apiVersion: keda.sh/v1alpha1
kind: ScaledObject
metadata:
annotations:
autoscaling.keda.sh/paused-replicas: "4"
# ...- Documentation
- OpenShift Dedicated
- Nodes
- Automatically scaling pods with the Custom Metrics Autoscaler Operator
- Pausing the custom metrics autoscaler history bug_report picture_as_pdf
OpenShift docs are moving and will soon only be available at docs.redhat.com, the home of all Red Hat product documentation. Explore the new docs experience today.
- About
- What's new
- Introduction to OpenShift Dedicated
- Architecture
- Red Hat OpenShift Cluster Manager
- Planning your environment
- Getting started
- Installing, accessing, and deleting OpenShift Dedicated clusters
- Support
- Web console
- CLI tools
- Cluster administration
- Security and compliance
- Authentication and authorization
- Authentication and authorization overview
- Understanding authentication
- Managing user-owned OAuth access tokens
- Configuring identity providers
- Revoking privileges and access to an OpenShift Dedicated cluster
- Managing administration roles and users
- Using RBAC to define and apply permissions
- Understanding and creating service accounts
- Using service accounts in applications
- Using a service account as an OAuth client
- Scoping tokens
- Using bound service account tokens
- Managing security context constraints
- Understanding and managing pod security admission
- Syncing LDAP groups
- Upgrading
- CI/CD
- Images
- Overview of images
- Overview of the Cluster Samples Operator
- Using the Cluster Samples Operator with an alternate registry
- Creating images
- Managing images
- Managing image streams
- Using image streams with Kubernetes resources
- Triggering updates on image stream changes
- Image configuration resources
- Using templates
- Using Ruby on Rails
- Using images
- Add-on services
- Storage
- Registry
- Operators
- Operators overview
- Understanding Operators
- User tasks
- Administrator tasks
- Developing Operators
- About the Operator SDK
- Installing the Operator SDK CLI
- Go-based Operators
- Ansible-based Operators
- Helm-based Operators
- Defining cluster service versions (CSVs)
- Working with bundle images
- Complying with pod security admission
- Validating Operators using the scorecard
- Validating Operator bundles
- High-availability or single-node cluster detection and support
- Configuring built-in monitoring with Prometheus
- Configuring leader election
- Object pruning utility
- Migrating package manifest projects to bundle format
- Operator SDK CLI reference
- Migrating to Operator SDK v0.1.0
- Networking
- Building applications
- Building applications overview
- Projects
- Creating applications
- Viewing application composition using the Topology view
- Working with Helm charts
- Deployments
- Quotas
- Using config maps with applications
- Monitoring project and application metrics using the Developer perspective
- Monitoring application health
- Editing applications
- Working with quotas
- Pruning objects to reclaim resources
- Idling applications
- Deleting applications
- Using the Red Hat Marketplace
- Nodes
- Overview of nodes
- Working with pods
- Automatically scaling pods with the Custom Metrics Autoscaler Operator
- Release notes
- Custom Metrics Autoscaler Operator overview
- Installing the custom metrics autoscaler
- Understanding the custom metrics autoscaler triggers
- Understanding the custom metrics autoscaler trigger authentications
- Pausing the custom metrics autoscaler
- Gathering audit logs
- Gathering debugging data
- Viewing Operator metrics
- Understanding how to add custom metrics autoscalers
- Removing the Custom Metrics Autoscaler Operator
- Controlling pod placement onto nodes (scheduling)
- About pod placement using the scheduler
- Placing pods relative to other pods using pod affinity and anti-affinity rules
- Controlling pod placement on nodes using node affinity rules
- Placing pods onto overcommited nodes
- Placing pods on specific nodes using node selectors
- Controlling pod placement using pod topology spread constraints
- Using jobs and daemon sets
- Working with nodes
- Working with containers
- Understanding containers
- Using Init Containers to perform tasks before a pod is deployed
- Using volumes to persist container data
- Mapping volumes using projected volumes
- Allowing containers to consume API objects
- Copying files to or from a container
- Executing remote commands in a container
- Using port forwarding to access applications in a container
- Working with clusters
- Observability
- Observability overview
- Monitoring
- Monitoring overview
- Accessing monitoring for user-defined projects
- Configuring the monitoring stack
- Disabling monitoring for user-defined projects
- Enabling alert routing for user-defined projects
- Managing metrics
- Managing alerts
- Reviewing monitoring dashboards
- Accessing third-party monitoring APIs
- Troubleshooting monitoring issues
- Config map reference for the Cluster Monitoring Operator
- Service Mesh
- Service Mesh 2.x
- About OpenShift Service Mesh
- Service Mesh 2.x release notes
- Service Mesh architecture
- Service Mesh deployment models
- Service Mesh and Istio differences
- Preparing to install Service Mesh
- Installing the Operators
- Creating the ServiceMeshControlPlane
- Adding workloads to a service mesh
- Enabling sidecar injection
- Upgrading Service Mesh
- Managing users and profiles
- Security
- Traffic management
- Metrics, logs, and traces
- Performance and scalability
- Deploying to production
- Federation
- Extensions
- 3scale WebAssembly for 2.1
- 3scale Istio adapter for 2.0
- Troubleshooting Service Mesh
- Control plane configuration reference
- Kiali configuration reference
- Jaeger configuration reference
- Uninstalling Service Mesh
- Service Mesh 2.x
- Serverless
×
Pausing the custom metrics autoscaler for a scaled object
You can pause and restart the autoscaling of a workload, as needed.
For example, you might want to pause autoscaling before performing cluster maintenance or to avoid resource starvation by removing non-mission-critical workloads.
Pausing a custom metrics autoscaler
You can pause the autoscaling of a scaled object by adding the autoscaling.keda.sh/paused-replicas annotation to the custom metrics autoscaler for that scaled object. The custom metrics autoscaler scales the replicas for that workload to the specified value and pauses autoscaling until the annotation is removed.
Procedure
-
Use the following command to edit the
ScaledObjectCR for your workload:$ oc edit ScaledObject scaledobject -
Add the
autoscaling.keda.sh/paused-replicasannotation with any value:apiVersion: keda.sh/v1alpha1 kind: ScaledObject metadata: annotations: autoscaling.keda.sh/paused-replicas: "4" (1) creationTimestamp: "2023-02-08T14:41:01Z" generation: 1 name: scaledobject namespace: my-project resourceVersion: '65729' uid: f5aec682-acdf-4232-a783-58b5b82f5dd01 Specifies that the Custom Metrics Autoscaler Operator is to scale the replicas to the specified value and stop autoscaling.
Restarting the custom metrics autoscaler for a scaled object
You can restart a paused custom metrics autoscaler by removing the autoscaling.keda.sh/paused-replicas annotation for that ScaledObject.
apiVersion: keda.sh/v1alpha1
kind: ScaledObject
metadata:
annotations:
autoscaling.keda.sh/paused-replicas: "4"
# ...Procedure
-
Use the following command to edit the
ScaledObjectCR for your workload:$ oc edit ScaledObject scaledobject -
Remove the
autoscaling.keda.sh/paused-replicasannotation.apiVersion: keda.sh/v1alpha1 kind: ScaledObject metadata: annotations: autoscaling.keda.sh/paused-replicas: "4" (1) creationTimestamp: "2023-02-08T14:41:01Z" generation: 1 name: scaledobject namespace: my-project resourceVersion: '65729' uid: f5aec682-acdf-4232-a783-58b5b82f5dd01 Remove this annotation to restart a paused custom metrics autoscaler.