TFRTenant Deployment

Deploy TFRTenant tenant environment.

Overview

TFRTenant is a namespace-scoped resource that creates tenant-isolated environments and middleware instances:

• Namespace - Tenant-dedicated namespace
• PostgreSQL - Database cluster via CloudNativePG
• MinIO - Object storage instance
• Redis - Caching service
• RabbitMQ - Message queue
• Grafana - Monitoring dashboard (optional)
• Jaeger - Distributed tracing (optional)

Prerequisites

TFRTenant depends on Operators installed by TFRCluster. Ensure TFRCluster status is Available before deployment.

---

Configuration

Core Fields

Field	Description	Example
name	TFRTenant resource name	enaic
namespace	Tenant namespace (auto-created)	staging-tenant
tenantType	Tenant type	enterprise, personal
storage	Storage configuration	See below
imagePullSecrets	Private registry credentials reference	See below

Storage Configuration

storage:
  enabled: true
  storageClass: "cbs"      # TKE uses cbs, Kind uses standard
  defaultSize: "10Gi"      # Default storage size for components

Storage Class Notes

• Tencent Cloud TKE: cbs (Cloud Block Storage)
• Kind / Docker Desktop: standard
• Shared Cache: Requires RWX-capable storage class (e.g., cfs)

Private Registry Credentials

Reference Secret created by TFRCluster:

imagePullSecrets:
  - secretName: "tfrs-private-registry"
    sourceNamespace: "tfr-image-credentials"

Component Configuration

Each component supports independent storage and credential configuration, inheriting tenant-level config when not specified:

postgresql:
  enabled: false           # Enable component-level config
  username: ""             # Defaults to TFRobotID
  password: ""             # Auto-generated by default
  storage:
    storageClass: ""       # Inherits from tenant-level
    defaultSize: "10Gi"

grafana:
  enabled: true
  adminUser: "admin"
  adminPassword: "admin"   # Use strong password in production

---

Example Configurations

Project provides example configurations:

• chart/tfrtenant/values-staging.yaml - Staging environment
• chart/tfrtenant/values-prod.yaml - Production environment
• chart/tfrtenant/values-kind.yaml - Local development environment

---

Kustomize Deployment

Create CR Manifest

# tfrtenant-staging.yaml
apiVersion: infra.tfrobotserver.cn/v1
kind: TFRTenant
metadata:
  name: enaic
spec:
  # Tenant type (Required)
  tenantType: enterprise

  # Tenant namespace (Required)
  namespace: staging-tenant

  # Storage configuration
  storage:
    storageClass: "cbs"      # TKE uses cbs, Kind uses standard
    defaultSize: "10Gi"

  # Private registry credentials (Reference Secret created by TFRCluster)
  imagePullSecrets:
    - secretName: "tfrs-private-registry"
      sourceNamespace: "tfr-image-credentials"

  # Grafana configuration
  grafana:
    adminUser: "admin"
    adminPassword: "your-secure-password"  # Use strong password in production

Field Notes

• tenantType and namespace are required fields
• The enabled field in Helm values files is for Chart rendering logic, not needed in CR
• Full field reference: config/samples/infra_v1_tfrtenant.yaml

Deploy

kubectl apply -f tfrtenant-staging.yaml
# tfrtenant.infra.tfrobotserver.cn/enaic created

Verify

kubectl get tfrtenant
kubectl get tfrtenant enaic -o json | jq '.status.components | to_entries[] | "\(.key): \(.value.state)"'

Real-time deployment monitoring

while true; do clear; echo "--- $(date +%H:%M:%S) ---"; \
  echo "=== Conditions ==="; \
  kubectl get tfrtenant enaic -o json 2>/dev/null \
    | jq -r '.status.conditions[]? | "\(.type): \(.status) - \(.message)"'; \
  echo; echo "=== Components ==="; \
  kubectl get tfrtenant enaic -o json 2>/dev/null \
    | jq -r '.status.components // {} | to_entries[] | "\(.key): \(.value.state)"'; \
  echo; echo "=== Pods ==="; \
  kubectl get pods -n staging-tenant 2>/dev/null; \
  sleep 10; done

Press Ctrl+C to exit.

Determine Deployment Completion

TFRTenant indicates deployment status via the Ready condition in status.conditions:

# Check Ready condition
kubectl get tfrtenant enaic -o jsonpath='{.status.conditions[?(@.type=="Ready")].status}'
# True = deployment complete, False = deploying or failed

# Automation script: wait for deployment to complete (10 minute timeout)
kubectl wait tfrtenant enaic --for=condition=Ready --timeout=600s

Uninstall

kubectl delete tfrtenant enaic

---

Helm Deployment

Deploy

Install from OCI registry

helm install tfrtenant oci://helm.cnb.cool/turingfocus/tfrs-operator/tfrtenant \
  --version 0.1.0 \
  --set name=enaic \
  --set namespace=staging-tenant \
  --set storage.enabled=true \
  --set storage.storageClass=cbs \
  --set grafana.enabled=true

Install with values file

helm install tfrtenant oci://helm.cnb.cool/turingfocus/tfrs-operator/tfrtenant \
  --version 0.1.0 \
  -f values-staging.yaml

Install from local source

helm install tfrtenant ./chart/tfrtenant \
  -f chart/tfrtenant/values-staging.yaml

Note

TFRTenant is a cluster-scoped resource, no need to specify --namespace. The tenant namespace is specified by the namespace field, and the Controller creates it automatically.

Verify

# Check TFRTenant status
kubectl get tfrtenant

# Wait for deployment to complete (10 minute timeout)
kubectl wait tfrtenant enaic --for=condition=Available --timeout=600s

# View components in tenant namespace
kubectl get pods -n staging-tenant

Real-time deployment monitoring

while true; do clear; echo "--- $(date +%H:%M:%S) ---"; \
  echo "=== Conditions ==="; \
  kubectl get tfrtenant enaic -o json 2>/dev/null \
    | jq -r '.status.conditions[]? | "\(.type): \(.status) - \(.message)"'; \
  echo; echo "=== Components ==="; \
  kubectl get tfrtenant enaic -o json 2>/dev/null \
    | jq -r '.status.components // {} | to_entries[] | "\(.key): \(.value.state)"'; \
  echo; echo "=== Pods ==="; \
  kubectl get pods -n staging-tenant 2>/dev/null; \
  sleep 10; done

Press Ctrl+C to exit.

Determine Deployment Completion

TFRTenant indicates deployment status via the Ready condition in status.conditions:

# Check Ready condition
kubectl get tfrtenant enaic -o jsonpath='{.status.conditions[?(@.type=="Ready")].status}'
# True = deployment complete, False = deploying or failed

# Automation script: wait for deployment to complete (10 minute timeout)
kubectl wait tfrtenant enaic --for=condition=Ready --timeout=600s

Uninstall

helm uninstall tfrtenant

Note

Uninstalling TFRTenant will delete the tenant namespace and all its resources (databases, storage, etc.). Ensure important data is backed up.