Configuration Reference¶

SLOpilot is configured through Helm values. Pass custom values using --values my-overrides.yaml when running the installer, or use helm upgrade directly after installation.

# During installation
./slopilot-install.sh --values my-overrides.yaml ...

# After installation
helm upgrade slopilot-rightsizing \
    oci://ghcr.io/valuematic/charts/slopilot-rightsizing \
    -n slopilot \
    --values my-overrides.yaml

Ingress¶

By default, SLOpilot is only reachable via kubectl port-forward or a NodePort/LoadBalancer Service. To expose it through an Ingress, enable it and configure the host:

ingress:
  enabled: true
  className: nginx          # or traefik, alb, etc.
  hosts:
    - host: slopilot.example.com
      paths:
        - path: /
          pathType: Prefix

For TLS, add a tls block referencing an existing kubernetes.io/tls Secret in the same namespace. See the controller-specific examples below.

When SLOpilot is behind an ingress controller, configure trusted_proxy_cidrs so that forwarded client IPs are respected:

security:
  trusted_proxy_cidrs:
    - "10.0.0.0/8"

Resources¶

Default resource requests and limits for the SLOpilot pod:

	CPU	Memory
Requests	100m	256Mi
Limits	500m	512Mi

For clusters with more than 200 monitored workloads, consider increasing these values.

Prometheus¶

SLOpilot deploys its own Prometheus instance by default. It is pre-configured with metric filtering, retention, and storage guardrails — no tuning is required.

If your cluster's default StorageClass does not support the default PVC size, you can override it:

prometheus-stack:
  server:
    persistentVolume:
      size: 40Gi              # default; adjust if needed

Tip

Both the application and Prometheus PVCs are annotated with helm.sh/resource-policy: keep and survive helm uninstall. Your data is preserved across reinstalls.

Network Policy¶

A NetworkPolicy is created by default. It restricts traffic to the minimum required:

Ingress: HTTP on port 8080 from all sources
Egress (DNS): UDP/TCP port 53 to all namespaces
Egress (Prometheus): ports 80 and 9090 within the same namespace
Egress (Kubernetes API): TCP ports 443 and 6443
Egress (License server): TCP port 443

Disable it with networkPolicy.enabled: false if your cluster does not use NetworkPolicies or you manage them externally.

Scheduling¶

Place SLOpilot on specific nodes using standard Kubernetes scheduling fields:

nodeSelector:
  workload-type: platform

tolerations:
  - key: platform
    operator: Equal
    value: "true"
    effect: NoSchedule

affinity and topologySpreadConstraints are also supported.

Ingress Examples¶

NGINXTraefikAWS ALB

ingress:
  enabled: true
  className: nginx
  annotations:
    nginx.ingress.kubernetes.io/ssl-redirect: "true"
  hosts:
    - host: slopilot.example.com
      paths:
        - path: /
          pathType: Prefix
  tls:
    secretName: slopilot-tls

ingress:
  enabled: true
  className: traefik
  annotations:
    traefik.ingress.kubernetes.io/router.tls: "true"
  hosts:
    - host: slopilot.example.com
      paths:
        - path: /
          pathType: Prefix
  tls:
    secretName: slopilot-tls

ingress:
  enabled: true
  className: alb
  annotations:
    alb.ingress.kubernetes.io/scheme: internal
    alb.ingress.kubernetes.io/target-type: ip
    alb.ingress.kubernetes.io/certificate-arn: arn:aws:acm:...
    alb.ingress.kubernetes.io/listen-ports: '[{"HTTPS": 443}]'
  hosts:
    - host: slopilot.example.com
      paths:
        - path: /
          pathType: Prefix

Complete Custom Values Example¶

The following my-overrides.yaml combines ingress, trusted proxies, increased resources, and node scheduling:

# my-overrides.yaml

# Expose via NGINX ingress with TLS
ingress:
  enabled: true
  className: nginx
  annotations:
    nginx.ingress.kubernetes.io/ssl-redirect: "true"
  hosts:
    - host: slopilot.example.com
      paths:
        - path: /
          pathType: Prefix
  tls:
    secretName: slopilot-tls

# Trust the in-cluster ingress controller's forwarded headers
security:
  trusted_proxy_cidrs:
    - "10.0.0.0/8"

# Increase application resource limits for large clusters
resources:
  requests:
    cpu: 200m
    memory: 512Mi
  limits:
    cpu: 1000m
    memory: 1Gi

# Schedule on dedicated infrastructure nodes
nodeSelector:
  workload-type: platform

tolerations:
  - key: platform
    operator: Equal
    value: "true"
    effect: NoSchedule

Apply it during installation:

./slopilot-install.sh \
    --username <ghcr-username> \
    --password <ghcr-token> \
    --license-key "SLOPILOT-XXXX" \
    --tag vX.Y.Z \
    --values my-overrides.yaml