truthound-dashboard 1.3.1__py3-none-any.whl → 1.4.1__py3-none-any.whl

truthound_dashboard/core/notifications/throttling/stores.py

@@ -0,0 +1,1859 @@
+"""Storage backends for throttling state.
+
+This module provides storage backends for tracking rate limit
+counters and token buckets.
+
+Storage Backends:
+    - InMemoryThrottlingStore: Simple in-memory storage with TTL and LRU eviction
+    - SQLiteThrottlingStore: Persistent SQLite storage
+    - RedisThrottlingStore: Redis-based storage for distributed deployments
+
+Features:
+    - TTL-based automatic expiration of old entries
+    - LRU eviction when max entries exceeded
+    - Periodic background cleanup task
+    - Configurable cleanup interval and max entries
+    - Memory usage metrics
+"""
+
+from __future__ import annotations
+
+import asyncio
+import json
+import logging
+import os
+import sqlite3
+import threading
+import time
+import weakref
+from abc import ABC, abstractmethod
+from collections import OrderedDict
+from dataclasses import dataclass, field
+from pathlib import Path
+from typing import TYPE_CHECKING, Any
+
+# Optional Redis dependency
+try:
+    import redis
+    import redis.asyncio
+
+    REDIS_AVAILABLE = True
+except ImportError:
+    REDIS_AVAILABLE = False
+    redis = None  # type: ignore[assignment]
+
+if TYPE_CHECKING:
+    import redis as redis_sync
+    import redis.asyncio as redis_async
+
+
+logger = logging.getLogger(__name__)
+
+
+@dataclass
+class ThrottlingEntry:
+    """A stored throttling entry.
+
+    Attributes:
+        key: Unique key identifying the throttled entity.
+        count: Current count within the window.
+        window_start: Start of the current window.
+        tokens: Current token count (for token bucket).
+        last_refill: Last token refill time.
+        last_accessed: Last access timestamp for LRU tracking.
+    """
+
+    key: str
+    count: int = 0
+    window_start: float = 0.0
+    tokens: float = 0.0
+    last_refill: float = 0.0
+    last_accessed: float = field(default_factory=time.time)
+    metadata: dict[str, Any] = field(default_factory=dict)
+
+
+@dataclass
+class ThrottlingMetrics:
+    """Metrics for throttling store operations.
+
+    Attributes:
+        total_entries: Current number of entries.
+        peak_entries: Maximum entries ever stored.
+        expired_removed: Number of entries removed by TTL expiration.
+        lru_evicted: Number of entries evicted by LRU policy.
+        cleanup_runs: Number of cleanup cycles executed.
+        last_cleanup_time: Timestamp of last cleanup.
+        last_cleanup_removed: Entries removed in last cleanup.
+        memory_bytes_estimate: Estimated memory usage in bytes.
+    """
+
+    total_entries: int = 0
+    peak_entries: int = 0
+    expired_removed: int = 0
+    lru_evicted: int = 0
+    cleanup_runs: int = 0
+    last_cleanup_time: float = 0.0
+    last_cleanup_removed: int = 0
+    memory_bytes_estimate: int = 0
+
+    def to_dict(self) -> dict[str, Any]:
+        """Convert metrics to dictionary."""
+        return {
+            "total_entries": self.total_entries,
+            "peak_entries": self.peak_entries,
+            "expired_removed": self.expired_removed,
+            "lru_evicted": self.lru_evicted,
+            "cleanup_runs": self.cleanup_runs,
+            "last_cleanup_time": self.last_cleanup_time,
+            "last_cleanup_removed": self.last_cleanup_removed,
+            "memory_bytes_estimate": self.memory_bytes_estimate,
+        }
+
+
+class BaseThrottlingStore(ABC):
+    """Abstract base class for throttling storage.
+
+    All stores must implement methods for getting and updating
+    throttling state.
+    """
+
+    @abstractmethod
+    def get(self, key: str) -> ThrottlingEntry | None:
+        """Get entry by key.
+
+        Args:
+            key: The throttling key.
+
+        Returns:
+            Entry if found, None otherwise.
+        """
+        ...
+
+    @abstractmethod
+    def set(self, entry: ThrottlingEntry) -> None:
+        """Set or update an entry.
+
+        Args:
+            entry: The entry to store.
+        """
+        ...
+
+    @abstractmethod
+    def increment(self, key: str, window_start: float) -> int:
+        """Increment counter and return new count.
+
+        If the entry doesn't exist or window has changed,
+        creates a new entry with count=1.
+
+        Args:
+            key: The throttling key.
+            window_start: Start of current window.
+
+        Returns:
+            New count value.
+        """
+        ...
+
+    @abstractmethod
+    def cleanup(self, max_age_seconds: int) -> int:
+        """Remove old entries.
+
+        Args:
+            max_age_seconds: Maximum age of entries to keep.
+
+        Returns:
+            Number of entries removed.
+        """
+        ...
+
+    @abstractmethod
+    def clear(self) -> None:
+        """Clear all entries."""
+        ...
+
+    def get_metrics(self) -> dict[str, Any]:
+        """Get store metrics.
+
+        Returns:
+            Dictionary with metrics data.
+        """
+        return {}
+
+
+class InMemoryThrottlingStore(BaseThrottlingStore):
+    """In-memory throttling storage with TTL and LRU eviction.
+
+    Thread-safe storage suitable for development and single-process
+    deployments with automatic memory management.
+
+    Features:
+        - TTL-based automatic expiration of old entries
+        - LRU eviction when max entries exceeded
+        - Periodic background cleanup task
+        - Configurable cleanup interval and max entries
+        - Memory usage metrics
+
+    Configuration via environment variables:
+        TRUTHOUND_THROTTLE_MAX_ENTRIES: Maximum entries (default: 10000)
+        TRUTHOUND_THROTTLE_DEFAULT_TTL: Default TTL in seconds (default: 3600)
+        TRUTHOUND_THROTTLE_CLEANUP_INTERVAL: Cleanup interval in seconds (default: 60)
+        TRUTHOUND_THROTTLE_LRU_EVICTION_PERCENT: Percent to evict on overflow (default: 10)
+
+    Example:
+        # Basic usage
+        store = InMemoryThrottlingStore()
+
+        # Custom configuration
+        store = InMemoryThrottlingStore(
+            max_entries=50000,
+            default_ttl=7200,
+            cleanup_interval=120,
+        )
+
+        # With background cleanup
+        await store.start_background_cleanup()
+        # ... use store ...
+        await store.stop_background_cleanup()
+    """
+
+    # Track all instances for proper cleanup
+    _instances: weakref.WeakSet["InMemoryThrottlingStore"] = weakref.WeakSet()
+
+    def __init__(
+        self,
+        max_entries: int | None = None,
+        default_ttl: int | None = None,
+        cleanup_interval: int | None = None,
+        lru_eviction_percent: int | None = None,
+        auto_start_cleanup: bool = False,
+    ) -> None:
+        """Initialize in-memory store with memory management.
+
+        Args:
+            max_entries: Maximum number of entries before LRU eviction.
+            default_ttl: Default TTL in seconds for entries.
+            cleanup_interval: Interval in seconds between cleanup runs.
+            lru_eviction_percent: Percentage of entries to evict on overflow.
+            auto_start_cleanup: Whether to auto-start background cleanup.
+        """
+        # Configuration from environment or parameters
+        self.max_entries = max_entries or int(
+            os.getenv("TRUTHOUND_THROTTLE_MAX_ENTRIES", "10000")
+        )
+        self.default_ttl = default_ttl or int(
+            os.getenv("TRUTHOUND_THROTTLE_DEFAULT_TTL", "3600")
+        )
+        self.cleanup_interval = cleanup_interval or int(
+            os.getenv("TRUTHOUND_THROTTLE_CLEANUP_INTERVAL", "60")
+        )
+        self.lru_eviction_percent = lru_eviction_percent or int(
+            os.getenv("TRUTHOUND_THROTTLE_LRU_EVICTION_PERCENT", "10")
+        )
+
+        # Use OrderedDict for LRU tracking
+        self._entries: OrderedDict[str, ThrottlingEntry] = OrderedDict()
+        self._lock = threading.RLock()
+
+        # Background cleanup task
+        self._cleanup_task: asyncio.Task[None] | None = None
+        self._cleanup_running = False
+        self._shutdown_event: asyncio.Event | None = None
+
+        # Metrics
+        self._metrics = ThrottlingMetrics()
+
+        # Track instance for cleanup
+        InMemoryThrottlingStore._instances.add(self)
+
+        # Auto-start background cleanup if requested
+        self._auto_start_cleanup = auto_start_cleanup
+
+    def get(self, key: str) -> ThrottlingEntry | None:
+        """Get entry by key and update access time for LRU."""
+        with self._lock:
+            entry = self._entries.get(key)
+            if entry is not None:
+                # Update access time
+                entry.last_accessed = time.time()
+                # Move to end for LRU (most recently used)
+                self._entries.move_to_end(key)
+            return entry
+
+    def set(self, entry: ThrottlingEntry) -> None:
+        """Set or update an entry with LRU tracking."""
+        entry.last_accessed = time.time()
+
+        with self._lock:
+            # Check if we need to evict entries
+            if len(self._entries) >= self.max_entries and entry.key not in self._entries:
+                self._evict_lru()
+
+            # Set entry
+            self._entries[entry.key] = entry
+            # Move to end for LRU
+            self._entries.move_to_end(entry.key)
+
+            # Update metrics
+            self._metrics.total_entries = len(self._entries)
+            if self._metrics.total_entries > self._metrics.peak_entries:
+                self._metrics.peak_entries = self._metrics.total_entries
+            self._update_memory_estimate()
+
+    def increment(self, key: str, window_start: float) -> int:
+        """Increment counter and return new count."""
+        with self._lock:
+            entry = self._entries.get(key)
+
+            if entry is None or entry.window_start != window_start:
+                # Check if we need to evict
+                if len(self._entries) >= self.max_entries and key not in self._entries:
+                    self._evict_lru()
+
+                # New window
+                entry = ThrottlingEntry(
+                    key=key,
+                    count=1,
+                    window_start=window_start,
+                    last_accessed=time.time(),
+                )
+                self._entries[key] = entry
+
+                # Update metrics
+                self._metrics.total_entries = len(self._entries)
+                if self._metrics.total_entries > self._metrics.peak_entries:
+                    self._metrics.peak_entries = self._metrics.total_entries
+
+                return 1
+
+            # Same window, increment
+            entry.count += 1
+            entry.last_accessed = time.time()
+            # Move to end for LRU
+            self._entries.move_to_end(key)
+
+            return entry.count
+
+    def cleanup(self, max_age_seconds: int) -> int:
+        """Remove old entries based on TTL."""
+        cutoff = time.time() - max_age_seconds
+        removed = 0
+
+        with self._lock:
+            # Collect expired keys
+            expired = [
+                key for key, entry in self._entries.items()
+                if entry.window_start < cutoff and entry.last_refill < cutoff
+            ]
+
+            # Remove expired entries
+            for key in expired:
+                del self._entries[key]
+                removed += 1
+
+            # Update metrics
+            self._metrics.expired_removed += removed
+            self._metrics.total_entries = len(self._entries)
+            self._metrics.cleanup_runs += 1
+            self._metrics.last_cleanup_time = time.time()
+            self._metrics.last_cleanup_removed = removed
+            self._update_memory_estimate()
+
+        if removed > 0:
+            logger.debug(f"Throttling store cleanup: removed {removed} expired entries")
+
+        return removed
+
+    def _evict_lru(self) -> None:
+        """Evict least recently used entries.
+
+        Called when max_entries is reached. Evicts a percentage
+        of entries based on lru_eviction_percent.
+        """
+        evict_count = max(1, int(len(self._entries) * self.lru_eviction_percent / 100))
+
+        # Remove oldest entries (at the beginning of OrderedDict)
+        for _ in range(evict_count):
+            if not self._entries:
+                break
+            # popitem(last=False) removes from beginning (oldest)
+            self._entries.popitem(last=False)
+            self._metrics.lru_evicted += 1
+
+        self._metrics.total_entries = len(self._entries)
+
+        logger.debug(f"Throttling store LRU eviction: evicted {evict_count} entries")
+
+    def _update_memory_estimate(self) -> None:
+        """Estimate memory usage of stored entries."""
+        # Rough estimate: key (~50 bytes) + entry object (~200 bytes)
+        entry_size_estimate = 250
+        self._metrics.memory_bytes_estimate = len(self._entries) * entry_size_estimate
+
+    def clear(self) -> None:
+        """Clear all entries."""
+        with self._lock:
+            self._entries.clear()
+            self._metrics.total_entries = 0
+            self._update_memory_estimate()
+
+    def get_metrics(self) -> dict[str, Any]:
+        """Get store metrics."""
+        with self._lock:
+            self._metrics.total_entries = len(self._entries)
+            self._update_memory_estimate()
+            return self._metrics.to_dict()
+
+    def count(self) -> int:
+        """Get total entry count."""
+        with self._lock:
+            return len(self._entries)
+
+    # =========================================================================
+    # Background Cleanup
+    # =========================================================================
+
+    async def start_background_cleanup(self) -> None:
+        """Start background cleanup task.
+
+        Creates an asyncio task that periodically runs cleanup
+        to remove expired entries.
+        """
+        if self._cleanup_task is not None and not self._cleanup_task.done():
+            logger.warning("Background cleanup already running")
+            return
+
+        self._cleanup_running = True
+        self._shutdown_event = asyncio.Event()
+        self._cleanup_task = asyncio.create_task(self._background_cleanup_loop())
+        logger.info(
+            f"Started throttling store background cleanup "
+            f"(interval={self.cleanup_interval}s, ttl={self.default_ttl}s)"
+        )
+
+    async def stop_background_cleanup(self) -> None:
+        """Stop background cleanup task."""
+        if not self._cleanup_running:
+            return
+
+        self._cleanup_running = False
+
+        if self._shutdown_event:
+            self._shutdown_event.set()
+
+        if self._cleanup_task:
+            try:
+                # Wait for task to complete with timeout
+                await asyncio.wait_for(self._cleanup_task, timeout=5.0)
+            except asyncio.TimeoutError:
+                self._cleanup_task.cancel()
+                try:
+                    await self._cleanup_task
+                except asyncio.CancelledError:
+                    pass
+
+        self._cleanup_task = None
+        self._shutdown_event = None
+        logger.info("Stopped throttling store background cleanup")
+
+    async def _background_cleanup_loop(self) -> None:
+        """Background task that periodically runs cleanup."""
+        while self._cleanup_running:
+            try:
+                # Wait for cleanup interval or shutdown
+                if self._shutdown_event:
+                    try:
+                        await asyncio.wait_for(
+                            self._shutdown_event.wait(),
+                            timeout=self.cleanup_interval,
+                        )
+                        # If we get here, shutdown was requested
+                        break
+                    except asyncio.TimeoutError:
+                        # Timeout means it's time to cleanup
+                        pass
+
+                # Run cleanup
+                removed = self.cleanup(self.default_ttl)
+                if removed > 0:
+                    logger.debug(
+                        f"Background cleanup: removed {removed} expired entries, "
+                        f"{self.count()} remaining"
+                    )
+
+            except asyncio.CancelledError:
+                logger.debug("Background cleanup task cancelled")
+                break
+            except Exception as e:
+                logger.error(f"Error in background cleanup: {e}")
+                # Continue running despite errors
+                await asyncio.sleep(self.cleanup_interval)
+
+    def __del__(self) -> None:
+        """Cleanup on deletion."""
+        # Note: Cannot await in __del__, so just flag for cleanup
+        self._cleanup_running = False
+
+    # =========================================================================
+    # Context Manager Support
+    # =========================================================================
+
+    async def __aenter__(self) -> "InMemoryThrottlingStore":
+        """Async context manager entry."""
+        if self._auto_start_cleanup:
+            await self.start_background_cleanup()
+        return self
+
+    async def __aexit__(
+        self, exc_type: Any, exc_val: Any, exc_tb: Any
+    ) -> None:
+        """Async context manager exit."""
+        await self.stop_background_cleanup()
+
+
+class SQLiteThrottlingStore(BaseThrottlingStore):
+    """SQLite-based persistent throttling storage.
+
+    Provides durable storage that survives process restarts.
+    """
+
+    def __init__(self, db_path: str | Path = "throttling.db") -> None:
+        """Initialize SQLite store.
+
+        Args:
+            db_path: Path to database file.
+        """
+        self.db_path = Path(db_path)
+        self._local = threading.local()
+        self._init_db()
+
+    def _get_connection(self) -> sqlite3.Connection:
+        """Get thread-local database connection."""
+        if not hasattr(self._local, "connection"):
+            self._local.connection = sqlite3.connect(
+                str(self.db_path),
+                check_same_thread=False,
+            )
+            self._local.connection.row_factory = sqlite3.Row
+        return self._local.connection
+
+    def _init_db(self) -> None:
+        """Initialize database schema."""
+        conn = self._get_connection()
+        conn.execute("""
+            CREATE TABLE IF NOT EXISTS throttling_entries (
+                key TEXT PRIMARY KEY,
+                count INTEGER NOT NULL DEFAULT 0,
+                window_start REAL NOT NULL,
+                tokens REAL NOT NULL DEFAULT 0,
+                last_refill REAL NOT NULL DEFAULT 0,
+                last_accessed REAL NOT NULL DEFAULT 0,
+                metadata TEXT
+            )
+        """)
+        conn.execute("""
+            CREATE INDEX IF NOT EXISTS idx_throttle_window
+            ON throttling_entries(window_start)
+        """)
+        conn.execute("""
+            CREATE INDEX IF NOT EXISTS idx_throttle_accessed
+            ON throttling_entries(last_accessed)
+        """)
+        conn.commit()
+
+    def get(self, key: str) -> ThrottlingEntry | None:
+        """Get entry by key."""
+        conn = self._get_connection()
+        now = time.time()
+
+        # Update last_accessed
+        cursor = conn.execute(
+            """
+            UPDATE throttling_entries
+            SET last_accessed = ?
+            WHERE key = ?
+            RETURNING key, count, window_start, tokens, last_refill, last_accessed, metadata
+            """,
+            (now, key),
+        )
+        row = cursor.fetchone()
+
+        if row is None:
+            # Fall back to SELECT if UPDATE returned nothing
+            cursor = conn.execute(
+                """
+                SELECT key, count, window_start, tokens, last_refill, last_accessed, metadata
+                FROM throttling_entries
+                WHERE key = ?
+                """,
+                (key,),
+            )
+            row = cursor.fetchone()
+
+        if row is None:
+            return None
+
+        metadata = {}
+        if row["metadata"]:
+            try:
+                metadata = json.loads(row["metadata"])
+            except json.JSONDecodeError:
+                pass
+
+        return ThrottlingEntry(
+            key=row["key"],
+            count=row["count"],
+            window_start=row["window_start"],
+            tokens=row["tokens"],
+            last_refill=row["last_refill"],
+            last_accessed=row["last_accessed"],
+            metadata=metadata,
+        )
+
+    def set(self, entry: ThrottlingEntry) -> None:
+        """Set or update an entry."""
+        conn = self._get_connection()
+        metadata_json = json.dumps(entry.metadata) if entry.metadata else None
+        now = time.time()
+
+        conn.execute(
+            """
+            INSERT OR REPLACE INTO throttling_entries
+            (key, count, window_start, tokens, last_refill, last_accessed, metadata)
+            VALUES (?, ?, ?, ?, ?, ?, ?)
+            """,
+            (
+                entry.key,
+                entry.count,
+                entry.window_start,
+                entry.tokens,
+                entry.last_refill,
+                now,
+                metadata_json,
+            ),
+        )
+        conn.commit()
+
+    def increment(self, key: str, window_start: float) -> int:
+        """Increment counter and return new count."""
+        conn = self._get_connection()
+        now = time.time()
+
+        # Check if entry exists and is in same window
+        cursor = conn.execute(
+            """
+            SELECT count, window_start FROM throttling_entries
+            WHERE key = ?
+            """,
+            (key,),
+        )
+        row = cursor.fetchone()
+
+        if row is None or row["window_start"] != window_start:
+            # New window
+            conn.execute(
+                """
+                INSERT OR REPLACE INTO throttling_entries
+                (key, count, window_start, tokens, last_refill, last_accessed)
+                VALUES (?, 1, ?, 0, 0, ?)
+                """,
+                (key, window_start, now),
+            )
+            conn.commit()
+            return 1
+
+        # Same window, increment
+        conn.execute(
+            """
+            UPDATE throttling_entries
+            SET count = count + 1, last_accessed = ?
+            WHERE key = ?
+            """,
+            (now, key),
+        )
+        conn.commit()
+
+        return row["count"] + 1
+
+    def cleanup(self, max_age_seconds: int) -> int:
+        """Remove old entries."""
+        conn = self._get_connection()
+        cutoff = time.time() - max_age_seconds
+
+        cursor = conn.execute(
+            """
+            DELETE FROM throttling_entries
+            WHERE window_start < ? AND last_refill < ?
+            """,
+            (cutoff, cutoff),
+        )
+        conn.commit()
+
+        return cursor.rowcount
+
+    def clear(self) -> None:
+        """Clear all entries."""
+        conn = self._get_connection()
+        conn.execute("DELETE FROM throttling_entries")
+        conn.commit()
+
+    def count(self) -> int:
+        """Get total entry count."""
+        conn = self._get_connection()
+        cursor = conn.execute("SELECT COUNT(*) FROM throttling_entries")
+        return cursor.fetchone()[0]
+
+    def get_metrics(self) -> dict[str, Any]:
+        """Get store metrics."""
+        return {
+            "total_entries": self.count(),
+            "db_path": str(self.db_path),
+        }
+
+    def close(self) -> None:
+        """Close database connection."""
+        if hasattr(self._local, "connection"):
+            self._local.connection.close()
+            del self._local.connection
+
+
+class RedisThrottlingStore(BaseThrottlingStore):
+    """Redis-based throttling store for distributed deployments.
+
+    Uses Redis for shared throttling state across multiple processes
+    or containers. Supports both sync and async operations with
+    connection pooling.
+
+    Features:
+        - Connection pool management with configurable pool size
+        - Automatic reconnection with exponential backoff
+        - TTL management for automatic expiration
+        - Graceful degradation (fallback to InMemory on Redis failure)
+        - Health check endpoint support
+        - Comprehensive metrics collection
+        - Async support for high-performance applications
+
+    Configuration via environment variables:
+        TRUTHOUND_THROTTLE_REDIS_URL: Redis connection URL (default: redis://localhost:6379/0)
+        TRUTHOUND_THROTTLE_REDIS_PREFIX: Key prefix (default: truthound:throttle:)
+        TRUTHOUND_THROTTLE_REDIS_TTL: Default TTL in seconds (default: 3600)
+        TRUTHOUND_THROTTLE_REDIS_POOL_SIZE: Connection pool size (default: 10)
+        TRUTHOUND_THROTTLE_REDIS_SOCKET_TIMEOUT: Socket timeout (default: 5.0)
+        TRUTHOUND_THROTTLE_REDIS_CONNECT_TIMEOUT: Connection timeout (default: 5.0)
+        TRUTHOUND_THROTTLE_REDIS_MAX_RETRIES: Max retry attempts (default: 3)
+        TRUTHOUND_THROTTLE_REDIS_RETRY_BASE_DELAY: Base delay for exponential backoff (default: 1.0)
+        TRUTHOUND_THROTTLE_FALLBACK_ENABLED: Enable fallback to InMemory (default: true)
+
+    Example:
+        # Basic usage
+        store = RedisThrottlingStore()
+
+        # Custom configuration
+        store = RedisThrottlingStore(
+            redis_url="redis://myredis:6379/1",
+            default_ttl=7200,
+            max_connections=20,
+        )
+
+        # With context manager
+        async with RedisThrottlingStore() as store:
+            result = throttler.allow("channel-1")
+
+    Note: Requires the 'redis' optional dependency.
+          Install with: pip install truthound-dashboard[redis]
+    """
+
+    # Hash field names
+    FIELD_COUNT = "count"
+    FIELD_WINDOW_START = "window_start"
+    FIELD_TOKENS = "tokens"
+    FIELD_LAST_REFILL = "last_refill"
+    FIELD_LAST_ACCESSED = "last_accessed"
+    FIELD_METADATA = "metadata"
+
+    def __init__(
+        self,
+        redis_url: str | None = None,
+        key_prefix: str | None = None,
+        default_ttl: int | None = None,
+        max_connections: int | None = None,
+        socket_timeout: float | None = None,
+        socket_connect_timeout: float | None = None,
+        max_retries: int | None = None,
+        retry_base_delay: float | None = None,
+        enable_fallback: bool | None = None,
+    ) -> None:
+        """Initialize Redis throttling store.
+
+        All parameters can be configured via environment variables if not
+        explicitly provided.
+
+        Args:
+            redis_url: Redis connection URL.
+            key_prefix: Prefix for all Redis keys.
+            default_ttl: Default TTL in seconds for entries.
+            max_connections: Maximum connections in the pool.
+            socket_timeout: Socket timeout in seconds.
+            socket_connect_timeout: Connection timeout in seconds.
+            max_retries: Maximum retry attempts for reconnection.
+            retry_base_delay: Base delay for exponential backoff.
+            enable_fallback: Enable fallback to InMemory on Redis failure.
+
+        Raises:
+            ImportError: If redis package is not installed.
+        """
+        if not REDIS_AVAILABLE:
+            raise ImportError(
+                "Redis support requires the 'redis' package. "
+                "Install with: pip install truthound-dashboard[redis] "
+                "or pip install redis"
+            )
+
+        # Configuration from environment or parameters
+        self.redis_url = redis_url or os.getenv(
+            "TRUTHOUND_THROTTLE_REDIS_URL", "redis://localhost:6379/0"
+        )
+        self.key_prefix = key_prefix or os.getenv(
+            "TRUTHOUND_THROTTLE_REDIS_PREFIX", "truthound:throttle:"
+        )
+        self.default_ttl = default_ttl or int(
+            os.getenv("TRUTHOUND_THROTTLE_REDIS_TTL", "3600")
+        )
+        self.max_connections = max_connections or int(
+            os.getenv("TRUTHOUND_THROTTLE_REDIS_POOL_SIZE", "10")
+        )
+        self.socket_timeout = socket_timeout or float(
+            os.getenv("TRUTHOUND_THROTTLE_REDIS_SOCKET_TIMEOUT", "5.0")
+        )
+        self.socket_connect_timeout = socket_connect_timeout or float(
+            os.getenv("TRUTHOUND_THROTTLE_REDIS_CONNECT_TIMEOUT", "5.0")
+        )
+        self.max_retries = max_retries or int(
+            os.getenv("TRUTHOUND_THROTTLE_REDIS_MAX_RETRIES", "3")
+        )
+        self.retry_base_delay = retry_base_delay or float(
+            os.getenv("TRUTHOUND_THROTTLE_REDIS_RETRY_BASE_DELAY", "1.0")
+        )
+
+        fallback_env = os.getenv("TRUTHOUND_THROTTLE_FALLBACK_ENABLED", "true")
+        self.enable_fallback = (
+            enable_fallback
+            if enable_fallback is not None
+            else fallback_env.lower() == "true"
+        )
+
+        # Connection pool for sync client
+        self._pool: redis.ConnectionPool | None = None
+        self._client: redis.Redis | None = None
+
+        # Connection pool for async client
+        self._async_pool: redis.asyncio.ConnectionPool | None = None
+        self._async_client: redis.asyncio.Redis | None = None
+
+        # Locks for thread-safe initialization
+        self._lock = threading.Lock()
+        self._async_lock: asyncio.Lock | None = None
+
+        # Fallback store for graceful degradation
+        self._fallback_store: InMemoryThrottlingStore | None = None
+        self._using_fallback = False
+
+        # Connection state tracking
+        self._connected = False
+        self._retry_count = 0
+        self._last_error: Exception | None = None
+        self._last_error_time: float | None = None
+
+        # Metrics
+        self._metrics = ThrottlingMetrics()
+        self._redis_errors = 0
+        self._fallback_count = 0
+        self._reconnections = 0
+
+        # Index tracking key
+        self._index_key = f"{self.key_prefix}index"
+
+    def _get_key(self, key: str) -> str:
+        """Get full Redis key for throttling entry.
+
+        Args:
+            key: The throttling key.
+
+        Returns:
+            Full Redis key with prefix.
+        """
+        return f"{self.key_prefix}entry:{key}"
+
+    def _create_pool(self) -> "redis.ConnectionPool":
+        """Create a connection pool for sync client.
+
+        Returns:
+            Configured connection pool.
+        """
+        return redis.ConnectionPool.from_url(
+            self.redis_url,
+            max_connections=self.max_connections,
+            socket_timeout=self.socket_timeout,
+            socket_connect_timeout=self.socket_connect_timeout,
+            retry_on_timeout=True,
+            decode_responses=True,
+        )
+
+    async def _create_async_pool(self) -> "redis.asyncio.ConnectionPool":
+        """Create a connection pool for async client.
+
+        Returns:
+            Configured async connection pool.
+        """
+        return redis.asyncio.ConnectionPool.from_url(
+            self.redis_url,
+            max_connections=self.max_connections,
+            socket_timeout=self.socket_timeout,
+            socket_connect_timeout=self.socket_connect_timeout,
+            retry_on_timeout=True,
+            decode_responses=True,
+        )
+
+    def _get_fallback_store(self) -> InMemoryThrottlingStore:
+        """Get or create fallback in-memory store.
+
+        Returns:
+            InMemoryThrottlingStore instance.
+        """
+        if self._fallback_store is None:
+            self._fallback_store = InMemoryThrottlingStore()
+        return self._fallback_store
+
+    def _calculate_backoff_delay(self) -> float:
+        """Calculate exponential backoff delay.
+
+        Returns:
+            Delay in seconds.
+        """
+        import random
+
+        # Exponential backoff with jitter
+        delay = self.retry_base_delay * (2 ** self._retry_count)
+        # Add jitter (up to 25% of delay)
+        jitter = delay * random.uniform(0, 0.25)
+        return min(delay + jitter, 60.0)  # Cap at 60 seconds
+
+    def _handle_redis_error(self, error: Exception, operation: str) -> None:
+        """Handle Redis errors with logging and metrics.
+
+        Args:
+            error: The exception that occurred.
+            operation: Name of the operation that failed.
+        """
+        self._redis_errors += 1
+        self._last_error = error
+        self._last_error_time = time.time()
+        self._connected = False
+
+        logger.error(
+            f"Redis throttling store error during {operation}: {error}",
+            extra={
+                "operation": operation,
+                "error_type": type(error).__name__,
+                "retry_count": self._retry_count,
+            },
+        )
+
+    def _try_reconnect_sync(self) -> bool:
+        """Attempt to reconnect to Redis synchronously.
+
+        Returns:
+            True if reconnection successful, False otherwise.
+        """
+        if self._retry_count >= self.max_retries:
+            logger.warning(
+                f"Max retries ({self.max_retries}) reached, using fallback"
+            )
+            return False
+
+        delay = self._calculate_backoff_delay()
+        logger.info(
+            f"Attempting Redis reconnection in {delay:.2f}s "
+            f"(attempt {self._retry_count + 1}/{self.max_retries})"
+        )
+
+        time.sleep(delay)
+        self._retry_count += 1
+
+        try:
+            # Close existing connections
+            if self._client:
+                try:
+                    self._client.close()
+                except Exception:
+                    pass
+                self._client = None
+
+            if self._pool:
+                try:
+                    self._pool.disconnect()
+                except Exception:
+                    pass
+                self._pool = None
+
+            # Create new connection
+            self._pool = self._create_pool()
+            self._client = redis.Redis(connection_pool=self._pool)
+
+            # Test connection
+            if self._client.ping():
+                self._connected = True
+                self._retry_count = 0
+                self._using_fallback = False
+                self._reconnections += 1
+                logger.info("Redis throttling store reconnection successful")
+                return True
+        except Exception as e:
+            logger.warning(f"Reconnection attempt failed: {e}")
+
+        return False
+
+    @property
+    def client(self) -> "redis.Redis":
+        """Get sync Redis client with connection pooling.
+
+        Creates the connection pool and client on first access.
+        Handles reconnection on failure.
+
+        Returns:
+            Redis client instance.
+        """
+        if self._client is None or not self._connected:
+            with self._lock:
+                if self._client is None or not self._connected:
+                    try:
+                        self._pool = self._create_pool()
+                        self._client = redis.Redis(connection_pool=self._pool)
+                        # Test connection
+                        self._client.ping()
+                        self._connected = True
+                        self._retry_count = 0
+                        logger.debug("Redis throttling store sync client connected")
+                    except Exception as e:
+                        self._handle_redis_error(e, "client_init")
+                        raise
+        return self._client
+
+    async def get_async_client(self) -> "redis.asyncio.Redis":
+        """Get async Redis client with connection pooling.
+
+        Creates the async connection pool and client on first access.
+
+        Returns:
+            Async Redis client instance.
+        """
+        if self._async_lock is None:
+            self._async_lock = asyncio.Lock()
+
+        if self._async_client is None or not self._connected:
+            async with self._async_lock:
+                if self._async_client is None or not self._connected:
+                    try:
+                        self._async_pool = await self._create_async_pool()
+                        self._async_client = redis.asyncio.Redis(
+                            connection_pool=self._async_pool
+                        )
+                        # Test connection
+                        await self._async_client.ping()
+                        self._connected = True
+                        self._retry_count = 0
+                        logger.debug("Redis throttling store async client connected")
+                    except Exception as e:
+                        self._handle_redis_error(e, "async_client_init")
+                        raise
+        return self._async_client
+
+    def _serialize_entry(self, entry: ThrottlingEntry) -> dict[str, str]:
+        """Serialize entry for Redis storage.
+
+        Args:
+            entry: The entry to serialize.
+
+        Returns:
+            Dictionary suitable for Redis HSET.
+        """
+        return {
+            self.FIELD_COUNT: str(entry.count),
+            self.FIELD_WINDOW_START: str(entry.window_start),
+            self.FIELD_TOKENS: str(entry.tokens),
+            self.FIELD_LAST_REFILL: str(entry.last_refill),
+            self.FIELD_LAST_ACCESSED: str(entry.last_accessed),
+            self.FIELD_METADATA: json.dumps(entry.metadata) if entry.metadata else "{}",
+        }
+
+    def _deserialize_entry(self, key: str, data: dict[str, str]) -> ThrottlingEntry:
+        """Deserialize entry from Redis storage.
+
+        Args:
+            key: The throttling key.
+            data: Dictionary from Redis HGETALL.
+
+        Returns:
+            ThrottlingEntry instance.
+        """
+        metadata = {}
+        if data.get(self.FIELD_METADATA):
+            try:
+                metadata = json.loads(data[self.FIELD_METADATA])
+            except json.JSONDecodeError:
+                pass
+
+        return ThrottlingEntry(
+            key=key,
+            count=int(data.get(self.FIELD_COUNT, 0)),
+            window_start=float(data.get(self.FIELD_WINDOW_START, 0)),
+            tokens=float(data.get(self.FIELD_TOKENS, 0)),
+            last_refill=float(data.get(self.FIELD_LAST_REFILL, 0)),
+            last_accessed=float(data.get(self.FIELD_LAST_ACCESSED, 0)),
+            metadata=metadata,
+        )
+
+    def get(self, key: str) -> ThrottlingEntry | None:
+        """Get entry by key.
+
+        Falls back to InMemory store on Redis failure if enabled.
+
+        Args:
+            key: The throttling key.
+
+        Returns:
+            Entry if found, None otherwise.
+        """
+        if self._using_fallback and self.enable_fallback:
+            return self._get_fallback_store().get(key)
+
+        try:
+            redis_key = self._get_key(key)
+            now = time.time()
+
+            # Get and update last_accessed atomically
+            pipe = self.client.pipeline()
+            pipe.hgetall(redis_key)
+            pipe.hset(redis_key, self.FIELD_LAST_ACCESSED, str(now))
+            pipe.expire(redis_key, self.default_ttl)
+            results = pipe.execute()
+
+            data = results[0]
+            if not data:
+                return None
+
+            return self._deserialize_entry(key, data)
+
+        except Exception as e:
+            self._handle_redis_error(e, "get")
+
+            if self.enable_fallback:
+                self._using_fallback = True
+                self._fallback_count += 1
+                logger.warning("Falling back to InMemory throttling store")
+                return self._get_fallback_store().get(key)
+
+            raise
+
+    async def get_async(self, key: str) -> ThrottlingEntry | None:
+        """Async get entry by key.
+
+        Args:
+            key: The throttling key.
+
+        Returns:
+            Entry if found, None otherwise.
+        """
+        if self._using_fallback and self.enable_fallback:
+            return self._get_fallback_store().get(key)
+
+        try:
+            client = await self.get_async_client()
+            redis_key = self._get_key(key)
+            now = time.time()
+
+            # Get and update last_accessed atomically
+            pipe = client.pipeline()
+            pipe.hgetall(redis_key)
+            pipe.hset(redis_key, self.FIELD_LAST_ACCESSED, str(now))
+            pipe.expire(redis_key, self.default_ttl)
+            results = await pipe.execute()
+
+            data = results[0]
+            if not data:
+                return None
+
+            return self._deserialize_entry(key, data)
+
+        except Exception as e:
+            self._handle_redis_error(e, "get_async")
+
+            if self.enable_fallback:
+                self._using_fallback = True
+                self._fallback_count += 1
+                logger.warning("Falling back to InMemory throttling store")
+                return self._get_fallback_store().get(key)
+
+            raise
+
+    def set(self, entry: ThrottlingEntry) -> None:
+        """Set or update an entry.
+
+        Args:
+            entry: The entry to store.
+        """
+        if self._using_fallback and self.enable_fallback:
+            self._get_fallback_store().set(entry)
+            return
+
+        try:
+            redis_key = self._get_key(entry.key)
+            entry.last_accessed = time.time()
+
+            pipe = self.client.pipeline()
+            pipe.hset(redis_key, mapping=self._serialize_entry(entry))
+            pipe.expire(redis_key, self.default_ttl)
+            pipe.sadd(self._index_key, entry.key)
+            pipe.expire(self._index_key, self.default_ttl * 2)
+            pipe.execute()
+
+        except Exception as e:
+            self._handle_redis_error(e, "set")
+
+            if self.enable_fallback:
+                self._using_fallback = True
+                self._fallback_count += 1
+                logger.warning("Falling back to InMemory throttling store")
+                self._get_fallback_store().set(entry)
+                return
+
+            raise
+
+    async def set_async(self, entry: ThrottlingEntry) -> None:
+        """Async set or update an entry.
+
+        Args:
+            entry: The entry to store.
+        """
+        if self._using_fallback and self.enable_fallback:
+            self._get_fallback_store().set(entry)
+            return
+
+        try:
+            client = await self.get_async_client()
+            redis_key = self._get_key(entry.key)
+            entry.last_accessed = time.time()
+
+            pipe = client.pipeline()
+            pipe.hset(redis_key, mapping=self._serialize_entry(entry))
+            pipe.expire(redis_key, self.default_ttl)
+            pipe.sadd(self._index_key, entry.key)
+            pipe.expire(self._index_key, self.default_ttl * 2)
+            await pipe.execute()
+
+        except Exception as e:
+            self._handle_redis_error(e, "set_async")
+
+            if self.enable_fallback:
+                self._using_fallback = True
+                self._fallback_count += 1
+                logger.warning("Falling back to InMemory throttling store")
+                self._get_fallback_store().set(entry)
+                return
+
+            raise
+
+    def increment(self, key: str, window_start: float) -> int:
+        """Increment counter and return new count.
+
+        Uses Lua script for atomic check-and-increment.
+
+        Args:
+            key: The throttling key.
+            window_start: Start of current window.
+
+        Returns:
+            New count value.
+        """
+        if self._using_fallback and self.enable_fallback:
+            return self._get_fallback_store().increment(key, window_start)
+
+        # Lua script for atomic increment with window check
+        lua_script = """
+        local key = KEYS[1]
+        local index_key = KEYS[2]
+        local window_start = tonumber(ARGV[1])
+        local now = tonumber(ARGV[2])
+        local ttl = tonumber(ARGV[3])
+        local raw_key = ARGV[4]
+
+        -- Get current window_start
+        local current_window = redis.call('HGET', key, 'window_start')
+
+        if current_window == false or tonumber(current_window) ~= window_start then
+            -- New window, reset count to 1
+            redis.call('HMSET', key,
+                'count', 1,
+                'window_start', window_start,
+                'tokens', 0,
+                'last_refill', 0,
+                'last_accessed', now,
+                'metadata', '{}'
+            )
+            redis.call('EXPIRE', key, ttl)
+            redis.call('SADD', index_key, raw_key)
+            redis.call('EXPIRE', index_key, ttl * 2)
+            return 1
+        else
+            -- Same window, increment
+            local new_count = redis.call('HINCRBY', key, 'count', 1)
+            redis.call('HSET', key, 'last_accessed', now)
+            redis.call('EXPIRE', key, ttl)
+            return new_count
+        end
+        """
+
+        try:
+            redis_key = self._get_key(key)
+            now = time.time()
+
+            result = self.client.eval(
+                lua_script,
+                2,  # number of keys
+                redis_key,
+                self._index_key,
+                str(window_start),
+                str(now),
+                str(self.default_ttl),
+                key,
+            )
+
+            return int(result)
+
+        except Exception as e:
+            self._handle_redis_error(e, "increment")
+
+            if self.enable_fallback:
+                self._using_fallback = True
+                self._fallback_count += 1
+                logger.warning("Falling back to InMemory throttling store")
+                return self._get_fallback_store().increment(key, window_start)
+
+            raise
+
+    async def increment_async(self, key: str, window_start: float) -> int:
+        """Async increment counter and return new count.
+
+        Args:
+            key: The throttling key.
+            window_start: Start of current window.
+
+        Returns:
+            New count value.
+        """
+        if self._using_fallback and self.enable_fallback:
+            return self._get_fallback_store().increment(key, window_start)
+
+        # Lua script for atomic increment
+        lua_script = """
+        local key = KEYS[1]
+        local index_key = KEYS[2]
+        local window_start = tonumber(ARGV[1])
+        local now = tonumber(ARGV[2])
+        local ttl = tonumber(ARGV[3])
+        local raw_key = ARGV[4]
+
+        local current_window = redis.call('HGET', key, 'window_start')
+
+        if current_window == false or tonumber(current_window) ~= window_start then
+            redis.call('HMSET', key,
+                'count', 1,
+                'window_start', window_start,
+                'tokens', 0,
+                'last_refill', 0,
+                'last_accessed', now,
+                'metadata', '{}'
+            )
+            redis.call('EXPIRE', key, ttl)
+            redis.call('SADD', index_key, raw_key)
+            redis.call('EXPIRE', index_key, ttl * 2)
+            return 1
+        else
+            local new_count = redis.call('HINCRBY', key, 'count', 1)
+            redis.call('HSET', key, 'last_accessed', now)
+            redis.call('EXPIRE', key, ttl)
+            return new_count
+        end
+        """
+
+        try:
+            client = await self.get_async_client()
+            redis_key = self._get_key(key)
+            now = time.time()
+
+            result = await client.eval(
+                lua_script,
+                2,
+                redis_key,
+                self._index_key,
+                str(window_start),
+                str(now),
+                str(self.default_ttl),
+                key,
+            )
+
+            return int(result)
+
+        except Exception as e:
+            self._handle_redis_error(e, "increment_async")
+
+            if self.enable_fallback:
+                self._using_fallback = True
+                self._fallback_count += 1
+                logger.warning("Falling back to InMemory throttling store")
+                return self._get_fallback_store().increment(key, window_start)
+
+            raise
+
+    def cleanup(self, max_age_seconds: int) -> int:
+        """Remove old entries.
+
+        Redis handles TTL automatically, but this method can force
+        cleanup of entries older than max_age_seconds.
+
+        Args:
+            max_age_seconds: Maximum age of entries to keep.
+
+        Returns:
+            Number of entries removed.
+        """
+        if self._using_fallback and self.enable_fallback:
+            return self._get_fallback_store().cleanup(max_age_seconds)
+
+        try:
+            cutoff = time.time() - max_age_seconds
+            removed = 0
+
+            # Get all keys from index
+            keys = self.client.smembers(self._index_key)
+
+            for key in keys:
+                redis_key = self._get_key(key)
+                data = self.client.hgetall(redis_key)
+
+                if not data:
+                    # Entry expired, remove from index
+                    self.client.srem(self._index_key, key)
+                    removed += 1
+                elif float(data.get(self.FIELD_WINDOW_START, 0)) < cutoff:
+                    # Entry is old, delete it
+                    self.client.delete(redis_key)
+                    self.client.srem(self._index_key, key)
+                    removed += 1
+
+            return removed
+
+        except Exception as e:
+            self._handle_redis_error(e, "cleanup")
+
+            if self.enable_fallback:
+                self._using_fallback = True
+                return self._get_fallback_store().cleanup(max_age_seconds)
+
+            raise
+
+    async def cleanup_async(self, max_age_seconds: int) -> int:
+        """Async remove old entries.
+
+        Args:
+            max_age_seconds: Maximum age of entries to keep.
+
+        Returns:
+            Number of entries removed.
+        """
+        if self._using_fallback and self.enable_fallback:
+            return self._get_fallback_store().cleanup(max_age_seconds)
+
+        try:
+            client = await self.get_async_client()
+            cutoff = time.time() - max_age_seconds
+            removed = 0
+
+            # Get all keys from index
+            keys = await client.smembers(self._index_key)
+
+            for key in keys:
+                redis_key = self._get_key(key)
+                data = await client.hgetall(redis_key)
+
+                if not data:
+                    await client.srem(self._index_key, key)
+                    removed += 1
+                elif float(data.get(self.FIELD_WINDOW_START, 0)) < cutoff:
+                    await client.delete(redis_key)
+                    await client.srem(self._index_key, key)
+                    removed += 1
+
+            return removed
+
+        except Exception as e:
+            self._handle_redis_error(e, "cleanup_async")
+
+            if self.enable_fallback:
+                self._using_fallback = True
+                return self._get_fallback_store().cleanup(max_age_seconds)
+
+            raise
+
+    def clear(self) -> None:
+        """Clear all throttling entries."""
+        if self._using_fallback and self.enable_fallback:
+            self._get_fallback_store().clear()
+            return
+
+        try:
+            # Get all keys from index
+            keys = self.client.smembers(self._index_key)
+
+            if keys:
+                # Delete all entry keys
+                redis_keys = [self._get_key(k) for k in keys]
+                self.client.delete(*redis_keys)
+
+            # Delete index
+            self.client.delete(self._index_key)
+
+        except Exception as e:
+            self._handle_redis_error(e, "clear")
+
+            if self.enable_fallback:
+                self._using_fallback = True
+                self._get_fallback_store().clear()
+                return
+
+            raise
+
+    async def clear_async(self) -> None:
+        """Async clear all throttling entries."""
+        if self._using_fallback and self.enable_fallback:
+            self._get_fallback_store().clear()
+            return
+
+        try:
+            client = await self.get_async_client()
+
+            # Get all keys from index
+            keys = await client.smembers(self._index_key)
+
+            if keys:
+                redis_keys = [self._get_key(k) for k in keys]
+                await client.delete(*redis_keys)
+
+            await client.delete(self._index_key)
+
+        except Exception as e:
+            self._handle_redis_error(e, "clear_async")
+
+            if self.enable_fallback:
+                self._using_fallback = True
+                self._get_fallback_store().clear()
+                return
+
+            raise
+
+    def count(self) -> int:
+        """Get total entry count."""
+        if self._using_fallback and self.enable_fallback:
+            return self._get_fallback_store().count()
+
+        try:
+            return self.client.scard(self._index_key)
+
+        except Exception as e:
+            self._handle_redis_error(e, "count")
+
+            if self.enable_fallback:
+                self._using_fallback = True
+                return self._get_fallback_store().count()
+
+            raise
+
+    async def count_async(self) -> int:
+        """Async get total entry count."""
+        if self._using_fallback and self.enable_fallback:
+            return self._get_fallback_store().count()
+
+        try:
+            client = await self.get_async_client()
+            return await client.scard(self._index_key)
+
+        except Exception as e:
+            self._handle_redis_error(e, "count_async")
+
+            if self.enable_fallback:
+                self._using_fallback = True
+                return self._get_fallback_store().count()
+
+            raise
+
+    def health_check(self) -> dict[str, Any]:
+        """Perform health check and return status.
+
+        Returns:
+            Dictionary with health status information.
+        """
+        import re
+
+        result = {
+            "healthy": False,
+            "connected": self._connected,
+            "using_fallback": self._using_fallback,
+            "redis_url": re.sub(r"://[^:]+:[^@]+@", "://***:***@", self.redis_url),
+        }
+
+        if self._using_fallback and self.enable_fallback:
+            result["healthy"] = True
+            result["mode"] = "fallback"
+            result["fallback_entries"] = self._get_fallback_store().count()
+            return result
+
+        try:
+            ping_ok = self.client.ping()
+
+            if ping_ok:
+                result["healthy"] = True
+                result["mode"] = "redis"
+                result["entries"] = self.count()
+
+                # Get Redis info
+                info = self.client.info(section="server")
+                result["redis_info"] = {
+                    "version": info.get("redis_version"),
+                    "uptime_seconds": info.get("uptime_in_seconds"),
+                }
+
+        except Exception as e:
+            result["error"] = str(e)
+            if self._last_error_time:
+                from datetime import datetime
+                result["last_error_time"] = datetime.fromtimestamp(
+                    self._last_error_time
+                ).isoformat()
+
+        return result
+
+    async def health_check_async(self) -> dict[str, Any]:
+        """Async perform health check and return status.
+
+        Returns:
+            Dictionary with health status information.
+        """
+        import re
+
+        result = {
+            "healthy": False,
+            "connected": self._connected,
+            "using_fallback": self._using_fallback,
+            "redis_url": re.sub(r"://[^:]+:[^@]+@", "://***:***@", self.redis_url),
+        }
+
+        if self._using_fallback and self.enable_fallback:
+            result["healthy"] = True
+            result["mode"] = "fallback"
+            result["fallback_entries"] = self._get_fallback_store().count()
+            return result
+
+        try:
+            client = await self.get_async_client()
+            ping_ok = await client.ping()
+
+            if ping_ok:
+                result["healthy"] = True
+                result["mode"] = "redis"
+                result["entries"] = await self.count_async()
+
+                info = await client.info(section="server")
+                result["redis_info"] = {
+                    "version": info.get("redis_version"),
+                    "uptime_seconds": info.get("uptime_in_seconds"),
+                }
+
+        except Exception as e:
+            result["error"] = str(e)
+            if self._last_error_time:
+                from datetime import datetime
+                result["last_error_time"] = datetime.fromtimestamp(
+                    self._last_error_time
+                ).isoformat()
+
+        return result
+
+    def get_metrics(self) -> dict[str, Any]:
+        """Get store metrics.
+
+        Returns:
+            Dictionary with metrics data.
+        """
+        metrics = {
+            "redis_errors": self._redis_errors,
+            "fallback_count": self._fallback_count,
+            "reconnections": self._reconnections,
+            "connected": self._connected,
+            "using_fallback": self._using_fallback,
+        }
+
+        if self._using_fallback and self._fallback_store:
+            metrics["fallback_metrics"] = self._fallback_store.get_metrics()
+        else:
+            try:
+                metrics["total_entries"] = self.count()
+            except Exception:
+                metrics["total_entries"] = -1
+
+        return metrics
+
+    def close(self) -> None:
+        """Close all connections and pools."""
+        if self._client is not None:
+            try:
+                self._client.close()
+            except Exception:
+                pass
+            self._client = None
+
+        if self._pool is not None:
+            try:
+                self._pool.disconnect()
+            except Exception:
+                pass
+            self._pool = None
+
+        self._connected = False
+
+    async def close_async(self) -> None:
+        """Async close all connections and pools."""
+        if self._async_client is not None:
+            try:
+                await self._async_client.close()
+            except Exception:
+                pass
+            self._async_client = None
+
+        if self._async_pool is not None:
+            try:
+                await self._async_pool.disconnect()
+            except Exception:
+                pass
+            self._async_pool = None
+
+        self._connected = False
+
+    def __enter__(self) -> "RedisThrottlingStore":
+        """Context manager entry."""
+        return self
+
+    def __exit__(self, exc_type: Any, exc_val: Any, exc_tb: Any) -> None:
+        """Context manager exit, closes connections."""
+        self.close()
+
+    async def __aenter__(self) -> "RedisThrottlingStore":
+        """Async context manager entry."""
+        return self
+
+    async def __aexit__(self, exc_type: Any, exc_val: Any, exc_tb: Any) -> None:
+        """Async context manager exit, closes connections."""
+        await self.close_async()
+
+
+# =============================================================================
+# Factory Function
+# =============================================================================
+
+
+class ThrottlingStoreType:
+    """Store type constants."""
+
+    MEMORY = "memory"
+    SQLITE = "sqlite"
+    REDIS = "redis"
+
+
+def create_throttling_store(
+    store_type: str | None = None,
+    **kwargs: Any,
+) -> BaseThrottlingStore:
+    """Factory function to create appropriate throttling store.
+
+    Selects the store type based on configuration or environment variables.
+
+    Environment variables:
+        TRUTHOUND_THROTTLE_STORE_TYPE: Store type (memory, sqlite, redis)
+        TRUTHOUND_THROTTLE_SQLITE_PATH: SQLite database path
+        TRUTHOUND_THROTTLE_REDIS_URL: Redis connection URL (enables redis)
+
+    Args:
+        store_type: Explicit store type override. If None, auto-detects.
+        **kwargs: Additional arguments passed to the store constructor.
+
+    Returns:
+        Configured BaseThrottlingStore instance.
+
+    Example:
+        # Auto-detect based on environment
+        store = create_throttling_store()
+
+        # Explicit type
+        store = create_throttling_store("redis", default_ttl=7200)
+
+        # SQLite with custom path
+        store = create_throttling_store("sqlite", db_path="/tmp/throttle.db")
+    """
+    # Determine store type
+    if store_type is None:
+        store_type = os.getenv("TRUTHOUND_THROTTLE_STORE_TYPE")
+
+    # Auto-detect if still None
+    if store_type is None:
+        redis_url = os.getenv("TRUTHOUND_THROTTLE_REDIS_URL")
+        if redis_url and REDIS_AVAILABLE:
+            store_type = ThrottlingStoreType.REDIS
+            logger.info(
+                "Auto-detected Redis throttling store from TRUTHOUND_THROTTLE_REDIS_URL"
+            )
+        elif os.getenv("TRUTHOUND_THROTTLE_SQLITE_PATH"):
+            store_type = ThrottlingStoreType.SQLITE
+            logger.info("Auto-detected SQLite store from TRUTHOUND_THROTTLE_SQLITE_PATH")
+        else:
+            store_type = ThrottlingStoreType.MEMORY
+            logger.info("Using default InMemory throttling store")
+
+    # Normalize store type
+    store_type = store_type.lower().strip()
+
+    # Create store based on type
+    if store_type == ThrottlingStoreType.MEMORY:
+        logger.info("Creating InMemory throttling store")
+        return InMemoryThrottlingStore(**kwargs)
+
+    elif store_type == ThrottlingStoreType.SQLITE:
+        db_path = kwargs.pop("db_path", None) or os.getenv(
+            "TRUTHOUND_THROTTLE_SQLITE_PATH", "throttling.db"
+        )
+        logger.info(f"Creating SQLite throttling store at {db_path}")
+        return SQLiteThrottlingStore(db_path=db_path)
+
+    elif store_type == ThrottlingStoreType.REDIS:
+        if not REDIS_AVAILABLE:
+            logger.warning(
+                "Redis not available, falling back to InMemory store. "
+                "Install with: pip install truthound-dashboard[redis]"
+            )
+            return InMemoryThrottlingStore(**kwargs)
+
+        logger.info("Creating Redis throttling store")
+        return RedisThrottlingStore(**kwargs)
+
+    else:
+        logger.warning(
+            f"Unknown store type '{store_type}', falling back to InMemory store"
+        )
+        return InMemoryThrottlingStore(**kwargs)