vertical-pod-autoscaler

Vertical Pod Autoscaler

Vertical Pod Autoscaler is a set of components that automatically adjust the amount of CPU and memory requested by pods running in the Kubernetes Cluster.

DISCLAIMER: This is an unofficial chart not supported by Vertical Pod Autoscaler authors.

TL;DR;

$ helm repo add cowboysysop https://cowboysysop.github.io/charts/
$ helm install my-release cowboysysop/vertical-pod-autoscaler

Introduction

This chart bootstraps a Vertical Pod Autoscaler deployment on a Kubernetes cluster using the Helm package manager.

Prerequisites

• Kubernetes >= 1.24
• Metrics Server >= 0.2 (you can use the bitnami/metrics-server chart)
• Helm >= 3.9

Installing

Install the chart using:

$ helm repo add cowboysysop https://cowboysysop.github.io/charts/
$ helm install my-release cowboysysop/vertical-pod-autoscaler

These commands deploy Vertical Pod Autoscaler on the Kubernetes cluster in the default configuration and with the release name my-release. The deployment configuration can be customized by specifying the customization parameters with the helm install command using the --values or --set arguments. Find more information in the configuration section of this document.

Upgrading

Upgrade the chart deployment using:

$ helm upgrade my-release cowboysysop/vertical-pod-autoscaler

The command upgrades the existing my-release deployment with the most latest release of the chart.

TIP: Use helm repo update to update information on available charts in the chart repositories.

Upgrading to version 9.0.0

The chart is now tested with Kubernetes >= 1.24 and Helm >= 3.9.

Future upgrades may introduce undetected breaking changes if you continue to use older versions.

Upgrading to version 8.0.0

Some parameters related to port management have been modified:

• Parameter admissionController.metrics.service.port has been renamed admissionController.metrics.service.ports.metrics.
• Parameter recommender.metrics.service.port has been renamed recommender.metrics.service.ports.metrics.
• Parameter updater.metrics.service.port has been renamed updater.metrics.service.ports.metrics.

Upgrading to version 7.0.0

Some parameters related to image management have been modified:

• Registry prefix in image.repository parameters is now configured in image.registry.
• Parameter imagePullSecrets has been renamed global.imagePullSecrets.

Upgrading to version 6.0.0

The application version is no more compatible with Kubernetes 1.19, 1.20 and 1.21.

Upgrading to version 5.0.0

The application validates that all fields that specify CPU and memory have supported resolution:

• CPU is a whole number of milli CPUs
• Memory is a whole number of bytes

Upgrading to version 4.0.0

The application version is no more compatible with Kubernetes 1.16.

Custom resource definitions are now created and upgraded with a pre-install/pre-upgrade job.

Upgrading to version 3.0.0

The chart is no more compatible with Helm 2.

Refer to the Helm documentation for more information.

Upgrading to version 2.0.0

The port names have been changed to be compatible with Istio service mesh.

Uninstalling

Uninstall the my-release deployment using:

$ helm uninstall my-release

The command deletes the release named my-release and frees all the kubernetes resources associated with the release.

TIP: Specify the --purge argument to the above command to remove the release from the store and make its name free for later use.

Delete the vpa-webhook-config mutating webhook configuration automatically created by Vertical Pod Autoscaler admission controller component using:

$ kubectl delete mutatingwebhookconfiguration vpa-webhook-config

Optionally, delete the custom resource definitions created by the chart using:

WARNING: It will also try to delete all instances of the custom resource definitions.

$ kubectl delete crd verticalpodautoscalers.autoscaling.k8s.io
$ kubectl delete crd verticalpodautoscalercheckpoints.autoscaling.k8s.io

Configuration

Global parameters

Name	Description	Default
global.imageRegistry	Global Docker image registry	""
global.imagePullSecrets	Global Docker registry secret names as an array	[]

Common parameters

Name	Description	Default
kubeVersion	Override Kubernetes version	""
nameOverride	Partially override vertical-pod-autoscaler.fullname template with a string (will prepend the release name)	""
fullnameOverride	Fully override vertical-pod-autoscaler.fullname template with a string	""
commonAnnotations	Annotations to add to all deployed objects	{}
commonLabels	Labels to add to all deployed objects	{}
extraDeploy	Array of extra objects to deploy with the release	[]

Admission controller parameters

Name	Description	Default
admissionController.enabled	Enable the component	true
admissionController.replicaCount	Number of replicas	1
admissionController.image.registry	Image registry	registry.k8s.io
admissionController.image.repository	Image repository	autoscaling/vpa-admission-controller
admissionController.image.tag	Image tag	1.2.1
admissionController.image.digest	Image digest	""
admissionController.image.pullPolicy	Image pull policy	IfNotPresent
admissionController.pdb.create	Specifies whether a pod disruption budget should be created	false
admissionController.pdb.minAvailable	Minimum number/percentage of pods that should remain scheduled	1
admissionController.pdb.maxUnavailable	Maximum number/percentage of pods that may be made unavailable	nil
admissionController.serviceAccount.create	Specifies whether a service account should be created	true
admissionController.serviceAccount.annotations	Service account annotations	{}
admissionController.serviceAccount.name	The name of the service account to use (Generated using the vertical-pod-autoscaler.fullname template if not set)	nil
admissionController.hostAliases	Pod host aliases	[]
admissionController.deploymentAnnotations	Additional deployment annotations	{}
admissionController.podAnnotations	Additional pod annotations	{}
admissionController.podLabels	Additional pod labels	{}
admissionController.podSecurityContext	Pod security context
admissionController.podSecurityContext.runAsNonRoot	Whether the container must run as a non-root user	true
admissionController.podSecurityContext.runAsUser	The UID to run the entrypoint of the container process	65534
admissionController.podSecurityContext.runAsGroup	The GID to run the entrypoint of the container process	65534
admissionController.hostNetwork	Use the host network	false
admissionController.priorityClassName	Priority class name	nil
admissionController.runtimeClassName	Runtime class name	""
admissionController.topologySpreadConstraints	Topology Spread Constraints for pod assignment	[]
admissionController.securityContext	Container security context	{}
admissionController.containerPorts.https	Container port for HTTPS	8000
admissionController.containerPorts.metrics	Container port for Metrics	8944
admissionController.livenessProbe.enabled	Enable liveness probe	true
admissionController.livenessProbe.initialDelaySeconds	Delay before the liveness probe is initiated	0
admissionController.livenessProbe.periodSeconds	How often to perform the liveness probe	10
admissionController.livenessProbe.timeoutSeconds	When the liveness probe times out	1
admissionController.livenessProbe.failureThreshold	Minimum consecutive failures for the liveness probe to be considered failed after having succeeded	3
admissionController.livenessProbe.successThreshold	Minimum consecutive successes for the liveness probe to be considered successful after having failed	1
admissionController.readinessProbe.enabled	Enable readiness probe	true
admissionController.readinessProbe.initialDelaySeconds	Delay before the readiness probe is initiated	0
admissionController.readinessProbe.periodSeconds	How often to perform the readiness probe	10
admissionController.readinessProbe.timeoutSeconds	When the readiness probe times out	1
admissionController.readinessProbe.failureThreshold	Minimum consecutive failures for the readiness probe to be considered failed after having succeeded	3
admissionController.readinessProbe.successThreshold	Minimum consecutive successes for the readiness probe to be considered successful after having failed	1
admissionController.startupProbe.enabled	Enable startup probe	false
admissionController.startupProbe.initialDelaySeconds	Delay before the startup probe is initiated	0
admissionController.startupProbe.periodSeconds	How often to perform the startup probe	10
admissionController.startupProbe.timeoutSeconds	When the startup probe times out	1
admissionController.startupProbe.failureThreshold	Minimum consecutive failures for the startup probe to be considered failed after having succeeded	3
admissionController.startupProbe.successThreshold	Minimum consecutive successes for the startup probe to be considered successful after having failed	1
admissionController.service.annotations	Service annotations	{}
admissionController.service.type	Service type	ClusterIP
admissionController.service.clusterIP	Static cluster IP address or None for headless service when service type is ClusterIP	nil
admissionController.service.sessionAffinity	Control where client requests go, to the same pod or round-robin	None
admissionController.service.sessionAffinityConfig	Additional settings for the sessionAffinity	{}
admissionController.service.ports.https	Service port for HTTPS (do not change it)	443
admissionController.resources	CPU/Memory resource requests/limits	{}
admissionController.nodeSelector	Node labels for pod assignment	{}
admissionController.tolerations	Tolerations for pod assignment	[]
admissionController.affinity	Map of node/pod affinities	{}
admissionController.extraArgs	Additional container arguments
admissionController.extraArgs.v	Number for the log level verbosity	2
admissionController.extraEnvVars	Additional container environment variables	[]
admissionController.extraEnvVarsCM	Name of existing ConfigMap containing additional container environment variables	nil
admissionController.extraEnvVarsSecret	Name of existing Secret containing additional container environment variables	nil
admissionController.extraVolumes	Optionally specify extra list of additional volumes	[]
admissionController.extraVolumeMounts	Optionally specify extra list of additional volumeMounts	[]
admissionController.metrics.service.annotations	Metrics service annotations	{}
admissionController.metrics.service.type	Metrics service type	ClusterIP
admissionController.metrics.service.clusterIP	Metrics static cluster IP address or None for headless service when service type is ClusterIP	nil
admissionController.metrics.service.ports.metrics	Metrics service port for Metrics	8944
admissionController.metrics.serviceMonitor.enabled	Specifies whether a service monitor should be created	false
admissionController.metrics.serviceMonitor.namespace	Namespace in which to create the service monitor	""
admissionController.metrics.serviceMonitor.annotations	Service monitor annotations	{}
admissionController.metrics.serviceMonitor.labels	Additional service monitor labels	{}
admissionController.metrics.serviceMonitor.jobLabel	The name of the label on the target service to use as the job name in Prometheus	""
admissionController.metrics.serviceMonitor.honorLabels	Whether to choose the metric’s labels on collisions with target labels	false
admissionController.metrics.serviceMonitor.interval	Interval at which metrics should be scraped	""
admissionController.metrics.serviceMonitor.scrapeTimeout	Timeout after which the scrape is ended	""
admissionController.metrics.serviceMonitor.metricRelabelings	Specify additional relabeling of metrics	[]
admissionController.metrics.serviceMonitor.relabelings	Specify general relabeling	[]
admissionController.tls.caCert	TLS CA certificate (Generated using the genCA function if not set)	""
admissionController.tls.cert	TLS certificate (Generated using the genSignedCert function if not set)	""
admissionController.tls.key	TLS private key (Generated using the genSignedCert function if not set)	""
admissionController.tls.existingSecret	Name of existing TLS Secret to use	""

Recommender parameters

Name	Description	Default
recommender.replicaCount	Number of replicas	1
recommender.image.registry	Image registry	registry.k8s.io
recommender.image.repository	Image repository	autoscaling/vpa-recommender
recommender.image.tag	Image tag	1.2.1
recommender.image.digest	Image digest	""
recommender.image.pullPolicy	Image pull policy	IfNotPresent
recommender.pdb.create	Specifies whether a pod disruption budget should be created	false
recommender.pdb.minAvailable	Minimum number/percentage of pods that should remain scheduled	1
recommender.pdb.maxUnavailable	Maximum number/percentage of pods that may be made unavailable	nil
recommender.serviceAccount.create	Specifies whether a service account should be created	true
recommender.serviceAccount.annotations	Service account annotations	{}
recommender.serviceAccount.name	The name of the service account to use (Generated using the vertical-pod-autoscaler.fullname template if not set)	nil
recommender.hostAliases	Pod host aliases	[]
recommender.deploymentAnnotations	Additional deployment annotations	{}
recommender.podAnnotations	Additional pod annotations	{}
recommender.podLabels	Additional pod labels	{}
recommender.podSecurityContext	Pod security context
recommender.podSecurityContext.runAsNonRoot	Whether the container must run as a non-root user	true
recommender.podSecurityContext.runAsUser	The UID to run the entrypoint of the container process	65534
recommender.podSecurityContext.runAsGroup	The GID to run the entrypoint of the container process	65534
recommender.priorityClassName	Priority class name	nil
recommender.runtimeClassName	Runtime class name	""
recommender.topologySpreadConstraints	Topology Spread Constraints for pod assignment	[]
recommender.securityContext	Container security context	{}
recommender.containerPorts.metrics	Container port for Metrics	8942
recommender.livenessProbe.enabled	Enable liveness probe	true
recommender.livenessProbe.initialDelaySeconds	Delay before the liveness probe is initiated	0
recommender.livenessProbe.periodSeconds	How often to perform the liveness probe	10
recommender.livenessProbe.timeoutSeconds	When the liveness probe times out	1
recommender.livenessProbe.failureThreshold	Minimum consecutive failures for the liveness probe to be considered failed after having succeeded	3
recommender.livenessProbe.successThreshold	Minimum consecutive successes for the liveness probe to be considered successful after having failed	1
recommender.readinessProbe.enabled	Enable readiness probe	true
recommender.readinessProbe.initialDelaySeconds	Delay before the readiness probe is initiated	0
recommender.readinessProbe.periodSeconds	How often to perform the readiness probe	10
recommender.readinessProbe.timeoutSeconds	When the readiness probe times out	1
recommender.readinessProbe.failureThreshold	Minimum consecutive failures for the readiness probe to be considered failed after having succeeded	3
recommender.readinessProbe.successThreshold	Minimum consecutive successes for the readiness probe to be considered successful after having failed	1
recommender.startupProbe.enabled	Enable startup probe	false
recommender.startupProbe.initialDelaySeconds	Delay before the startup probe is initiated	0
recommender.startupProbe.periodSeconds	How often to perform the startup probe	10
recommender.startupProbe.timeoutSeconds	When the startup probe times out	1
recommender.startupProbe.failureThreshold	Minimum consecutive failures for the startup probe to be considered failed after having succeeded	3
recommender.startupProbe.successThreshold	Minimum consecutive successes for the startup probe to be considered successful after having failed	1
recommender.resources	CPU/Memory resource requests/limits	{}
recommender.nodeSelector	Node labels for pod assignment	{}
recommender.tolerations	Tolerations for pod assignment	[]
recommender.affinity	Map of node/pod affinities	{}
recommender.extraArgs	Additional container arguments
recommender.extraArgs.v	Number for the log level verbosity	2
recommender.extraEnvVars	Additional container environment variables	[]
recommender.extraEnvVarsCM	Name of existing ConfigMap containing additional container environment variables	nil
recommender.extraEnvVarsSecret	Name of existing Secret containing additional container environment variables	nil
recommender.extraVolumes	Optionally specify extra list of additional volumes	[]
recommender.extraVolumeMounts	Optionally specify extra list of additional volumeMounts	[]
recommender.metrics.service.annotations	Metrics service annotations	{}
recommender.metrics.service.type	Metrics service type	ClusterIP
recommender.metrics.service.clusterIP	Metrics static cluster IP address or None for headless service when service type is ClusterIP	nil
recommender.metrics.service.ports.metrics	Metrics service port for Metrics	8942
recommender.metrics.serviceMonitor.enabled	Specifies whether a service monitor should be created	false
recommender.metrics.serviceMonitor.namespace	Namespace in which to create the service monitor	""
recommender.metrics.serviceMonitor.annotations	Service monitor annotations	{}
recommender.metrics.serviceMonitor.labels	Additional service monitor labels	{}
recommender.metrics.serviceMonitor.jobLabel	The name of the label on the target service to use as the job name in Prometheus	""
recommender.metrics.serviceMonitor.honorLabels	Whether to choose the metric’s labels on collisions with target labels	false
recommender.metrics.serviceMonitor.interval	Interval at which metrics should be scraped	""
recommender.metrics.serviceMonitor.scrapeTimeout	Timeout after which the scrape is ended	""
recommender.metrics.serviceMonitor.metricRelabelings	Specify additional relabeling of metrics	[]
recommender.metrics.serviceMonitor.relabelings	Specify general relabeling	[]

Updater parameters

Name	Description	Default
updater.enabled	Enable the component	true
updater.replicaCount	Number of replicas	1
updater.image.registry	Image registry	registry.k8s.io
updater.image.repository	Image repository	autoscaling/vpa-updater
updater.image.tag	Image tag	1.2.1
updater.image.digest	Image digest	""
updater.image.pullPolicy	Image pull policy	IfNotPresent
updater.pdb.create	Specifies whether a pod disruption budget should be created	false
updater.pdb.minAvailable	Minimum number/percentage of pods that should remain scheduled	1
updater.pdb.maxUnavailable	Maximum number/percentage of pods that may be made unavailable	nil
updater.serviceAccount.create	Specifies whether a service account should be created	true
updater.serviceAccount.annotations	Service account annotations	{}
updater.serviceAccount.name	The name of the service account to use (Generated using the vertical-pod-autoscaler.fullname template if not set)	nil
updater.hostAliases	Pod host aliases	[]
updater.deploymentAnnotations	Additional deployment annotations	{}
updater.podAnnotations	Additional pod annotations	{}
updater.podLabels	Additional pod labels	{}
updater.podSecurityContext	Pod security context
updater.podSecurityContext.runAsNonRoot	Whether the container must run as a non-root user	true
updater.podSecurityContext.runAsUser	The UID to run the entrypoint of the container process	65534
updater.podSecurityContext.runAsGroup	The GID to run the entrypoint of the container process	65534
updater.priorityClassName	Priority class name	nil
updater.runtimeClassName	Runtime class name	""
updater.topologySpreadConstraints	Topology Spread Constraints for pod assignment	[]
updater.securityContext	Container security context	{}
updater.containerPorts.metrics	Container port for Metrics	8943
updater.livenessProbe.enabled	Enable liveness probe	true
updater.livenessProbe.initialDelaySeconds	Delay before the liveness probe is initiated	0
updater.livenessProbe.periodSeconds	How often to perform the liveness probe	10
updater.livenessProbe.timeoutSeconds	When the liveness probe times out	1
updater.livenessProbe.failureThreshold	Minimum consecutive failures for the liveness probe to be considered failed after having succeeded	3
updater.livenessProbe.successThreshold	Minimum consecutive successes for the liveness probe to be considered successful after having failed	1
updater.readinessProbe.enabled	Enable readiness probe	true
updater.readinessProbe.initialDelaySeconds	Delay before the readiness probe is initiated	0
updater.readinessProbe.periodSeconds	How often to perform the readiness probe	10
updater.readinessProbe.timeoutSeconds	When the readiness probe times out	1
updater.readinessProbe.failureThreshold	Minimum consecutive failures for the readiness probe to be considered failed after having succeeded	3
updater.readinessProbe.successThreshold	Minimum consecutive successes for the readiness probe to be considered successful after having failed	1
updater.startupProbe.enabled	Enable startup probe	false
updater.startupProbe.initialDelaySeconds	Delay before the startup probe is initiated	0
updater.startupProbe.periodSeconds	How often to perform the startup probe	10
updater.startupProbe.timeoutSeconds	When the startup probe times out	1
updater.startupProbe.failureThreshold	Minimum consecutive failures for the startup probe to be considered failed after having succeeded	3
updater.startupProbe.successThreshold	Minimum consecutive successes for the startup probe to be considered successful after having failed	1
updater.resources	CPU/Memory resource requests/limits	{}
updater.nodeSelector	Node labels for pod assignment	{}
updater.tolerations	Tolerations for pod assignment	[]
updater.affinity	Map of node/pod affinities	{}
updater.extraArgs	Additional container arguments
updater.extraArgs.v	Number for the log level verbosity	2
updater.extraEnvVars	Additional container environment variables	[]
updater.extraEnvVarsCM	Name of existing ConfigMap containing additional container environment variables	nil
updater.extraEnvVarsSecret	Name of existing Secret containing additional container environment variables	nil
updater.extraVolumes	Optionally specify extra list of additional volumes	[]
updater.extraVolumeMounts	Optionally specify extra list of additional volumeMounts	[]
updater.metrics.service.annotations	Metrics service annotations	{}
updater.metrics.service.type	Metrics service type	ClusterIP
updater.metrics.service.clusterIP	Metrics static cluster IP address or None for headless service when service type is ClusterIP	nil
updater.metrics.service.ports.metrics	Metrics service port for Metrics	8943
updater.metrics.serviceMonitor.enabled	Specifies whether a service monitor should be created	false
updater.metrics.serviceMonitor.namespace	Namespace in which to create the service monitor	""
updater.metrics.serviceMonitor.annotations	Service monitor annotations	{}
updater.metrics.serviceMonitor.labels	Additional service monitor labels	{}
updater.metrics.serviceMonitor.jobLabel	The name of the label on the target service to use as the job name in Prometheus	""
updater.metrics.serviceMonitor.honorLabels	Whether to choose the metric’s labels on collisions with target labels	false
updater.metrics.serviceMonitor.interval	Interval at which metrics should be scraped	""
updater.metrics.serviceMonitor.scrapeTimeout	Timeout after which the scrape is ended	""
updater.metrics.serviceMonitor.metricRelabelings	Specify additional relabeling of metrics	[]
updater.metrics.serviceMonitor.relabelings	Specify general relabeling	[]

CRDs parameters

Name	Description	Default
crds.image.registry	Image registry	docker.io
crds.image.repository	Image repository	bitnami/kubectl
crds.image.tag	Image tag	1.29.3
crds.image.digest	Image digest	""
crds.image.pullPolicy	Image pull policy	IfNotPresent
crds.podAnnotations	Additional pod annotations	{}
crds.podLabels	Additional pod labels	{}
crds.podSecurityContext	Pod security context
crds.podSecurityContext.runAsNonRoot	Whether the container must run as a non-root user	true
crds.podSecurityContext.runAsUser	The UID to run the entrypoint of the container process	1001
crds.podSecurityContext.runAsGroup	The GID to run the entrypoint of the container process	1001
crds.securityContext	Container security context	{}
crds.resources	CPU/Memory resource requests/limits	{}
crds.nodeSelector	Node labels for pod assignment	{}
crds.tolerations	Tolerations for pod assignment	[]
crds.affinity	Map of node/pod affinities	{}

Tests parameters

Name	Description	Default
tests.image.registry	Image registry	ghcr.io
tests.image.repository	Image repository	cowboysysop/pytest
tests.image.tag	Image tag	1.0.41
tests.image.digest	Image digest	""
tests.image.pullPolicy	Image pull policy	IfNotPresent

Setting parameters

Specify the parameters you which to customize using the --set argument to the helm install command. For instance,

$ helm install my-release \
    --set nameOverride=my-name cowboysysop/vertical-pod-autoscaler

The above command sets the nameOverride to my-name.

Alternatively, a YAML file that specifies the values for the above parameters can be provided while installing the chart. For example,

$ helm install my-release \
    --values values.yaml cowboysysop/vertical-pod-autoscaler

TIP: You can use the default values.yaml.

Limitations

Due to hard-coded values in Vertical Pod Autoscaler, the chart configuration has some limitations:

• Admission controller component service name is vpa-webhook
• Admission controller component service port is 443
• Mutating webhook configuration name automatically created by admission controller component is vpa-webhook-config

License

The source code of this chart is under MIT License.

It also uses source code under Apache 2.0 License from the Bitnami repository.