Skip to content

3.15.1 — 2025-11-30

← 3.15.2 | 3.15.0 → | ↑ 3.x series

Added

Database Adapters - ScyllaDB Enhancements

  • Retry Policies - Added configurable retry policies for handling transient failures in ScyllaDB adapter

    • Exponential backoff retry policy with configurable min/max intervals and max retries
    • Fallthrough retry policy for no-retry scenarios
    • Downgrading consistency retry policy for automatic consistency level adjustment
    • Configuration options: RETRY_POLICY, RETRY_MAX_NUM_RETRIES, RETRY_MIN_INTERVAL, RETRY_MAX_INTERVAL

  • Health Checks - Added comprehensive health check functionality

    • Connection status verification with is_connected() method for both sync and async adapters
    • Detailed health metrics including cluster state, host availability, and latency
    • health_check() method returning comprehensive cluster health information
    • Configurable health check timeout via HEALTH_CHECK_TIMEOUT configuration

  • Exception Handling - Improved error handling with centralized exception management

    • Centralized _handle_scylladb_exception() method for consistent error handling
    • Specific error messages for connection, query execution, and configuration issues
    • Proper exception chaining for better debugging and error tracing

  • Helper Methods - Added convenience methods for common database operations

    • insert() method with TTL support for time-based data expiration
    • update() method with conditional updates and TTL support
    • count() method for counting rows with optional WHERE clause filtering
    • exists() method for checking record existence based on conditions
    • close() method for explicit connection cleanup and resource management

  • Connection Pool Monitoring - Added connection pool statistics and monitoring capabilities

    • get_pool_stats() method providing detailed metrics on pool utilization
    • Metrics include connections per host, in-flight requests, and pool health
    • Configurable pool parameters: MAX_CONNECTIONS_PER_HOST, MIN_CONNECTIONS_PER_HOST, CORE_CONNECTIONS_PER_HOST
    • MAX_REQUESTS_PER_CONNECTION configuration for request throttling
    • Optional pool monitoring via ENABLE_CONNECTION_POOL_MONITORING flag

  • Prepared Statement Caching - Implemented caching mechanism for prepared statements

    • LRU cache with configurable size via PREPARED_STATEMENT_CACHE_SIZE
    • Configurable TTL via PREPARED_STATEMENT_CACHE_TTL_SECONDS
    • Automatic cache invalidation after TTL expiration
    • Enable/disable via ENABLE_PREPARED_STATEMENT_CACHE configuration
    • Improved performance for frequently executed queries

  • Data Center Awareness - Added support for multi-datacenter deployments

    • Local datacenter configuration via LOCAL_DC for optimized query routing
    • Replication strategy configuration: REPLICATION_STRATEGY (SimpleStrategy, NetworkTopologyStrategy)
    • Per-datacenter replication factor configuration via REPLICATION_CONFIG
    • Improved latency for geographically distributed deployments

Documentation

  • ScyllaDB Adapter Documentation - Added comprehensive documentation for ScyllaDB adapter
    • Complete usage examples for all adapter methods
    • Configuration guide with all available options
    • Best practices for connection pooling and performance tuning
    • Examples for retry policies, health checks, and TTL usage
    • Multi-datacenter deployment configuration examples

Testing

  • Enhanced BDD Test Suite - Expanded Behave test scenarios for ScyllaDB adapter
    • Added test scenarios for TTL functionality
    • Added test scenarios for helper methods (count, exists, insert, update)
    • Added test scenarios for health checks and connection status
    • Added test scenarios for connection pool monitoring
    • Improved test coverage for error handling and edge cases