Slow Query Backlog Masks Redis Connection Pool Exhaustion
warningRedis slowlog entries accumulating (redis.slowlog.length rising) can indicate operations blocking on network or disk I/O, exhausting connection pools and causing cascading failures in dependent services even when Redis CPU appears healthy.
Monitor redis.slowlog.length trending upward and redis.slowlog.micros.95percentile increasing while redis.clients.blocked grows. This pattern indicates operations taking longer than expected, holding connections and blocking application threads. May coincide with redis.persistence.rdb_last_bgsave_time_sec spikes during snapshot operations.
Query redis slowlog to identify problematic commands. Evaluate whether KEYS, SMEMBERS, or other O(N) operations are running on large datasets. Consider moving expensive operations to dedicated Redis instances. If correlated with persistence events, tune save intervals or switch to AOF. Increase connection pool timeouts in applications to handle persistence latency spikes.