Redis OOM blocks BullMQ job processing and ingestion

Redis runs out of memory when BullMQ queue fills up and workers cannot process events fast enough. Results in 'OOM command not allowed when used memory > maxmemory' errors on job operations like trace-delete, preventing new data ingestion.

Configure Redis with no-eviction cache policy (recommended). Scale worker containers to increase processing throughput and prevent queue buildup. Monitor queue lengths using statsd, AWS ECS metrics, or BullMQ manually to identify problematic queues. When OOM occurs, web container rejects new tracing events with 5xx codes; worker continues processing pending events until Redis accepts new operations.