python.redis.unbounded_keys

Stability High

Detects Redis key patterns that can grow unboundedly.

Why It Matters

Unbounded Redis keys:

• Memory exhaustion — Redis runs out of memory
• Performance degradation — Large datasets slow operations
• Service crash — OOM kills Redis process

Example

# ❌ Before (unbounded keys)
def track_user_activity(user_id: str, action: str):
    key = f"user:{user_id}:actions"
    redis.lpush(key, action)  # Grows forever!

# ✅ After (bounded with max length)
MAX_ACTIONS = 1000

def track_user_activity(user_id: str, action: str):
    key = f"user:{user_id}:actions"
    pipe = redis.pipeline()
    pipe.lpush(key, action)
    pipe.ltrim(key, 0, MAX_ACTIONS - 1)  # Keep last 1000
    pipe.execute()

What Unfault Detects

• LPUSH/RPUSH without LTRIM
• SET accumulating without expiry
• Unbounded ZADD operations

Related Rules

• python.redis.missing_ttl