Skip to content

Glass Instrumentation Configuration Reference

This page provides a complete reference for all configuration parameters available in the Glass Instrumentation Helm chart.

Shared Configuration

Many parameters use shared patterns documented in Common Configuration, including image configuration, resource requirements, and security contexts.

Cluster Identity

Cluster identification and labeling for SPOG organization.

Parameter Default Description
clusterId Unique cluster identifier for SPOG. Used to identify this cluster in the SPOG control plane. Must follow Kubernetes naming conventions (lowercase alphanumeric characters or '-'). Can be empty string (optional).

Example: prod-eu-west-1
labels {} Cluster labels for organization and filtering. Values can be a single string or an array of strings for multi-valued labels.

Example:
environment: production
region: eu-west-1
team:
- platform
- security

Reserved Label Keys

The following keywords are reserved and should not be used as label keys: and, or, not, in, like, contains, group, by, cluster_id. These conflict with the GlassQL filter query syntax.

Labels defined here are used throughout SPOG for:

Global Settings

Global configuration values that apply to all services in the chart.

Parameter Default Description
global.image.registry Override container registry for all images (e.g., 'myregistry.local:8085')

Example: myregistry.local:8085
global.image.repository Override repository path for all images (e.g., 'myproject')

Example: myproject
global.image.pullPolicy IfNotPresent Set image pull policy for all containers

Options: Always, IfNotPresent, Never
global.imagePullSecrets {} Define inline registry credentials for pulling private images (accepts arbitrary keys with object values)

Example:
myIPSSecret:
  email: admin@registry.example.com
  password: regPassword
  registry: registry.example.com:5000
  username: regUsername
global.imagePullSecretsList [] Reference existing Kubernetes image pull secrets by name

Example:
- global-imagepullsecrets-1
- global-imagepullsecrets-2
global.openshift.enabled false Enable OpenShift compatibility mode

Default Pod Configuration

Security, scheduling, and availability settings inherited by all services. The defaults implement Pod Security Standards (Restricted profile) for secure, production-ready deployments.

Parameter Default Description
containerSecurityContext 
allowPrivilegeEscalation: false
capabilities:
  drop:
  - ALL
readOnlyRootFilesystem: true
runAsNonRoot: true
seccompProfile:
  type: RuntimeDefault
Default container security context for all services. Defaults implement Pod Security Standards (Restricted profile).
podSecurityContext 
fsGroup: 10001
fsGroupChangePolicy: OnRootMismatch
runAsGroup: 10001
runAsNonRoot: true
runAsUser: 10001
seccompProfile:
  type: RuntimeDefault
Default pod security context for all services. Runs as non-root user 10001 with seccomp enabled.
podAnnotations {} Default pod annotations for all services. Useful for Prometheus scrape config or policy enforcement. (accepts arbitrary string key-value pairs)

Example:
prometheus.io/port: '8080'
prometheus.io/scrape: 'true'
podLabels {} Default pod labels for all services. Added to all pods in addition to standard labels. (accepts arbitrary string key-value pairs)
nodeSelector {} Default node selector for all services. Schedule pods only on nodes with matching labels. (accepts arbitrary string key-value pairs)

Example:
workload-type: spog
tolerations [] Default tolerations for all services. Allow pods to schedule on nodes with matching taints.

Example:
- effect: NoSchedule
  key: dedicated
  operator: Equal
  value: spog
affinity {} Default affinity rules for all services. For simple anti-affinity, use antiAffinityPreset instead.
antiAffinityPreset preferred Pod anti-affinity preset to distribute pods across nodes. 'preferred' allows co-location if needed; 'required' enforces different nodes. ('soft'/'hard' are aliases for backward compatibility.)

Options: soft, hard, preferred, required
podDisruptionBudget 
minAvailable: 1
Default pod disruption budget for all services. Controls voluntary pod evictions during maintenance.

Individual services can override these settings. Use nodeSelector and tolerations for dedicated node pools, and antiAffinityPreset: required for strict high-availability requirements.

Product Configuration

Identifies this deployment within the SPOG ecosystem.

Parameter Default Description
powerdns.product glass-instrumentation Product identifier used for resource naming and Kubernetes labeling

Example: glass-instrumentation

Instrumentation Services

Services deployed per CloudControl cluster that provide monitoring, logging, and management capabilities. Each service follows the serviceConfig pattern with common properties for replicas, logging, resources, and scheduling.

Parameter Default Description
discovery 
affinity: {}
enabled: true
logFormat: json
logLevel: info
nodeSelector: {}
podAnnotations: {}
podLabels: {}
replicas: 1
resources:
  limits:
    cpu: 500m
    memory: 512Mi
  requests:
    cpu: 100m
    memory: 128Mi
serviceAccount:
  annotations: {}
  create: true
  name: glass-discovery
tolerations: []
Discovery service configuration - handles service discovery and cluster announcement
dynamicSnapshots 
affinity: {}
enabled: true
logFormat: json
logLevel: info
nodeSelector: {}
podAnnotations: {}
podLabels: {}
replicas: 1
resources:
  limits:
    cpu: 500m
    memory: 512Mi
  requests:
    cpu: 100m
    memory: 128Mi
serviceAccount:
  annotations: {}
  create: true
  name: glass-dynamic-snapshots
tolerations: []
Dynamic Snapshots service configuration - provides real-time state snapshots
logStreamer 
affinity: {}
enabled: true
logFormat: json
logLevel: info
nodeSelector: {}
podAnnotations: {}
podLabels: {}
replicas: 1
resources:
  limits:
    cpu: 500m
    memory: 512Mi
  requests:
    cpu: 100m
    memory: 128Mi
serviceAccount:
  annotations: {}
  create: true
  name: glass-log-streamer
tolerations: []
Log Streamer service configuration - aggregates logs from pods
k8sManager 
affinity: {}
enabled: true
logFormat: json
logLevel: info
nodeSelector: {}
podAnnotations: {}
podLabels: {}
replicas: 1
resources:
  limits:
    cpu: 500m
    memory: 512Mi
  requests:
    cpu: 100m
    memory: 128Mi
serviceAccount:
  annotations: {}
  create: true
  name: glass-k8s-manager
tolerations: []
Kubernetes Manager service configuration - monitors Kubernetes resources
ccApiAdapter 
affinity: {}
enabled: true
headers:
  X-API-Key:
    secretKeyRef:
      key: token_full
      name: cc-api
logFormat: json
logLevel: info
nodeSelector: {}
podAnnotations: {}
podLabels: {}
replicas: 1
resources:
  limits:
    cpu: 500m
    memory: 512Mi
  requests:
    cpu: 100m
    memory: 128Mi
serviceAccount:
  annotations: {}
  create: true
  name: glass-cc-api-adapter
tolerations: []
CloudControl API Adapter service configuration - bridges CloudControl API. Supports custom HTTP headers for authentication.
dnsCheck 
affinity: {}
concurrentQueries: 10
dnsRetries: 3
dnsTimeout: 5s
enabled: true
logFormat: json
logLevel: info
logTimeWindow: 1m
nodeSelector: {}
podAnnotations: {}
podLabels: {}
replicas: 1
resources:
  limits:
    cpu: 500m
    memory: 512Mi
  requests:
    cpu: 100m
    memory: 128Mi
serviceAccount:
  annotations: {}
  create: true
  name: glass-dns-check
tolerations: []
DNS Check service configuration - validates and executes DNS queries.

Additional settings:
• dnsTimeout: Query timeout (default: 5s)
• dnsRetries: Retry attempts on failure (default: 3)
• concurrentQueries: Max parallel queries (default: 10)
• logTimeWindow: Log aggregation window (default: 1m)

Container Images

Container image configuration for all services. Set top-level defaults for registry, project, and pullPolicy, or override per-service using the imageConfig structure.

Parameter Default Description
images 
ccApiAdapter:
  name: rest-api-adapter
  pullPolicy: IfNotPresent
  registry: registry.open-xchange.com
  repository: glass
  tag: latest
discovery:
  name: disco
  pullPolicy: IfNotPresent
  registry: registry.open-xchange.com
  repository: glass
  tag: latest
dnsCheck:
  name: dns-check
  pullPolicy: IfNotPresent
  registry: registry.open-xchange.com
  repository: glass
  tag: latest
dynamicSnapshots:
  name: dynamic-snapshots
  pullPolicy: IfNotPresent
  registry: registry.open-xchange.com
  repository: glass
  tag: latest
k8sManager:
  name: k8s-manager
  pullPolicy: IfNotPresent
  registry: registry.open-xchange.com
  repository: glass
  tag: latest
logStreamer:
  name: log-streamer
  pullPolicy: IfNotPresent
  registry: registry.open-xchange.com
  repository: glass
  tag: latest
Container image configuration for all Glass Instrumentation services

Observability

Prometheus metrics and OpenTelemetry logging configuration for monitoring instrumentation services.

Parameter Default Description
prometheus.operator.available false Set to true if Prometheus Operator is available to create PodMonitor objects
prometheus.operator.scrape.interval 10s Scrape interval for Prometheus metrics collection (duration format: e.g., '10s', '1m', '30s')

Example: 10s
prometheus.annotations true Set to true to add Prometheus scrape annotations on pods (for Prometheus deployments without Prometheus Operator)
otelLogging.enabled false Enable OTEL log forwarding to an OTLP collector
otelLogging.endpoint OTLP collector endpoint (e.g., 'otel-collector:4318' for HTTP or 'otel-collector:4317' for gRPC)

Example: otel-collector:4318
otelLogging.protocol grpc OTLP protocol to use for log export

Options: grpc, http
otelLogging.timeout 10s Export timeout duration (e.g., '10s', '30s', '1m')

Example: 10s

See Also