Oracle CRD

What is Oracle CRD?

Oracle is a Kubernetes Custom Resource Definition (CRD) maintained by KubeDB. It provides a declarative configuration for Oracle database instances in your Kubernetes cluster, enabling you to manage Oracle databases using native Kubernetes tools and practices.

Key Features

• Native Kubernetes Integration: Manage Oracle databases using kubectl and Kubernetes APIs
• Flexible Deployment Modes: Support for both Standalone and DataGuard configurations
• High Availability: Built-in support for Oracle DataGuard for disaster recovery
• Resource Management: Fine-grained control over CPU, memory, and storage resources
• Security Controls: Built-in security context configuration for enhanced security
• Automated Operations: Simplified deployment and management of complex Oracle setups

Deployment Modes

KubeDB Oracle supports two primary deployment modes:

1. Standalone Mode

• Single instance Oracle database deployment
• Suitable for development and testing environments
• Simplified configuration and management

2. DataGuard Mode

• High availability configuration with primary and standby databases
• Built-in disaster recovery capabilities
• Support for synchronous and asynchronous replication
• Observer process for automated failover

Oracle CRD Specification

Like any official Kubernetes resource, an Oracle object has standard fields:

• apiVersion: Specifies the API version (kubedb.com/v1alpha2)
• kind: Defines the resource type (Oracle)
• metadata: Contains name, namespace, and other metadata
• spec: Defines the desired state of your Oracle instance

Sample Oracle Configuration

1. Standalone Mode Configuration

apiVersion: kubedb.com/v1alpha2
kind: Oracle
metadata:
  name: oracle
  namespace: demo
spec:
  version: 21.3.0
  mode: Standalone
  edition: enterprise
  replicas: 1
  storageType: Durable
  storage:
    accessModes:
      - ReadWriteOnce
    resources:
      requests:
        storage: 30Gi
  podTemplate:
    spec:
      securityContext:
        fsGroup: 54321
        runAsGroup: 54321
        runAsUser: 54321
      containers:
        - name: oracle
          resources:
            limits:
              cpu: "4"
              memory: 10Gi
            requests:
              cpu: "2"
              memory: 3Gi
  deletionPolicy: Delete

$ kubectl create -f https://github.com/kubedb/docs/raw/v2026.6.5-rc.1/docs/examples/oracle/quickstart/standalone.yaml
oracle.kubedb.com/oracle created

2. DataGuard Mode Configuration

apiVersion: kubedb.com/v1alpha2
kind: Oracle
metadata:
  name: oracle-sample
  namespace: demo
spec:
  version: 21.3.0
  mode: DataGuard
  edition: enterprise
  replicas: 3
  
  # DataGuard Specific Configuration
  dataGuard:
    protectionMode: MaximumProtection
    standbyType: PHYSICAL
    syncMode: SYNC
    applyLagThreshold: 0
    transportLagThreshold: 0
    fastStartFailover:
      fastStartFailoverThreshold: 15
    observer:
      podTemplate:
        spec:
          containers:
          - name: observer
            resources:
              limits:
                cpu: "1"
                memory: 2Gi
              requests:
                cpu: 500m
                memory: 2Gi
      storage:
        resources:
          requests:
            storage: 1Gi
            
  # Storage Configuration
  storageType: Durable
  storage:
    accessModes:
    - ReadWriteOnce
    resources:
      requests:
        storage: 30Gi
        
  # Pod Configuration
  podTemplate:
    spec:
      securityContext:
        fsGroup: 54321
        runAsGroup: 54321
        runAsUser: 54321
      serviceAccountName: oracle-sample
      containers:
      - name: oracle
        resources:
          limits:
            cpu: "4"
            memory: 10Gi
          requests:
            cpu: "1500m"
            memory: 4Gi
      - name: oracle-coordinator
        resources:
          limits:
            memory: 256Mi
          requests:
            cpu: 200m
            memory: 256Mi
  
  deletionPolicy: Delete

Configuration Parameters Explained

Core Parameters

• version: Oracle database version (e.g., 21.3.0)
• mode: Deployment mode (Standalone or DataGuard)
• edition: Oracle edition (enterprise, standard)
• replicas: Number of database instances

DataGuard Specific Parameters

• protectionMode: Defines the data protection mode (MaximumProtection, MaximumAvailability, MaximumPerformance)
• standbyType: Type of standby database (PHYSICAL)
• syncMode: Synchronization mode between primary and standby (SYNC, ASYNC)
• applyLagThreshold: Maximum acceptable lag in applying changes
• transportLagThreshold: Maximum acceptable transport lag
• fastStartFailover: Configuration for automated failover
• observer: Configuration for the DataGuard observer process

Storage Configuration

• storageType: Type of storage (Durable for persistent storage)
• storage: Kubernetes PVC configuration
  • accessModes: Volume mount access modes
  • resources: Storage resource requests and limits

Pod Configuration

• podTemplate: Pod customization options
  • securityContext: Security settings
  • containers: Resource configurations for various containers
  • serviceAccountName: Kubernetes service account to use

Lifecycle Management

• deletionPolicy: Resource cleanup policy on deletion

Best Practices

1. Deployment Mode Selection:

   • Use Standalone mode for development/testing
   • Choose DataGuard mode for production environments requiring high availability

2. Resource Planning:

   • Allocate sufficient resources based on workload
   • Consider overhead for DataGuard synchronization
   • Plan observer resources carefully

3. Storage Configuration:

   • Use enterprise-grade storage for production
   • Configure appropriate storage class
   • Plan for backup storage

4. Security:

   • Implement proper security contexts
   • Use service accounts with minimum required permissions
   • Regular security patches and updates

5. Monitoring and Maintenance:

   • Monitor replication lag in DataGuard setups
   • Regular backup testing
   • Keep track of resource usage patterns

6. High Availability:

   • Configure appropriate FastStart Failover thresholds
   • Regular failover testing
   • Monitor observer health