Helm values and component configuration

Helm value paths are relative to a chart. A bundled subchart value and a standalone chart value can have different prefixes. Match the selected chart version before editing configuration; the Agent’s collection flag does not install every controller it can observe.

Configuration ownership

Use Install with Helm for core install and Argo CD/Terraform/ARM variants. When the Agent manages a component, KedifyConfiguration describes its desired installation state. Keep Helm/GitOps and that resource’s ownership consistent; a direct change to a managed dependent release can be reverted.

Agent feature and collection options

These feature paths can describe a newer release than the Agent v0.7.0 defaults below. Use the values for your installed chart version.

Full set of configuration options is available in Kedify Agent chart values. Here are some notable options:

• agent.features.podResourceProfilesEnabled: Enables the controller for PodResourceProfile to be started by Kedify Agent. The feature is further described on its own docs. By default, it is true. Env var equivalent: PRP_ENABLED.

• agent.features.prpRequiresAnnotatedPods: If enabled, the controller for PodResourceProfile requires the target pods (where container resources should be modified) to have annotation prp.kedify.io/reconcile=enabled on them. This is a performance optimization so that more pod events can be ignored by the controller. Default: true. Env var equivalent: PRP_REQUIRES_ANNOTATED_PODS.

• agent.features.podResourceAutoscalersEnabled: Enables the controller for Pod Resource Autoscaler. Default is false; set it to true to enable PRA. Env var equivalent: PRA_ENABLED.

• global.features.multicluster.enabled: Enables the selected Multi-Cluster Scaling model. With type: distributed, it enables the Agent DSO/DSJ controllers and bundled KEDA raw metrics gRPC. With type: dedicated, it enables remote reconciliation in bundled KEDA/KPA and disables Agent DSO/DSJ controllers.

• global.features.multicluster.type: distributed (default) or dedicated. Dedicated mode requires keda.enabled=true and kpa.enabled=true, selects KPA as the default autoscaler, and consumes ready ClusterRegistration resources and their referenced Secrets. Leave namespace watch restrictions and KEDA label sharding disabled in the dedicated controller releases. See dedicated setup.

• agent.features.multicluster.enabled: Enables only the distributed Agent controllers and cannot configure bundled KEDA/KPA. It is not a dedicated-mode selector. The previous flat distributedScaledObjectsEnabled / distributedScaledJobsEnabled switches remain supported during the deprecation period; they must be disabled with dedicated mode.

• global.features.multicluster.vClusterDiscoveryEnabled: Discovers hosted vClusters through their native kubeconfig Secrets. In the 0.8.1 bundle it also installs the five live KEDA/KPA CRDs into members and creates managed kubeconfig Secrets and ClusterRegistration resources after bootstrap. All five source CRDs, including KPA’s, must exist centrally. This grants cluster-wide read access to Secrets/Services, source CRD read access, and registration management permissions in the Agent namespace. The equivalent agent.features.multicluster.vClusterDiscoveryEnabled setting is also supported; discovery does not itself select a scaling model. See vCluster setup.

• agent.multicluster.localCluster.enabled: Registers the management cluster itself as an optional member. In 0.8.1, the chart creates a ClusterRegistration and the Agent materializes its referenced kubeconfig Secret. Local registration is opt-in; dedicated remote controllers do not automatically reconcile management-cluster workloads.

• agent.features.scalingPoliciesEnabled: Enables the controller for Scaling Policy. Default is true; set it to false to disable the controller. Env var equivalent: SCALING_POLICY_ENABLED.

• agent.features.kedifyPodAutoscalerEnabled: Enables KPA telemetry and Dashboard support. Default is false. A compatible Agent, Dashboard, KEDA, and KPA release set is required.

• agent.features.scaleAdaptersEnabled: Enables the controller for ScaleAdapter, which lets KEDA scale resources with an incomplete /scale subresource, such as Agones Fleets, or with no /scale subresource at all. Default is false; the agent additionally needs RBAC through agent.extraRbacRules: get and update on the target kinds’ /scale subresource, or get, list, watch and update on the whole resource for targets adapted through replica field paths. Env var equivalent: SCALE_ADAPTER_ENABLED.

• agent.features.recommendationsForLabeledNamespaces: Limits Kubernetes container utilization scraping to opted-in namespaces. Default: true. Env var equivalent: METRICS_FOR_LABELED_NAMESPACES_ENABLED.

• agent.metricsWatchNamespace: Comma-separated list of namespaces where Kubernetes container utilization metrics should be collected. When agent.features.recommendationsForLabeledNamespaces=true, explicit namespaces in this list are added to the labeled namespaces. When label-gated collection is disabled, an unset or empty value means all namespaces. Env var equivalent: METRICS_WATCH_NAMESPACE.

• agent.kedifyProxy.clusterWide: Enables the cluster-wide deployment of kedify-proxy, shared across all namespaces in the cluster. This variable should be set to true to configure a single instance of kedify-proxy in the KEDA installation namespace for centralized scaling. The cluster-wide setup is supported only for configurations using Istio’s VirtualService. For more details, see Kedify Proxy.

• agent.kedifyProxy.globalValues: Set of options to configure kedify-proxy deployments. See Kedify Proxy helm chart for more details. This can also be fine-tuned per namespace through agent.kedifyProxy.namespacedValues.${NAMESPACE}.

Read the values for your installed release

helm list --all-namespaces
helm get values RELEASE -n NAMESPACE -a
helm show values kedifykeda/kedify-agent --version CHART_VERSION

Treat live values as sensitive: they may contain credentials. Compare the selected chart’s schema/defaults with the deployed values before changing them. Availability records prerequisites; Insights settings explains namespace selection; observation explains native and existing-stack telemetry.

Agent chart v0.7.0 snapshot

Chart-relative path	Type	Snapshot default
clusterName	string	"cluster-1"
disableSchemaValidation	boolean	false
global.features.podResourceProfilesDownscaleGuardEnabled	boolean	true
global.features.podResourceProfilesEnabled	boolean	true
global.features.podResourceAutoscalersEnabled	boolean	false
global.features.argoRolloutsEnabled	boolean	false
global.features.prpRequiresAnnotatedPods	boolean	true
global.features.praRequiresAnnotatedPods	boolean	false
global.features.distributedScaledObjectsEnabled	boolean	false
global.features.distributedScaledJobsEnabled	boolean	false
global.features.scaleAdaptersEnabled	boolean	false
global.features.scalingPoliciesEnabled	boolean	true
global.features.recommendationsForLabeledNamespaces	boolean	true
global.features.multitenantKEDAEnabled	boolean	false
global.features.kedifyPodAutoscalerEnabled	boolean	false
agent.replicas	number	1
agent.orgId	string	""
agent.apiKey	string	""
agent.createApiKeySecret	boolean	true
agent.agentId	string	""
agent.kedifyServer	string	"service.kedify.io:443"
agent.cloudAccountID	string	""
agent.logging.level	string	"info"
agent.logging.format	string	"console"
agent.logging.timeEncoding	string	"RFC3339"
agent.logging.stackTracesEnabled	boolean	false
agent.kedifyProxy.clusterWide	boolean	false
agent.kedifyProxy.chart	string	"/helm/kedify/kedify-proxy.tgz"
agent.kedifyProxy.chartVersion	string	"v0.0.12"
agent.kedifyProxy.proxyConfigmapWatcherEnabled	boolean	true
agent.kedifyProxy.globalValues.logging.format	string	"plaintext"
agent.kedifyProxy.globalEnvoyConfigs.route.idle_timeout	string	"3600s"
agent.kedifyProxy.globalEnvoyConfigs.cluster.lb_policy	string	"ROUND_ROBIN"
agent.kedifyProxy.namespacedValues	object	{}
agent.kedifyProxy.namespacedEnvoyConfigs	object	{}
agent.createCrds	boolean	true
agent.createKedifyConfiguration	boolean	true
agent.kedifyConfiguration.agentAutoUpdate	boolean	false
agent.kedifyConfiguration.mode	string	"Helm"
agent.kedifyConfiguration.httpAddonHelm.enabled	boolean	true
agent.kedifyConfiguration.predictorAddonHelm.enabled	boolean	true
agent.kedifyConfiguration.otelAddonHelm.enabled	boolean	true
agent.kedifyConfiguration.kedaHelm.enabled	boolean	true
agent.kedifyConfiguration.telemetry.controlPlane.interval	number	30
agent.kedifyConfiguration.telemetry.controlPlane.disabled	boolean	false
agent.kedifyConfiguration.telemetry.kedaResources.interval	number	30
agent.kedifyConfiguration.telemetry.kedaResources.disabled	boolean	false
agent.kedifyConfiguration.telemetry.kubernetesResources.interval	number	30
agent.kedifyConfiguration.telemetry.kubernetesResources.disabled	boolean	true
agent.image.tag	string	"v0.7.0"
agent.image.repository	string	"ghcr.io/kedify/agent"
agent.image.pullPolicy	string	"Always"
agent.imagePullSecrets	array	[]
agent.kubectlImage.tag	string	"v1.33.1"
agent.kubectlImage.repository	string	"ghcr.io/kedify/kubectl"
agent.kubectlImage.pullPolicy	string	"Always"
agent.metricsWatchNamespace	string	""
agent.noBanner	boolean	false
agent.noColor	boolean	false
agent.nameOverride	string	""
agent.fullnameOverride	string	""
agent.features.podResourceProfilesDownscaleGuardEnabled	boolean	true
agent.features.podResourceProfilesEnabled	boolean	true
agent.features.podResourceAutoscalersEnabled	boolean	false
agent.features.argoRolloutsEnabled	boolean	false
agent.features.prpRequiresAnnotatedPods	boolean	true
agent.features.praRequiresAnnotatedPods	boolean	false
agent.features.distributedScaledObjectsEnabled	boolean	false
agent.features.distributedScaledJobsEnabled	boolean	false
agent.features.scaleAdaptersEnabled	boolean	false
agent.features.scalingPoliciesEnabled	boolean	true
agent.features.recommendationsForLabeledNamespaces	boolean	true
agent.features.multitenantKEDAEnabled	boolean	false
agent.features.kedifyPodAutoscalerEnabled	boolean	false
agent.multicluster.localCluster.enabled	boolean	false
agent.multicluster.localCluster.name	string	"multicluster-local"
agent.autowire.serviceSyncPeriod	string	"500ms"
agent.autowire.ingressCooldownPeriod	string	"3600s"
agent.rbac.create	boolean	true
agent.rbac.shouldBeAbleToInstallKeda	boolean	false
agent.rbac.shouldBeAbleToInstallHttpAddon	boolean	false
agent.rbac.selfUpdates	boolean	false
agent.rbac.readMetrics	boolean	true
agent.rbac.readLogs	boolean	true
agent.rbac.readStatefulSets	boolean	false
agent.rbac.readDaemonSets	boolean	false
agent.rbac.readHpas	boolean	false
agent.rbac.readNodes	boolean	true
agent.rbac.readPods	boolean	true
agent.rbac.readDeploymentsClusterwide	boolean	true
agent.rbac.readDeployments	boolean	true
agent.rbac.readServices	boolean	true
agent.rbac.ingressAutoWire	boolean	true
agent.rbac.kedifyConfig	boolean	false
agent.extraRbacRules	array	[]
agent.extraObjects	array	[]
agent.extraPreInstallObjects	array	[]
agent.extraPostInstallObjects	array	[]
agent.serviceAccount.create	boolean	true
agent.serviceAccount.automount	boolean	true
agent.serviceAccount.annotations	object	{}
agent.podAnnotations.kubectl.kubernetes.io/default-container	string	"manager"
agent.podLabels	object	{}
agent.resources.limits.cpu	string	"500m"
agent.resources.limits.memory	string	"1024Mi"
agent.resources.requests.cpu	string	"10m"
agent.resources.requests.memory	string	"300Mi"
agent.volumes	array	[]
agent.volumeMounts	array	[]
agent.podSecurityContext	object	{}
agent.securityContext.runAsNonRoot	boolean	true
agent.securityContext.allowPrivilegeEscalation	boolean	false
agent.securityContext.readOnlyRootFilesystem	boolean	false
agent.securityContext.seccompProfile.type	string	"RuntimeDefault"
agent.securityContext.capabilities.drop	array	["ALL"]
agent.nodeSelector	object	{}
agent.tolerations	array	[]
agent.affinity	object	{}
agent.extraArgs	object	{}
hooks	string	"helm"
keda.enabled	boolean	false
keda.prometheus.operator.enabled	boolean	true
keda.prometheus.metricServer.enabled	boolean	true
keda.prometheus.webhooks.enabled	boolean	true
keda-add-ons-http.enabled	boolean	false
keda-add-ons-http.interceptor.scaledObject.waitForCrd	boolean	true
otel-add-on.enabled	boolean	false
otel-add-on.fullnameOverride	string	"kedify-otel-scaler"
otel-add-on.otelOperator.enabled	boolean	true
kedify-predictor.enabled	boolean	false
kedify-predictor.fullnameOverride	string	"kedify-predictor"
autoscaling-checks.enabled	boolean	false
autoscaling-checks.http.enabled	boolean	true
autoscaling-checks.prometheus.enabled	boolean	false
autoscaling-checks.otel.enabled	boolean	false
autoscaling-checks.cpu.enabled	boolean	true
autoscaling-checks.memory.enabled	boolean	false

Standalone KPA chart snapshot

Chart-relative path	Type	Snapshot default
replicaCount	number	2
image.repository	string	"ghcr.io/kedify/kpa"
image.tag	string	""
image.pullPolicy	string	"IfNotPresent"
image.hardened	boolean	false
imagePullSecrets	array	[]
nameOverride	string	""
fullnameOverride	string	""
serviceAccount.create	boolean	true
serviceAccount.annotations	object	{}
serviceAccount.name	string	""
crds.install	boolean	true
watchNamespace	string	""
controller.syncPeriod	string	"15s"
controller.minimumSyncPeriod	string	"5s"
controller.infrastructureRetryMax	string	"30s"
controller.maxConcurrentReconciles	number	5
controller.downscaleStabilization	string	"5m"
controller.tolerance	number	0.1
controller.cpuInitializationPeriod	string	"5m"
controller.initialReadinessDelay	string	"30s"
controller.kubeAPIQPS	number	20
controller.kubeAPIBurst	number	30
controller.kedaRequestTimeout	string	"3s"
controller.leaderElect	boolean	true
controller.leaderElectionID	string	""
controller.logLevel	string	"info"
controller.extraArgs	array	[]
keda.metricsAddress	string	"keda-operator.keda.svc:9666"
keda.metricsAuthority	string	"keda-operator.keda.svc"
keda.certificateSecret	string	"kedaorg-certs"
keda.certificateMountPath	string	"/certs"
podAnnotations.prometheus.io/scrape	string	"true"
podAnnotations.prometheus.io/port	string	"8080"
podAnnotations.prometheus.io/path	string	"/metrics"
podLabels.app.kubernetes.io/part-of	string	"kedify-pod-autoscaler"
podSecurityContext.runAsNonRoot	boolean	true
podSecurityContext.runAsUser	number	65532
podSecurityContext.runAsGroup	number	65532
podSecurityContext.fsGroup	number	65532
podSecurityContext.seccompProfile.type	string	"RuntimeDefault"
securityContext.allowPrivilegeEscalation	boolean	false
securityContext.capabilities.drop	array	["ALL"]
securityContext.readOnlyRootFilesystem	boolean	true
securityContext.runAsNonRoot	boolean	true
resources.requests.cpu	string	"50m"
resources.requests.memory	string	"64Mi"
resources.limits.cpu	string	"500m"
resources.limits.memory	string	"256Mi"
service.annotations	object	{}
service.metricsPort	number	8080
serviceMonitor.enabled	boolean	false
serviceMonitor.labels	object	{}
serviceMonitor.interval	string	"30s"
serviceMonitor.scrapeTimeout	string	"10s"
prometheusRule.enabled	boolean	false
prometheusRule.labels	object	{}
podDisruptionBudget.enabled	boolean	true
podDisruptionBudget.minAvailable	number	1
networkPolicy.enabled	boolean	false
networkPolicy.apiServerPorts	array	[443, 6443]
networkPolicy.prometheusNamespaceSelector	object	{}
networkPolicy.prometheusPodSelector	object	{}
networkPolicy.kedaNamespaceSelector	object	{}
networkPolicy.kedaPodSelector.matchLabels.app.kubernetes.io/name	string	"keda-operator"
nodeSelector	object	{}
tolerations	array	[]
affinity	object	{}
topologySpreadConstraints	array	[]
priorityClassName	string	""
terminationGracePeriodSeconds	number	10

Continue with this topic

Related capabilities: Kedify Agent and data collection.