100 Oracle RAC Interview Questions and Answers: Complete Guide for Database Professionals

Oracle Real Application Clusters (RAC) is a critical technology for enterprise database environments, providing high availability, scalability, and load distribution. Whether you're a DBA, system administrator, or database architect, mastering Oracle RAC is essential for career advancement.

This comprehensive guide covers 100 carefully curated Oracle RAC interview questions, ranging from basic concepts to advanced troubleshooting scenarios. Perfect for interview preparation, certification study, or knowledge assessment.

Table of Contents

1. Basic Oracle RAC Concepts (Questions 1-20)
2. RAC Architecture and Components (Questions 21-40)
3. Grid Infrastructure and ASM (Questions 41-55)
4. RAC Administration (Questions 56-70)
5. Performance and Monitoring (Questions 71-85)
6. Troubleshooting and Advanced Topics (Questions 86-100)

Basic Oracle RAC Concepts (Questions 1-20)

1. What is Oracle RAC?

Answer: Oracle Real Application Clusters (RAC) is a clustering solution that allows multiple Oracle database instances to run simultaneously on different servers while accessing the same shared database. It provides high availability, scalability, and load distribution by enabling multiple nodes to work together as a single system.

2. What are the main benefits of Oracle RAC?

Answer:

• High Availability: Automatic failover if one node fails
• Scalability: Add nodes to increase processing power
• Load Distribution: Workload distributed across multiple nodes
• Reduced Downtime: Rolling upgrades and maintenance
• Improved Performance: Parallel processing capabilities

3. What is the difference between Oracle RAC and Oracle RAC One Node?

Answer:

• Oracle RAC: Multiple instances running on multiple nodes simultaneously
• Oracle RAC One Node: Single instance running on one node at a time, with automatic failover to other nodes when needed. Provides high availability without the complexity of multiple active instances.

4. What is a cluster in Oracle RAC context?

Answer: A cluster is a group of interconnected servers (nodes) that work together to provide a unified computing resource. In Oracle RAC, the cluster consists of multiple nodes running Oracle instances that access the same shared database storage.

5. What is the minimum number of nodes required for Oracle RAC?

Answer: The minimum number of nodes for Oracle RAC is 2. However, Oracle RAC can scale up to 100 nodes in Oracle Database 12c and later versions.

6. What is a RAC instance?

Answer: A RAC instance is an Oracle database instance running on a specific node in the RAC cluster. Each node typically runs one instance, and all instances access the same shared database files stored on shared storage.

7. What is the difference between single-instance Oracle and Oracle RAC?

Answer:

• Single Instance: One Oracle instance on one server accessing database files
• Oracle RAC: Multiple Oracle instances on multiple servers accessing the same shared database files
• Architecture: RAC requires cluster software, interconnect, and shared storage
• Complexity: RAC is more complex to install, configure, and maintain

8. What is Cache Fusion in Oracle RAC?

Answer: Cache Fusion is Oracle RAC's technology that enables sharing of cached data blocks between instances without writing to disk. It maintains data consistency across all instances by transferring data blocks directly through the private interconnect, eliminating the need for disk I/O for data sharing.

9. What is a VIP (Virtual IP) in Oracle RAC?

Answer: Virtual IP (VIP) is a virtual IP address assigned to each node in the RAC cluster. It provides transparent application failover by allowing client connections to automatically redirect to surviving nodes when a node fails, without waiting for TCP timeouts.

10. What is the purpose of SCAN (Single Client Access Name)?

Answer: SCAN provides a single name for clients to access any Oracle RAC database running in the cluster. It consists of:

• A single network name that resolves to multiple IP addresses
• Load balancing across available nodes
• Simplified client configuration
• Automatic failover capabilities

11. What is Grid Infrastructure?

Answer: Grid Infrastructure is the foundation software that enables Oracle RAC. It includes:

• Oracle Clusterware: Cluster management software
• Oracle ASM: Automatic Storage Management
• Oracle ACFS: ASM Cluster File System
• Network configuration: VIP, SCAN management

12. What are the different types of Oracle RAC configurations?

Answer:

• Oracle RAC: Traditional multi-instance configuration
• Oracle RAC One Node: Single instance with failover capability
• Oracle RAC on Extended Distance Clusters: Geographically dispersed nodes
• Oracle Flex Cluster: Hub and Leaf node architecture

13. What is the interconnect in Oracle RAC?

Answer: The interconnect is a private high-speed network that connects all nodes in the RAC cluster. It's used for:

• Cache Fusion traffic (data block transfers)
• Cluster heartbeat messages
• Inter-instance communication
• Lock management coordination

14. What is shared storage in Oracle RAC?

Answer: Shared storage is a storage system accessible by all nodes in the RAC cluster. It stores:

• Database files (datafiles, control files, redo logs)
• Grid Infrastructure files
• ASM disk groups
• Common requirements: Raw devices, ASM, or cluster file systems

15. What is Oracle Clusterware?

Answer: Oracle Clusterware is the cluster management software that:

• Manages cluster membership
• Provides high availability for cluster resources
• Monitors node health and performs automatic failover
• Manages VIPs, databases, and services
• Coordinates between cluster nodes

16. What are the prerequisites for Oracle RAC installation?

Answer:

• Hardware: Multiple servers with identical architecture
• Network: Public and private networks configured
• Storage: Shared storage accessible by all nodes
• Operating System: Certified OS with required packages
• Time Synchronization: NTP or CTSS configured
• DNS: Proper hostname resolution

17. What is the difference between OCR and Voting Disk?

Answer:

• OCR (Oracle Cluster Registry): Stores cluster configuration information, resource profiles, and policy information
• Voting Disk: Used for cluster membership decisions and prevents split-brain scenarios by providing quorum mechanism

18. What is a split-brain scenario in clustering?

Answer: Split-brain occurs when cluster nodes lose communication with each other but continue operating independently, potentially leading to data corruption. Oracle RAC prevents this using voting disks and quorum mechanisms to determine which nodes can continue operating.

19. What is TAF (Transparent Application Failover)?

Answer: TAF is an Oracle Net feature that provides automatic failover for client connections in case of node or instance failure. Types include:

• BASIC: Establishes new connection after failure
• PRECONNECT: Maintains backup connections
• SESSION: Recreates session state
• SELECT: Restores interrupted queries

20. What is FAN (Fast Application Notification)?

Answer: FAN is Oracle's mechanism for publishing cluster events to clients and applications. It provides:

• Immediate notification of node failures
• Load balancing advisories
• Service state changes
• Integration with connection pools for faster failover

RAC Architecture and Components (Questions 21-40)

21. Describe the Oracle RAC architecture components.

Answer: Oracle RAC architecture consists of:

• Shared Storage: Database files accessible by all nodes
• Grid Infrastructure: Clusterware and ASM
• Oracle Instances: Multiple instances accessing shared database
• Interconnect: Private network for inter-node communication
• Public Network: Client connectivity
• Virtual IPs: Failover mechanism for client connections

22. What is GRD (Global Resource Directory)?

Answer: GRD maintains cluster-wide resource information including:

• Lock ownership and modes
• Resource mastery assignments
• Block locations in different instances
• Coordinates resource access across instances
• Distributed across all instances for fault tolerance

23. What are the different Oracle RAC background processes?

Answer: Key RAC-specific background processes:

• LMS (Global Cache Service): Manages cache fusion
• LMD (Global Enqueue Service Daemon): Manages global locks
• LMON (Lock Monitor): Monitors global locks and performs recovery
• LCK (Lock Process): Handles non-cache fusion locks
• RMSn (Oracle RAC Management Processes): Coordinates UNDO and parallel query

24. What is GCS (Global Cache Service)?

Answer: GCS manages cache coherency across RAC instances by:

• Coordinating access to data blocks
• Managing block transfers between instances
• Maintaining consistency of cached data
• Implementing cache fusion technology
• Tracking block locations and states

25. What is GES (Global Enqueue Service)?

Answer: GES manages global locks and enqueues across RAC instances:

• Coordinates access to shared resources
• Manages row-level locks, table locks, and dictionary locks
• Prevents conflicts between instances
• Handles lock conversion and deadlock detection
• Maintains lock information in GRD

26. Explain the concept of instance membership in Oracle RAC.

Answer: Instance membership determines which instances are active in the cluster:

• Managed by Oracle Clusterware
• Uses voting disks for quorum decisions
• Monitors node health through heartbeat
• Automatically evicts failed nodes
• Triggers instance recovery when needed

27. What is GIPCD (Grid Inter-Process Communication Daemon)?

Answer: GIPCD is a Grid Infrastructure daemon that:

• Facilitates communication between Grid Infrastructure processes
• Provides OS-independent IPC mechanism
• Runs as a privileged user
• Essential for Grid Infrastructure startup
• Manages shared memory and semaphores

28. What is the role of MDNSD (Multicast Domain Name Service Daemon)?

Answer: MDNSD resolves cluster node names and provides:

• Local name resolution for cluster nodes
• Support for GNS (Grid Naming Service)
• Multicast DNS functionality
• Integration with cluster networking
• Automatic service discovery

29. What is GPNPD (Grid Plug and Play Profile Daemon)?

Answer: GPNPD manages Grid Plug and Play profiles:

• Stores cluster configuration information
• Provides centralized profile management
• Enables easy node addition to cluster
• Maintains network and storage configuration
• Supports cluster expansion operations

30. What are the different disk group types in ASM?

Answer: ASM disk group types:

• NORMAL: 2-way mirroring (default)
• HIGH: 3-way mirroring for maximum protection
• EXTERNAL: No ASM mirroring (relies on external redundancy)
• FLEX: Flexible redundancy with file-level mirroring options

31. What is the purpose of ASM instance?

Answer: ASM instance manages automatic storage:

• Manages disk groups and their metadata
• Provides storage virtualization
• Handles extent allocation and deallocation
• Performs automatic load balancing
• Runs on each node in RAC cluster

32. What is ACFS (ASM Cluster File System)?

Answer: ACFS is a cluster file system built on ASM:

• Provides POSIX-compliant file system
• Supports general-purpose file storage
• Offers high availability and scalability
• Includes snapshot and replication features
• Integrated with Grid Infrastructure

33. What is the difference between CRS and HAS?

Answer:

• CRS (Cluster Ready Services): Multi-node cluster stack
• HAS (High Availability Services): Single-node cluster stack
• CRS manages resources across multiple nodes
• HAS manages resources on single node (like Oracle Restart)
• Both provide high availability capabilities

34. What is Oracle Flex Cluster?

Answer: Oracle Flex Cluster is an enhanced cluster architecture:

• Hub Nodes: Full cluster members with direct shared storage access
• Leaf Nodes: Lightweight members accessing storage through Hub nodes
• Supports up to 64 Hub nodes and 64 Leaf nodes per cluster
• Provides better scalability and resource efficiency

35. What is Policy-Managed Database?

Answer: Policy-Managed Database automatically manages instance placement:

• Uses server pools to define node groups
• Automatically starts instances based on pool policies
• Provides dynamic load balancing
• Simplifies database administration
• Adapts to changing cluster membership

36. What is Server Pool in Oracle RAC?

Answer: Server Pools are logical groups of nodes:

• Define where specific databases run
• Enable workload isolation
• Support automatic instance management
• Provide resource allocation control
• Allow dynamic node assignment

37. What is the role of ONS (Oracle Notification Service)?

Answer: ONS provides cluster event notification:

• Publishes FAN events to applications
• Enables fast connection failover
• Supports load balancing advisories
• Integrates with connection pools
• Provides both local and remote notification

38. What is HAIP (Highly Available IP)?

Answer: HAIP provides redundancy for private interconnect:

• Creates virtual interfaces for redundancy
• Automatically fails over between network interfaces
• Provides load balancing across interfaces
• Simplifies interconnect configuration
• Available in Grid Infrastructure 11.2 and later

39. What is Cluster Verification Utility (CVU)?

Answer: CVU validates cluster configuration:

• Verifies prerequisites before installation
• Checks system requirements and configuration
• Validates network and storage setup
• Provides detailed reports on issues
• Helps troubleshoot cluster problems

40. What is Grid Naming Service (GNS)?

Answer: GNS provides dynamic name resolution:

• Automatically manages cluster name resolution
• Eliminates need for static DNS entries
• Supports dynamic VIP assignment
• Simplifies cluster network management
• Integrates with DHCP for IP assignment

Grid Infrastructure and ASM (Questions 41-55)

41. What are the components of Grid Infrastructure?

Answer: Grid Infrastructure includes:

• Oracle Clusterware: Cluster management and high availability
• Oracle ASM: Automatic Storage Management
• Oracle ACFS: ASM Cluster File System
• Network Services: VIP, SCAN, GNS management
• Time Synchronization: CTSS (Cluster Time Synchronization Service)

42. What is ASM disk group compatibility?

Answer: ASM disk group compatibility determines:

• ASM Compatibility: Minimum ASM version for disk group access
• Database Compatibility: Minimum database version for disk group access
• ADVM Compatibility: Minimum version for ASM Dynamic Volume Manager
• Enables new features while maintaining backward compatibility

43. What are ASM allocation units?

Answer: ASM allocation units are:

• Basic allocation units for space management
• Default size: 1MB (can be 1MB, 2MB, 4MB, 8MB, 16MB, 32MB, 64MB)
• Larger allocation units improve performance for large databases
• Cannot be changed after disk group creation
• Affects maximum file size and performance

44. What is ASM Fast Mirror Resync?

Answer: Fast Mirror Resync optimizes mirror resynchronization:

• Tracks changed extents during disk offline periods
• Only resynchronizes changed data upon disk return
• Significantly reduces resync time
• Maintains change tracking bitmap
• Automatic when disk offline period is short

45. What is ASM Preferred Read?

Answer: ASM Preferred Read optimizes read performance:

• Directs reads to preferred failure groups
• Reduces network traffic in extended clusters
• Configurable at disk group or file level
• Improves performance in geographically distributed clusters
• Automatically fails over to other copies if needed

46. What is the purpose of SPFILE in ASM?

Answer: SPFILE in ASM provides:

• Centralized parameter file storage
• Automatic backup and recovery
• Shared access across RAC instances
• Version control and change tracking
• High availability for critical configuration

47. What is ASM Dynamic Volume Manager (ADVM)?

Answer: ADVM provides:

• Volume management layer above ASM
• Support for non-Oracle files on ASM storage
• Foundation for ACFS implementation
• Dynamic volume creation and management
• Integration with Grid Infrastructure

48. What are ASM disk discovery methods?

Answer: ASM disk discovery methods:

• ASM_DISKSTRING: Pattern for disk discovery
• ASMLib: Oracle's kernel driver for disk management
• Raw devices: Direct device access
• Block devices: Standard block device access
• Multipath devices: For redundant storage paths

49. What is ASM rebalancing?

Answer: ASM rebalancing redistributes data:

• Automatically triggered when disks are added/removed
• Can be manually initiated
• Runs with configurable power levels (0-11)
• Maintains database availability during operation
• Optimizes storage utilization and performance

50. What is the purpose of voting disks in Grid Infrastructure?

Answer: Voting disks provide:

• Cluster membership quorum mechanism
• Prevention of split-brain scenarios
• Node eviction decisions during network partitions
• Cluster integrity maintenance
• Automatic redundancy management (odd number required)

51. What is OCR backup and recovery?

Answer: OCR backup includes:

• Automatic backups: Every 4 hours, daily, weekly
• Manual backups: Using ocrconfig utility
• Recovery: Restore from automatic or manual backups
• Multiple copies: Stored on different storage locations
• Export/Import: For disaster recovery scenarios

52. What is ASM disk group mounting process?

Answer: ASM disk group mounting involves:

1. Reading disk headers to identify disk group membership
2. Validating disk group metadata consistency
3. Constructing in-memory disk group structures
4. Making disk group available for database access
5. Enabling automatic rebalancing if needed

53. What is the role of ASM metadata?

Answer: ASM metadata includes:

• Disk group configuration: Disk membership and attributes
• File directory: File names and locations
• Extent maps: Physical-to-logical block mapping
• Free space maps: Available space tracking
• Alias directory: File aliases and symbolic links

54. What is ASM disk scrubbing?

Answer: ASM disk scrubbing:

• Proactively reads all allocated extents
• Detects and reports media failures
• Triggers automatic repair of corrupted blocks
• Runs as background maintenance activity
• Configurable through ASM parameters

55. What is Cluster Synchronization Services (CSS)?

Answer: CSS provides:

• Node membership management
• Cluster heartbeat mechanism
• Node eviction coordination
• Time synchronization services
• Integration with voting disk mechanism

RAC Administration (Questions 56-70)

56. How do you add a node to an existing RAC cluster?

Answer: Steps to add a node:

1. Prepare new node: Install OS, configure networking
2. Copy Grid Infrastructure: Clone or install Grid Infrastructure
3. Run addNode.sh: Add node to cluster configuration
4. Configure ASM: Extend ASM instance to new node
5. Add database instance: Create instance on new node
6. Update TNS configuration: Add new node to tnsnames.ora

57. How do you remove a node from RAC cluster?

Answer: Steps to remove a node:

1. Relocate services: Move services away from target node
2. Stop database instance: Shutdown instance on target node
3. Stop ASM instance: Shutdown ASM on target node
4. Run deleteNode.sh: Remove node from cluster
5. Clean up: Remove Grid Infrastructure from target node
6. Update configuration: Modify TNS and other configurations

58. How do you perform rolling upgrades in Oracle RAC?

Answer: Rolling upgrade process:

1. Upgrade Grid Infrastructure: Patch one node at a time
2. Upgrade Oracle Database: Out-of-place upgrade
3. Patch database instances: Rolling patch application
4. Maintain service availability: Keep other nodes running
5. Validate upgrade: Test functionality after each node

59. How do you manage services in Oracle RAC?

Answer: Service management using SRVCTL:

-- Create service
srvctl add service -d mydb -s myservice -r node1,node2

-- Start service
srvctl start service -d mydb -s myservice

-- Stop service
srvctl stop service -d mydb -s myservice

-- Relocate service
srvctl relocate service -d mydb -s myservice -i oldnode -t newnode

60. What is SRVCTL utility?

Answer: SRVCTL (Server Control) utility:

• Command-line tool for RAC administration
• Manages databases, instances, services, and listeners
• Integrates with Oracle Clusterware
• Provides cluster-aware start/stop operations
• Essential for RAC database management

61. How do you check RAC cluster status?

Answer: Commands to check cluster status:

# Cluster status
crsctl check cluster -all

# Resource status
crsctl status resource -t

# Database status
srvctl status database -d mydb

# Node applications
srvctl status nodeapps

# ASM status
srvctl status asm

62. How do you perform backup and recovery in RAC?

Answer: RAC backup and recovery:

• RMAN: Centralized backup from any node
• Archive log management: Coordinate across instances
• Shared backup location: Accessible by all nodes
• Recovery coordination: Single instance performs recovery
• Flashback features: Cluster-wide flashback operations

63. How do you manage undo in Oracle RAC?

Answer: RAC undo management:

• Automatic Undo Management: Each instance has its own undo tablespace
• Instance-specific: UNDO_TABLESPACE parameter per instance
• Sizing: Calculate based on longest transaction and retention
• Cleanup: Automatic cleanup of undo from failed instances
• Flashback: Requires proper undo retention configuration

64. How do you configure load balancing in RAC?

Answer: Load balancing configuration:

• Connection-time: Distribute new connections across nodes
• Runtime: Redirect connections based on node load
• Service-based: Configure preferred and available instances
• Client-side: Configure LOAD_BALANCE=yes in TNS
• Server-side: Use service weights and runtime balancing

65. How do you monitor RAC performance?

Answer: RAC performance monitoring:

• AWR reports: Cluster-wide and instance-specific reports
• Grid Control/Cloud Control: Centralized monitoring
• V$ views: RAC-specific dynamic performance views
• Enterprise Manager: Real-time performance monitoring
• Custom scripts: Monitor interconnect, wait events, and services

66. What are the important RAC-specific V$ views?

Answer: Key RAC V$ views:

• V$ACTIVE_INSTANCES: Active instances in cluster
• V$CLUSTER_INTERCONNECTS: Interconnect information
• V$GC_*: Global cache statistics
• V$GES_*: Global enqueue statistics
• V$INSTANCE: Instance information including thread#
• V$SERVICE_STATS: Service usage statistics

67. How do you handle patch management in RAC?

Answer: RAC patch management:

• Rolling patches: Apply without downtime
• Grid Infrastructure patches: Patch cluster software first
• Database patches: Apply to Oracle home
• Patch conflicts: Check for conflicts before application
• Validation: Test patches in development environment

68. How do you configure RAC for disaster recovery?

Answer: RAC disaster recovery setup:

• Data Guard: Primary RAC to standby RAC configuration
• Extended clusters: Geographically distributed nodes
• Storage replication: Synchronous or asynchronous replication
• Network redundancy: Multiple paths between sites
• Automated failover: Configure automatic disaster recovery

69. How do you troubleshoot interconnect issues?

Answer: Interconnect troubleshooting:

• Check network configuration: Verify IP addresses and routing
• Monitor network statistics: Check for packet loss and errors
• Review alert logs: Look for interconnect-related messages
• Use oifcfg: Verify interconnect interface configuration
• Network tools: ping, traceroute, iperf for network testing

70. How do you manage temporary tablespaces in RAC?

Answer: RAC temporary tablespace management:

• Shared temporary tablespace: Accessible by all instances
• Local temporary tablespaces: Instance-specific temporary spaces
• Automatic management: Use locally managed temporary tablespaces
• Sizing: Size based on concurrent user workload
• Performance: Monitor temporary segment usage across instances

Performance and Monitoring (Questions 71-85)

71. What are the key RAC-specific wait events?

Answer: Important RAC wait events:

• gc buffer busy: Block being modified on another instance
• gc current block 2-way/3-way: Block transfer from remote instance
• gc cr request: Consistent read block request
• gc current grant busy: Wait for current block grant
• enq: TX - allocate ITL entry: ITL shortages in RAC

72. How do you identify interconnect performance issues?

Answer: Interconnect performance analysis:

• Monitor GC wait events: High wait times indicate issues
• Check network utilization: Use OS tools like sar, iostat
• Review cluster statistics: V$CLUSTER_INTERCONNECTS
• Analyze AWR reports: Look for interconnect-related waits
• Network latency tests: Measure round-trip times

73. What is cache fusion and how does it affect performance?

Answer: Cache fusion performance impact:

• Positive: Eliminates disk I/O for shared data
• Negative: Network overhead for block transfers
• Latency: Interconnect latency affects response time
• Optimization: Proper application design reduces transfers
• Monitoring: Track gc wait events and transfer statistics

74. How do you optimize RAC performance?

Answer: RAC performance optimization:

• Application design: Minimize inter-instance communication
• Instance affinity: Direct related transactions to same instance
• Interconnect tuning: Use dedicated high-speed networks
• Storage optimization: Use ASM for automatic load balancing
• Service configuration: Optimize service placement and routing

75. What are Global Cache (GC) statistics?

Answer: GC statistics monitor cache fusion:

• gc blocks received: Blocks received from remote instances
• gc blocks served: Blocks served to remote instances
• gc current blocks received: Current mode blocks received
• gc cr blocks received: Consistent read blocks received
• gc blocks lost: Blocks lost due to communication failures

76. How do you monitor ASM performance in RAC?

Answer: ASM performance monitoring:

• V$ASM_DISKGROUP_STAT: Disk group I/O statistics
• V$ASM_DISK_STAT: Individual disk performance
• AWR ASM reports: Comprehensive ASM performance analysis
• asmcmd: Command-line monitoring tools
• Grid Control: GUI-based ASM monitoring

77. What is the impact of interconnect bandwidth on RAC?

Answer: Interconnect bandwidth impact:

• Low bandwidth: Increases gc wait events
• High bandwidth: Improves cache fusion performance
• Latency vs. bandwidth: Latency often more critical than bandwidth
• Multiple interfaces: Use bonding or HAIP for better throughput
• Monitoring: Track bytes sent/received across interconnect

78. How do you analyze RAC-specific AWR reports?

Answer: RAC AWR analysis focus areas:

• Global Cache Efficiency: Cache fusion hit ratios
• Interconnect Statistics: Block transfer rates and times
• Service Statistics: Load distribution across instances
• Wait Events: RAC-specific waits and their impact
• Instance Comparison: Performance variations between nodes

79. What are the common performance bottlenecks in RAC?

Answer: Common RAC bottlenecks:

• Interconnect contention: Insufficient bandwidth or high latency
• Lock contention: Global locks causing serialization
• Load imbalance: Uneven workload distribution
• Storage contention: Hot spots on shared storage
• Application design: Excessive inter-instance communication

80. How do you implement workload balancing in RAC?

Answer: Workload balancing strategies:

• Connection-time balancing: Distribute connections at startup
• Runtime balancing: Redirect based on current load
• Service-based routing: Use services to control placement
• Application partitioning: Partition data and users by instance
• Load balancing advisories: Use FAN events for dynamic balancing

81. What is the role of LMS (Lock Manager Service) in performance?

Answer: LMS process performance impact:

• Block transfers: Handles cache fusion block transfers
• CPU utilization: High LMS CPU indicates heavy cache fusion
• Multiple LMS processes: Scale with interconnect bandwidth
• Tuning: Adjust _gcs_server_processes parameter
• Monitoring: Track LMS wait events and CPU usage

82. How do you monitor RAC cluster performance using Enterprise Manager?

Answer: EM RAC monitoring features:

• Cluster database home page: Overall cluster health
• Performance pages: Real-time performance metrics
• Interconnect analysis: Network performance monitoring
• Service management: Service performance and availability
• Automatic alerts: Proactive problem notification

83. What are the key metrics for RAC health monitoring?

Answer: Essential RAC health metrics:

• Instance availability: All instances running and accessible
• Interconnect performance: Low latency and high throughput
• Global cache efficiency: High cache fusion hit ratios
• Service response time: Application-level performance
• Storage performance: Balanced I/O across disk groups

84. How do you troubleshoot slow queries in RAC environment?

Answer: RAC query troubleshooting:

• Identify wait events: Check for RAC-specific waits
• Execution plan analysis: Verify optimal plan selection
• Instance affinity: Ensure data locality when possible
• Parallel execution: Monitor cross-instance parallel execution
• Resource contention: Check for lock and latch contention

85. What is Cluster Ready Services (CRS) performance monitoring?

Answer: CRS performance monitoring:

• Resource monitoring: Track resource state changes
• Event processing: Monitor CRS event handling performance
• Voting disk I/O: Monitor voting disk response times
• OCR performance: Monitor OCR access times
• Log analysis: Review CRS logs for performance issues

Troubleshooting and Advanced Topics (Questions 86-100)

86. How do you troubleshoot node eviction in Oracle RAC?

Answer: Node eviction troubleshooting:

1. Check alert logs: Review CRS and database alert logs
2. Examine voting disk access: Verify storage connectivity
3. Network issues: Check interconnect connectivity
4. System resources: Monitor CPU, memory, and I/O
5. Time synchronization: Verify NTP configuration
6. Review cluster logs: Check ocssd.log for eviction reasons

87. What causes split-brain scenarios and how do you prevent them?

Answer: Split-brain prevention:

• Voting disks: Odd number provides quorum mechanism
• Network redundancy: Multiple interconnect paths
• Proper configuration: Correct voting disk placement
• Monitoring: Proactive network monitoring
• Storage reliability: Ensure voting disk accessibility

88. How do you troubleshoot ASM diskgroup mount failures?

Answer: ASM diskgroup troubleshooting:

1. Check disk visibility: Verify all disks are accessible
2. Examine ASM logs: Review ASM alert log for errors
3. Disk header validation: Check for corruption
4. Permission issues: Verify disk ownership and permissions
5. Network connectivity: Ensure shared storage access
6. Force mount: Use FORCE option if necessary

89. How do you resolve Oracle Clusterware startup issues?

Answer: Clusterware startup troubleshooting:

1. Check prerequisites: Verify network and storage configuration
2. Review logs: Examine CRS logs in $GRID_HOME/log
3. Voting disk access: Ensure voting disks are accessible
4. OCR integrity: Verify OCR is not corrupted
5. Dependencies: Check service dependencies and order
6. Manual startup: Use crsctl commands for manual intervention

90. What are common causes of high interconnect traffic?

Answer: High interconnect traffic causes:

• Poorly designed applications: Excessive cross-instance access
• Full table scans: Large table scans across instances
• Hot blocks: Frequently accessed blocks causing transfers
• Lock contention: Global locks increasing communication
• Parallel queries: Cross-instance parallel execution

91. How do you troubleshoot SCAN connectivity issues?

Answer: SCAN troubleshooting:

1. DNS resolution: Verify SCAN name resolves to multiple IPs
2. Network connectivity: Test connectivity to SCAN IPs
3. Listener status: Check SCAN listener status
4. Grid Infrastructure: Verify GI resources are online
5. Client configuration: Check client TNS configuration
6. Firewall settings: Ensure ports are open

92. How do you handle corrupted OCR?

Answer: OCR corruption handling:

1. Automatic recovery: OCR automatically recovers from mirrors
2. Manual restore: Restore from automatic backups
3. Recreation: Recreate OCR using cluster configuration
4. Export/import: Use OCR export for disaster recovery
5. Multiple copies: Maintain OCR on different storage types

93. What is the procedure for RAC database disaster recovery?

Answer: RAC disaster recovery procedure:

1. Assessment: Determine extent of failure
2. Failover: Switch to standby site if available
3. Recovery: Restore database from backups
4. Cluster rebuild: Reconstruct cluster if necessary
5. Synchronization: Resync primary and standby
6. Testing: Validate all functionality

94. How do you troubleshoot service relocation failures?

Answer: Service relocation troubleshooting:

1. Check service configuration: Verify preferred/available instances
2. Resource dependencies: Ensure all dependencies are satisfied
3. Network connectivity: Verify target node accessibility
4. Instance status: Confirm target instance is running
5. Resource limits: Check for resource constraints
6. Manual relocation: Use srvctl to manually relocate

95. What are the steps to recover from voting disk corruption?

Answer: Voting disk recovery:

1. Stop cluster: Shutdown all cluster nodes
2. Restore voting disks: Restore from backups or recreate
3. Update OCR: Update voting disk locations in OCR
4. Start cluster: Restart cluster services
5. Verify functionality: Test cluster operations
6. Backup: Create new backups after recovery

96. How do you troubleshoot TAF (Transparent Application Failover) issues?

Answer: TAF troubleshooting:

1. Check TAF configuration: Verify connect descriptor settings
2. Service configuration: Ensure TAF is enabled for service
3. Network connectivity: Test backup connection paths
4. Application compatibility: Verify application supports TAF
5. Monitor failover: Check TAF statistics and events
6. Testing: Simulate failures to test TAF functionality

97. What is the approach for debugging cache fusion performance?

Answer: Cache fusion debugging:

1. Wait event analysis: Focus on gc-related wait events
2. Transfer statistics: Monitor block transfer rates
3. Interconnect analysis: Check network performance
4. Application review: Identify inefficient access patterns
5. Instance placement: Optimize data and session placement
6. Hardware validation: Verify interconnect hardware performance

98. How do you resolve ASM disk rebalancing issues?

Answer: ASM rebalancing troubleshooting:

1. Check disk status: Verify all disks are online
2. Monitor progress: Use V$ASM_OPERATION for status
3. Resource allocation: Adjust rebalance power setting
4. Space availability: Ensure sufficient free space
5. Error analysis: Review ASM logs for errors
6. Manual intervention: Stop/restart rebalancing if needed

99. What are the best practices for RAC log file management?

Answer: RAC log management practices:

• Centralized logging: Use shared location for critical logs
• Log rotation: Implement automatic log rotation
• Retention policies: Define log retention periods
• Monitoring: Set up log monitoring and alerting
• Backup: Include logs in backup strategy
• Analysis tools: Use log analysis tools for troubleshooting

100. How do you perform root cause analysis for RAC issues?

Answer: RCA methodology for RAC:

1. Problem definition: Clearly define the issue and impact
2. Data collection: Gather logs, traces, and performance data
3. Timeline analysis: Establish chronology of events
4. Component analysis: Examine each RAC component
5. Pattern identification: Look for recurring patterns
6. Root cause identification: Determine underlying cause
7. Solution implementation: Apply appropriate fixes
8. Validation: Verify resolution and prevent recurrence

Conclusion

This comprehensive collection of 100 Oracle RAC interview questions covers the essential knowledge areas for database professionals working with Oracle Real Application Clusters. From basic concepts to advanced troubleshooting scenarios, these questions will help you:

• Prepare for interviews with confidence
• Assess your knowledge of Oracle RAC technologies
• Identify learning areas for professional development
• Understand best practices for RAC administration

Key Takeaways

1. Architecture Understanding: Master the fundamental components and their interactions
2. Practical Skills: Focus on hands-on administration and troubleshooting
3. Performance Optimization: Learn to identify and resolve performance bottlenecks
4. High Availability: Understand how to maintain service availability
5. Troubleshooting: Develop systematic approaches to problem resolution

Next Steps

• Practice with hands-on RAC environments
• Study Oracle documentation and best practices guides
• Join Oracle communities and forums for knowledge sharing
• Consider Oracle certification paths for formal validation
• Stay updated with latest Oracle RAC features and enhancements

Whether you're preparing for an interview, planning a certification, or enhancing your database administration skills, this guide provides a solid foundation for Oracle RAC expertise.

Ready to advance your Oracle RAC career? Explore our comprehensive database administration courses and gain hands-on experience with expert-led training programs.