configure storage: By default, containers operate using ephemeral storage or transient local storage. The ephemeral storage has a lifetime limitation, so you must configure persistent storage to store the data for a long time. You can configure storage by using one of the following methods:

    • Dynamic provisioning: You can dynamically provision storage on demand by defining and creating storage classes that control different levels of storage, including storage access.

    • Static provisioning: Cluster administrators can use Kubernetes persistent volumes to make existing storage available to a cluster by supporting various device configurations and mount options.

  • Configure users: OAuth access tokens allow users to authenticate themselves to the API. As a cluster administrator, you can configure OAuth to specify an identity provider, use role-based access control to define and apply permissions to users, and install an Operator from OperatorHub.

  • Manage alerts and notifications: As a cluster administrator, you can view firing alerts by default from the Alerting UI of the web console. You can also configure {product-title} to send alert notifications to external systems so that you learn about important issues with your cluster.