Configuration

This page covers the controller manager's runtime flags. For resource-level configuration (hub connection, form policies, templates), follow the links in each section.

Controller Flags

Set in config/manager/manager.yaml under spec.template.spec.containers[].args:

args:
  # Leader election
  - --leader-elect=true                        # required in HA deployments

  # Concurrency (tune for cluster scale)
  - --hub-concurrency=3                        # concurrent hub syncs (default: 3)
  - --form-concurrency=5                       # concurrent form reconciliations (default: 5)
  - --node-concurrency=10                      # concurrent node reconciliations (default: 10)

  # Metrics and health
  - --metrics-bind-address=:8443               # Prometheus scrape endpoint (0 = disabled)
  - --health-probe-bind-address=:8081          # liveness/readiness probes
  - --metrics-secure=true                      # serve metrics over HTTPS

  # TLS (cert-manager provisions these automatically)
  - --webhook-cert-path=/tmp/k8s-webhook-server/serving-certs
  - --metrics-cert-path=/tmp/k8s-metrics-server/serving-certs

  # Logging
  - --zap-log-level=info                       # debug | info | error
  - --zap-encoder=json                         # json | console
  - --zap-devel=false                          # true enables development mode

Concurrency guidance:

Cluster scale	hub	form	node
Small (<50 nodes)	3	5	10
Medium (50–500)	5	8	20
Large (500+)	8	10	30

For more tuning options, see Performance.

Resource Limits

The shipped manifests and chart use conservative defaults:

resources:
  requests:
    cpu: 10m
    memory: 64Mi
  limits:
    cpu: 500m
    memory: 128Mi

Increase memory before scaling beyond a few hundred LynqNodes — the controller-runtime cache grows with total managed object count. For observed benchmarks, a sizing table, and a CPU/memory model, see Resource Sizing.

Configuration by Topic

Topic	Where to configure	Reference
Database connection, sync interval, column mappings	LynqHub spec	Datasource
Resource blueprints, rollout settings	LynqForm spec	Templates
CreationPolicy, DeletionPolicy, ConflictPolicy, PatchStrategy	Per-resource in LynqForm	Policies
Prometheus metrics, alerting	Operator flags + Prometheus config	Monitoring
RBAC, network policies, credentials	Cluster config + SecretRef	Security

See Also

• Installation — deploying the operator
• Performance — scaling and optimization
• Security — RBAC and credential management