Prometheus Helm Values Reference

Overview

This document provides the configuration reference for the Prometheus monitoring system deployed in the Fawkes platform. Prometheus collects metrics from platform services and applications for observability and alerting.

Helm Chart: prometheus/prometheus (Prometheus Community chart)

Chart Repository: https://prometheus-community.github.io/helm-charts

Values File Location: platform/apps/prometheus/values.yaml

---

Server Configuration

server.resources

Resource allocation for the Prometheus server.

Field	Type	Required	Default	Description
server.resources.requests.cpu	String	No	100m	CPU request (millicores).
server.resources.requests.memory	String	No	256Mi	Memory request.
server.resources.limits.cpu	String	No	200m	Maximum CPU allocation.
server.resources.limits.memory	String	No	512Mi	Maximum memory allocation.

Fawkes Defaults:

server:
  resources:
    requests:
      memory: 256Mi
      cpu: 100m
    limits:
      memory: 512Mi
      cpu: 200m

Scaling Guidelines:

Environment	CPU Request	Memory Request	CPU Limit	Memory Limit
Development	100m	256Mi	200m	512Mi
Staging	200m	512Mi	500m	1Gi
Production	500m	1Gi	2	4Gi

---

server.service

Kubernetes Service configuration for the Prometheus server.

Field	Type	Required	Default	Description
server.service.type	String	No	ClusterIP	Service type: ClusterIP, NodePort, LoadBalancer.
server.service.port	Integer	No	80	Service port.
server.service.targetPort	Integer	No	9090	Container port.

Fawkes Default:

server:
  service:
    type: ClusterIP

Access Pattern: Prometheus is accessed via Ingress, not directly exposed.

---

server.persistentVolume

Persistent storage for metrics data.

Field	Type	Required	Default	Description
server.persistentVolume.enabled	Boolean	No	true	Enable persistent storage.
server.persistentVolume.size	String	No	8Gi	Storage size.
server.persistentVolume.storageClass	String	No	standard	StorageClass name.
server.persistentVolume.accessModes	Array[String]	No	["ReadWriteOnce"]	Volume access modes.

Retention Recommendations:

Retention Period	Recommended Size	Use Case
7 days	8Gi	Development
15 days	20Gi	Staging
30 days	50Gi	Production
90 days	150Gi	Long-term analysis

Example:

server:
  persistentVolume:
    enabled: true
    size: 50Gi
    storageClass: gp3

---

server.retention

Metrics retention configuration.

Field	Type	Required	Default	Description
server.retention	String	No	15d	Time-based retention (e.g., 7d, 30d).
server.retentionSize	String	No	-	Size-based retention (e.g., 10GB, 50GB). Overrides time-based retention.

Example:

server:
  retention: "30d"
  retentionSize: "45GB"

---

Alerting Configuration

alertmanager.enabled

Enable or disable Alertmanager integration.

Field	Type	Required	Default	Description
alertmanager.enabled	Boolean	No	true	Deploy Alertmanager for alert routing.

Fawkes Default: true

---

alertmanagerFiles

Alert routing configuration.

Field	Type	Required	Default	Description
alertmanagerFiles.alertmanager.yml	Object	No	{}	Alertmanager configuration (receivers, routes).

Example:

alertmanagerFiles:
  alertmanager.yml:
    global:
      resolve_timeout: 5m
    route:
      group_by: ["alertname", "cluster", "service"]
      group_wait: 10s
      group_interval: 10s
      repeat_interval: 12h
      receiver: "mattermost"
    receivers:
      - name: "mattermost"
        webhook_configs:
          - url: "https://mattermost.fawkes.example.com/hooks/alerts"

---

Scrape Configurations

serverFiles.prometheus.yml

Prometheus scrape configuration.

Field	Type	Required	Default	Description
serverFiles.prometheus.yml.scrape_configs	Array[Object]	No	[]	List of scrape jobs.

Default Scrape Targets:

Job	Target	Metrics
prometheus	Prometheus itself	Self-monitoring metrics.
kubernetes-apiservers	Kubernetes API server	Control plane metrics.
kubernetes-nodes	Kubelet on each node	Node metrics (CPU, memory, disk).
kubernetes-pods	Pods with prometheus.io/scrape=true annotation	Application metrics.
kubernetes-service-endpoints	Service endpoints	Service-level metrics.

Example Custom Scrape Job:

serverFiles:
  prometheus.yml:
    scrape_configs:
      - job_name: "jenkins"
        static_configs:
          - targets: ["jenkins.jenkins.svc.cluster.local:8080"]
        metrics_path: "/prometheus"
        scrape_interval: 30s

---

Service Discovery

kubeStateMetrics.enabled

Enable Kube State Metrics for Kubernetes resource metrics.

Field	Type	Required	Default	Description
kubeStateMetrics.enabled	Boolean	No	true	Deploy kube-state-metrics.

Metrics Collected:

• Deployment status (replicas, available replicas)
• Pod status (phase, restarts, resource usage)
• Node status (allocatable resources, conditions)
• PersistentVolumeClaim status (phase, capacity)

---

nodeExporter.enabled

Enable Node Exporter for node-level metrics.

Field	Type	Required	Default	Description
nodeExporter.enabled	Boolean	No	true	Deploy node-exporter DaemonSet.

Metrics Collected:

• CPU usage and load average
• Memory usage (total, available, buffers, caches)
• Disk I/O and space
• Network traffic and errors

---

Ingress Configuration

server.ingress

Ingress configuration for accessing Prometheus UI.

Field	Type	Required	Default	Description
server.ingress.enabled	Boolean	No	false	Enable Ingress resource creation.
server.ingress.annotations	Object	No	{}	Ingress annotations (e.g., cert-manager.io/cluster-issuer).
server.ingress.hosts	Array[String]	No	[]	Hostnames for Ingress rules.
server.ingress.tls	Array[Object]	No	[]	TLS configuration.

Example:

server:
  ingress:
    enabled: true
    annotations:
      cert-manager.io/cluster-issuer: letsencrypt-prod
      nginx.ingress.kubernetes.io/auth-type: basic
      nginx.ingress.kubernetes.io/auth-secret: prometheus-basic-auth
    hosts:
      - prometheus.fawkes.example.com
    tls:
      - secretName: prometheus-tls
        hosts:
          - prometheus.fawkes.example.com

---

Federation Configuration

server.global

Global Prometheus configuration for federation.

Field	Type	Required	Default	Description
server.global.scrape_interval	String	No	15s	Default scrape interval.
server.global.scrape_timeout	String	No	10s	Default scrape timeout.
server.global.evaluation_interval	String	No	15s	Rule evaluation interval.
server.global.external_labels	Object	No	{}	Labels added to all metrics (e.g., cluster, environment).

Example:

server:
  global:
    scrape_interval: 15s
    evaluation_interval: 15s
    external_labels:
      cluster: "fawkes-prod"
      environment: "production"

---

Security Configuration

server.podSecurityPolicy

Pod Security Policy for Prometheus server.

Field	Type	Required	Default	Description
server.podSecurityPolicy.enabled	Boolean	No	false	Enable PodSecurityPolicy (deprecated in K8s 1.25+).

Note: Use Kyverno policies instead for modern Kubernetes versions.

---

Complete Example

server:
  resources:
    requests:
      memory: 1Gi
      cpu: 500m
    limits:
      memory: 4Gi
      cpu: 2
  service:
    type: ClusterIP
  persistentVolume:
    enabled: true
    size: 50Gi
    storageClass: gp3
  retention: "30d"
  global:
    scrape_interval: 15s
    external_labels:
      cluster: "fawkes-prod"
  ingress:
    enabled: true
    hosts:
      - prometheus.fawkes.example.com
    tls:
      - secretName: prometheus-tls
        hosts:
          - prometheus.fawkes.example.com

alertmanager:
  enabled: true

kubeStateMetrics:
  enabled: true

nodeExporter:
  enabled: true

---

See Also

• Prometheus Official Documentation
• Prometheus Helm Chart
• View DORA Metrics
• Unified Telemetry Explanation