:::caution Auto-generated This page is auto-generated from the Helm chart source. Do not edit directly. Regenerate with make docs-helm-charts. :::

rag-server

RAG server

Version	1.1.0
Type	application

Quick Start

# Add and install the chart
helm install rag-server oci://ghcr.io/caipe-io/charts/rag-server --version 1.1.0

# Upgrade an existing release
helm upgrade rag-server oci://ghcr.io/caipe-io/charts/rag-server --version 1.1.0

Customizing Values

Override default values using --set flags or a custom values file:

# Override individual values
helm install rag-server oci://ghcr.io/caipe-io/charts/rag-server --version 1.1.0 \
  --set replicaCount=2

# Use a custom values file
helm install rag-server oci://ghcr.io/caipe-io/charts/rag-server --version 1.1.0 \
  -f custom-values.yaml

# Show all configurable values
helm show values oci://ghcr.io/caipe-io/charts/rag-server --version 1.1.0

Reading the Values Table

Column	Meaning
Key	Dot-separated path into values.yaml (e.g. image.repository)
Type	Go/Helm data type (string, int, bool, object, list)
Default	Value used when not overridden
Description	What the parameter controls

Values

Key	Type	Default	Description
affinity	object	{}
enableGraphRag	bool	true
env	object	{}
envFrom	list	[]
extraVolumes	list	[]
fullnameOverride	string	"rag-server"
image.pullPolicy	string	"IfNotPresent"
image.repository	string	"ghcr.io/caipe-io/caipe-rag-server"
image.tag	string	""
imagePullSecrets	list	[]
initContainers	list	[]
livenessProbe.failureThreshold	int	3
livenessProbe.httpGet.path	string	"/health"
livenessProbe.httpGet.port	string	"http"
livenessProbe.periodSeconds	int	10
livenessProbe.timeoutSeconds	int	5
llmSecrets.secretName	string	"llm-secret"
nameOverride	string	""
nodeSelector	object	{}
podAnnotations	object	{}
podLabels	object	{}
podSecurityContext.runAsNonRoot	bool	true
readinessProbe.failureThreshold	int	3
readinessProbe.httpGet.path	string	"/healthz"
readinessProbe.httpGet.port	string	"http"
readinessProbe.periodSeconds	int	5
readinessProbe.timeoutSeconds	int	3
replicaCount	int	1
resources	object	{}
revisionHistoryLimit	int	3
securityContext.allowPrivilegeEscalation	bool	false
securityContext.capabilities.drop[0]	string	"ALL"
securityContext.readOnlyRootFilesystem	bool	false
securityContext.runAsNonRoot	bool	true
securityContext.runAsUser	int	1001
securityContext.seccompProfile.type	string	"RuntimeDefault"
service.port	int	9446
service.type	string	"ClusterIP"
serviceAccount.annotations	object	{}
serviceAccount.automount	bool	false
serviceAccount.create	bool	false
serviceAccount.name	string	""
startupProbe.failureThreshold	int	30
startupProbe.httpGet.path	string	"/health"
startupProbe.httpGet.port	string	"http"
startupProbe.initialDelaySeconds	int	10
startupProbe.periodSeconds	int	10
startupProbe.timeoutSeconds	int	5
tolerations	list	[]
vpa.controlledResources[0]	string	"cpu"
vpa.controlledResources[1]	string	"memory"
vpa.controlledValues	string	"RequestsAndLimits"
vpa.enabled	bool	false
vpa.maxAllowed	object	{}
vpa.minAllowed.cpu	string	"50m"
vpa.minAllowed.memory	string	"128Mi"
vpa.updateMode	string	"InPlaceOrRecreate"