Openshift TOC

Openshift TOC

OpenShift-customized-Training

Prerequisites

  • Aws/Azure Account
  • Resources used
    • 3 LoadBalancers
    • Route 53
    • AWS s3 bucket
    • 1 VPC
    • 5 Ip addresses
    • NAT gateways
    • 8 VMS with 4 cpu and 8 GB of RAM each

Duration:

OpenShift Container Platform architecture

  • Overview of Red Hat Enterprise Linux CoreOS (RHCOS)

  • Crio OverView

  • Podman Overview

  • Overview of journactl

Hands on Lab

  • Crio
  • Podman
  • Journalctl

Installation and update

  • Understand the underline infrstracture/resources requirements.
  • Know Quay.io
  • Know Redhat Registry
  • know ignition files
  • Installation with IPI
  • Installation with user-provisioned infrastructure
  • Installation on vmware(optional)
  • Multi tenant Openshift Installation
  • Configure Network Policy
  • Installation with customized network plugins
  • Troubleshooting installation issues
    • Gathering logs from a failed installation
    • Manually gathering logs with SSH access to your host(s)
    • Manually gathering logs without SSH access to your host(s)
    • Getting debug information from the installation program

Post_installation_configuration This task will take most of the time

  • Configuration of Authentication with Htpasswd

  • Configuration of Authentication with Azure AD

  • Remove the default virtual admin user (kubeadmin)

  • Secure Api with ssl certificate

  • Secure Route with Route

  • Setting the Ingress Controller

  • Restricting the API server to private

  • Configuration Default Quota project template

  • Configure default limits

  • Restrict user for LoadBalancer service.

  • Configure Alert Manager

  • Updating the global cluster pull secret

  • Configure Autoscaling for nodes

  • Create infrastracture nodes

  • Move all infr related services to infra nodes

  • OpenShift Internal Registry

  • Router pods

  • Monitoring pods

  • Logging pods

OpenShift Backup and DR

  • Installation and Configuration of Kasten/Velero
  • Setup the backup of etcd
  • Recovering from the etcd backup

Post-installation node tasks

  • Adding RHEL compute nodes to a cluster if needed
  • Configuring Machine health checks
  • Limitations when deploying machine health checks
  • Node host best practices
  • Configure different type of profile
  • Updating ssh keys for master and worker nodes

Post-installation network configuration

  • Enabling the cluster-wide proxy
  • Configuring ingress cluster traffic
  • Configuring network policy
  • Configuring multitenant isolation by using network policy

Post-installation storage configuration

  • Dynamic provisioning
  • Defining a storage class
  • Using Azure file for RWX
  • Installation and configuration cephcluster with rook operator to achieve below:
  • Block storage
  • File storage
  • Object Storage

Know OpenShift Internal Registry

  • Configuring additional trust stores for image registry access
  • Configuring storage credentials for the Image Registry Operator

OpenShift Scc

  • Understanding default scc
  • Creating and user custom scc

Pod Scheduling*

  • Default scheduling
  • Infrastructure Topological Levels
  • Affinity
  • Anti Affinity
  • Advanced scheduling
    • Pod Affinity and Anti-affinity
    • Node Affinity
    • Node Selectors
    • Taints and Tolerations
  • Custom scheduling
  • Deploying the Scheduler

Troubleshoot

  • Pod related issues

    • Router/Registry Not deploying to correct node
    • Registry not showing contents of NFS mount (persistent volume)
    • Hosts Can No Longer Resolve Each Other During Anisble Install
    • Failure to deploy registry (permissions issues)
    • Application Pod fails to deploy

  • Issues with Nodes

    • Nodes being reported as ready, but builds failing
    • Node reporting NotReady
    • Nodes report ready but ETCD health check fails
    • Atomic-openshift-node service fails to start

  • Registry issues

    • OpenShift builds fail trying to push image using a wrong IP address for the registry
    • OpenShift build error: failed to push image while using NFS persistent storage
    • Failure to push image to OpenShift’s Registry when backed by shared storage

  • Quotas and Limitranges

    • Must make a non-zero request for cpu

  • Installation Fails…​

    • Web Console Public URL on a different Port
    • UI Redirecting to the URL of the masters instead of the LB
    • Intermittent Login issues (htpasswd)
    • Build Issues
    • oc new-app runs s2i instead of Docker build
    • Binary Build Fails, citing “BadRequest”

  • Issues related to Identity

    • user is unable to login
    • user has two identities
    • How to impersonate user
    • login with service account

Migration from Ocp3 to ocp 4