- Documentation
- OKD
- Upgrading Clusters
- Upgrade methods and strategies
- About
- What's New?
- Getting Started
- Architecture
- Container Security Guide
- Installing Clusters
- Upgrading Clusters
- Configuring Clusters
- Overview
- Setting up the Registry
- Setting up a Router
- Deploying Red Hat CloudForms
- Prometheus Cluster Monitoring
- Accessing and Configuring the Red Hat Registry
- Master and Node Configuration
- OpenShift Ansible Broker Configuration
- Adding Hosts to an Existing Cluster
- Loading the Default Image Streams and Templates
- Configuring Custom Certificates
- Redeploying Certificates
- Configuring Authentication and User Agent
- Syncing groups with LDAP
- Configuring LDAP failover
- Configuring the SDN
- Configuring Nuage SDN
- Configuring NSX-T SDN
- Configuring Kuryr SDN
- Configuring for AWS
- Configuring for Red Hat Virtualization
- Configuring for OpenStack
- Configuring for Google Compute Engine
- Configuring for Azure
- Configuring for VMware vSphere
- Configuring Local Volumes
- Configuring Persistent Storage
- Overview
- Using NFS
- Using GlusterFS
- Using OpenStack Cinder
- Using Ceph RBD
- Using AWS Elastic Block Store
- Using GCE Persistent Disk
- Using iSCSI
- Using Fibre Channel
- Using Azure Disk
- Using Azure File
- Using FlexVolume
- Using VMware vSphere volumes for persistent storage
- Using Local Volume
- Using Container Storage Interface (CSI)
- Using OpenStack Manila shares
- Dynamic Provisioning and Creating Storage Classes
- Volume Security
- Selector-Label Volume Binding
- Enabling Controller-managed Attachment and Detachment
- Persistent Volume Snapshots
- Using hostPath
- Persistent Storage Examples
- Overview
- Sharing an NFS PV Across Two Pods
- Using Ceph RBD for Persistent Storage
- Using Ceph RBD for dynamic provisioning
- Complete Example Using GlusterFS
- Complete Example Using GlusterFS for Dynamic Provisioning
- Mounting Volumes To Privileged Pods
- Mount Propagation
- Switching an Integrated OpenShift Container Registry to GlusterFS
- Binding Persistent Volumes by Label
- Using StorageClasses for Dynamic Provisioning
- Using StorageClasses for Existing Legacy Storage
- Configuring Azure Blob Storage for Integrated Container Image Registry
- Configuring Ephemeral Storage
- Working with HTTP Proxies
- Configuring Global Build Defaults and Overrides
- Configuring Pipeline Execution
- Configuring Route Timeouts
- Configuring Native Container Routing
- Routing from Edge Load Balancers
- Aggregating Container Logs
- Aggregate Logging Sizing Guidelines
- Enabling Cluster Metrics
- Customizing the Web Console
- Deploying External Persistent Volume Provisioners
- Installing the Operator Framework (Technology Preview)
- Uninstalling Operator Lifecycle Manager
- Day Two Operations Guide
- Cluster Administration
- Overview
- Managing Nodes
- Restoring your cluster
- Replacing a master host
- Managing Users
- Managing Projects
- Managing Pods
- Managing Networking
- Configuring Service Accounts
- Managing Role-based Access Control
- Image Policy
- Image Signatures
- Scoped Tokens
- Monitoring Images
- Managing Security Context Constraints
- Scheduling
- Overview
- Default Scheduling
- Descheduling
- Custom Scheduling
- Controlling Pod Placement
- Pod Priority and Preemption
- Advanced Scheduling
- Advanced Scheduling and Node Affinity
- Advanced Scheduling and Pod Affinity/Anti-affinity
- Advanced Scheduling and Node Selectors
- Advanced Scheduling and Taints and Tolerations
- Setting Quotas
- Setting Multi-Project Quotas
- Pruning objects
- Extending the Kubernetes API with Custom Resources
- Garbage Collection
- Allocating Node Resources
- Overcommitting
- Out of Resource Handling
- Setting Limit Ranges
- Node Problem Detector
- Assigning Unique External IPs for Ingress Traffic
- Monitoring and Debugging Routers
- High Availability
- IPtables
- Securing Builds by Strategy
- Restricting Application Capabilities Using Seccomp
- Sysctls
- Encrypting Data at Datastore Layer
- Encrypting traffic between nodes with IPsec
- Building Dependency Trees
- Replacing a failed etcd member
- Restoring etcd quorum
- Troubleshooting Networking
- Diagnostics Tool
- Idling Applications
- Analyzing Cluster Capacity
- Configuring the cluster auto-scaler in AWS
- Disabling Features using Feature Gates
- Kuryr SDN Administration
- Scaling and Performance Guide
- Overview
- Recommended Installation Practices
- Recommended Host Practices
- Optimizing Compute Resources
- Optimizing Persistent Storage
- Optimizing Ephemeral Storage
- Network Optimization
- Routing Optimization
- Scaling Cluster Metrics
- Scaling Cluster Monitoring
- Tested Maximums per Cluster
- Using Cluster Loader
- Using CPU Manager
- Managing Huge Pages
- Optimizing On GlusterFS Storage
- Developer Guide
- Overview
- Application Life Cycle Management
- Authentication
- Authorization
- Projects
- Migrating Applications
- Tutorials
- Builds
- deployments
- Templates
- Opening a Remote Shell to Containers
- Service Accounts
- Managing Images
- Quotas and Limit Ranges
- Getting Traffic into a Cluster
- Routes
- Integrating External Services
- Using Device Manager
- Using Device Plug-ins
- Secrets
- ConfigMaps
- Downward API
- Projected Volumes
- Using Daemonsets
- Pod Autoscaling
- Managing Volumes
- Using Persistent Volumes
- Expanding Persistent Volumes
- Executing Remote Commands
- Copying Files
- Port Forwarding
- Shared Memory
- Application Health
- Events
- Managing Environment Variables
- Jobs
- OpenShift Pipeline
- Cron Jobs
- Create from URL
- Creating an object from a custom resource definition
- Application memory sizing
- Creating Images
- Using Images
- CLI Reference
- Ansible Playbook Bundle Development Guide
- Operators
- API reference
- API list
- Common object reference
- core
- About core
- Binding [core/v1]
- ComponentStatus [core/v1]
- ConfigMap [core/v1]
- Endpoints [core/v1]
- Event [core/v1]
- LimitRange [core/v1]
- Namespace [core/v1]
- Node [core/v1]
- PersistentVolumeClaim [core/v1]
- PersistentVolume [core/v1]
- Pod [core/v1]
- PodTemplate [core/v1]
- ReplicationController [core/v1]
- ResourceQuota [core/v1]
- Secret [core/v1]
- ServiceAccount [core/v1]
- Service [core/v1]
- admissionregistration.k8s.io
- apiregistration.k8s.io
- apps
- apps.openshift.io
- authentication.k8s.io
- authorization.k8s.io
- authorization.openshift.io
- About authorization.openshift.io
- ClusterRoleBinding [authorization.openshift.io/v1]
- ClusterRole [authorization.openshift.io/v1]
- LocalResourceAccessReview [authorization.openshift.io/v1]
- LocalSubjectAccessReview [authorization.openshift.io/v1]
- ResourceAccessReview [authorization.openshift.io/v1]
- RoleBindingRestriction [authorization.openshift.io/v1]
- RoleBinding [authorization.openshift.io/v1]
- Role [authorization.openshift.io/v1]
- SelfSubjectRulesReview [authorization.openshift.io/v1]
- SubjectAccessReview [authorization.openshift.io/v1]
- SubjectRulesReview [authorization.openshift.io/v1]
- autoscaling
- batch
- build.openshift.io
- certificates.k8s.io
- events.k8s.io
- image.openshift.io
- network.openshift.io
- networking.k8s.io
- oauth.openshift.io
- policy
- project.openshift.io
- quota.openshift.io
- rbac.authorization.k8s.io
- route.openshift.io
- scheduling.k8s.io
- security.openshift.io
- storage.k8s.io
- template.openshift.io
- user.openshift.io
- CRI-O Runtime
Upgrade methods and strategies
When new versions of OKD are released, you can upgrade your existing cluster to apply the latest enhancements and bug fixes. For OpenShift Origin, see the Releases page on GitHub to review the latest changes.
The OKD upgrade process uses Ansible playbooks to automate the tasks needed to upgrade your cluster. You must use the inventory file that you used during initial installation or during the last time that the upgrade was successful to run the upgrade playbook. Using this method allows you to choose between either upgrade strategy: in-place upgrades or blue-green deployments.
Unless noted otherwise, node and masters within a major version are forward and backward compatible across one minor version, so upgrading your cluster should go smoothly. However, you should not run mismatched versions longer than necessary to upgrade the entire cluster.
Before upgrading, ensure that all OKD services are running well. In the event of a control plane upgrade failure, check the versions of your masters to ensure that all versions are the same. If your masters are all the same version, re-run the upgrade. If they differ, downgrade the masters to match the lower versioned master, then re-run the upgrade.
Upgrade Strategies
There are two strategies you can take for performing the OKD cluster upgrade: in-place upgrades or blue-green deployments.
In-place Upgrades
With in-place upgrades, the cluster upgrade is performed on all hosts in a single, running cluster: first masters and then nodes. Pods are evacuated off of nodes and recreated on other running nodes before a node upgrade begins; this helps reduce downtime of user applications.
Blue-green deployments
The blue-green deployment upgrade method follows a similar flow to the in-place method: masters and etcd servers are still upgraded first, however a parallel environment is created for new nodes instead of upgrading them in-place.
This method allows administrators to switch traffic from the old set of nodes (e.g., the "blue" deployment) to the new set (e.g., the "green" deployment) after the new deployment has been verified. If a problem is detected, it is also then easy to rollback to the old deployment quickly.