fil/charts/kreuzberg/values.yaml

# -- Number of replicas.
# WARNING: When cache.enabled=true and cache.accessModes=[ReadWriteOnce], only one
# replica can mount the PVC at a time. Keep replicaCount: 1 with RWO storage, or
# switch to ReadWriteMany storage before increasing replicas. With RWO + multiple
# replicas the deployment strategy must be Recreate (not RollingUpdate).
replicaCount: 1

# -- Deployment strategy. When cache is enabled with ReadWriteOnce storage,
# set to Recreate to avoid Multi-Attach errors during rolling updates.
strategy:
  type: Recreate

image:
  # -- Container image registry
  registry: ghcr.io
  # -- Container image repository
  repository: kreuzberg-dev/kreuzberg
  # -- Image tag. Defaults to Chart.AppVersion when empty.
  # Use "latest" for the full image (Tesseract + PaddleOCR + layout models)
  # or "core" for the minimal image (no pre-downloaded models).
  tag: ""
  # -- Image pull policy
  pullPolicy: IfNotPresent

# -- Image pull secrets for private registries
imagePullSecrets: []

# -- Override the chart name
nameOverride: ""
# -- Override the full release name
fullnameOverride: ""

serviceAccount:
  # -- Create a ServiceAccount
  create: true
  # -- Annotations for the ServiceAccount
  annotations: {}
  # -- Override the ServiceAccount name (defaults to release fullname)
  name: ""

# -- Pod-level annotations
podAnnotations: {}

# -- Pod-level security context
podSecurityContext:
  runAsNonRoot: true
  runAsUser: 1000
  runAsGroup: 1000
  fsGroup: 1000
  seccompProfile:
    type: RuntimeDefault

# -- Container-level security context
securityContext:
  allowPrivilegeEscalation: false
  readOnlyRootFilesystem: true
  capabilities:
    drop: ["ALL"]

service:
  # -- Service type
  type: ClusterIP
  # -- Service port
  port: 80

ingress:
  # -- Enable Ingress
  enabled: false
  # -- Ingress class name (e.g. "nginx")
  className: ""
  # -- Ingress annotations
  annotations: {}
  # -- Ingress hosts
  hosts:
    - host: kreuzberg.local
      paths:
        - path: /
          pathType: Prefix
  # -- Ingress TLS configuration
  tls: []
  #  - secretName: kreuzberg-tls
  #    hosts:
  #      - kreuzberg.local

# -- Container resource requests and limits
resources:
  requests:
    memory: "512Mi"
    cpu: "500m"
  limits:
    memory: "2Gi"
    cpu: "2000m"

autoscaling:
  # -- Enable HorizontalPodAutoscaler
  enabled: false
  # -- Minimum replicas. Note: if podDisruptionBudget.minAvailable equals this
  # value, scale-down will be blocked. Set minReplicas lower than minAvailable
  # or raise minAvailable accordingly.
  minReplicas: 1
  # -- Maximum replicas
  maxReplicas: 10
  # -- Target CPU utilization (percent)
  targetCPUUtilizationPercentage: 80
  # -- Target memory utilization (percent). Leave unset to disable.
  # targetMemoryUtilizationPercentage: 80

# -- Node selector for pod scheduling
nodeSelector: {}

# -- Tolerations for pod scheduling
tolerations: []

# -- Affinity rules for pod scheduling
affinity: {}

# -- Extra environment variables
extraEnv: []
#  - name: KREUZBERG_CORS_ORIGINS
#    value: "https://app.example.com"
#  - name: KREUZBERG_MAX_UPLOAD_SIZE_MB
#    value: "500"

# -- Kreuzberg-specific configuration
kreuzberg:
  # -- Log level: trace, debug, info, warn, error
  logLevel: "info"
  # -- Tesseract data prefix path (must match the container image)
  tessdataPrefix: "/usr/share/tesseract-ocr/5/tessdata"
  # -- Default OCR language
  ocrLanguage: "eng"

cache:
  # -- Enable persistent cache for embedding models and downloaded assets.
  # Models range from 90 MB to 1.2 GB and are re-downloaded on every pod
  # restart without a PVC.
  enabled: true
  # -- Storage size for the cache PVC
  size: 2Gi
  # -- StorageClass for the cache PVC (empty string uses cluster default)
  storageClass: ""
  # -- Access modes for the cache PVC
  accessModes:
    - ReadWriteOnce
  # -- Run an init container to chown the cache directory to UID 1000.
  # Most block-backed storage classes (EBS, GKE PD, Azure Disk) handle
  # ownership automatically via the pod's fsGroup, so you can set this
  # to false there. Set to true for NFS or other storage that does not
  # honour fsGroup on mount.
  initChown: true

# -- Disable Kubernetes service-discovery environment variable injection.
# Kubernetes injects {SVCNAME}_PORT=tcp://<clusterIP>:<port> for every
# Service in the namespace. When the release is named "kreuzberg" this
# injects KREUZBERG_PORT which the binary parses as a u16 and panics.
# CoreDNS makes these variables unnecessary in all modern clusters.
enableServiceLinks: false

podDisruptionBudget:
  # -- Enable PodDisruptionBudget
  enabled: false
  # -- Minimum available pods during disruption
  minAvailable: 1