Design a logging and monitoring solution for a mission-critical data pipeline.

Section 1 — The Context (The 'Why')
Mission-critical pipelines require observability that survives pipeline failures. The primary challenge is correlation across components, alerting on business metrics, and retaining history without unbounded cost. A naive approach logs to the same storage as pipeline data—when the pipeline fails, logs may be lost.

Section 2 — The Diagram

[Pipeline Jobs] --> [CloudWatch] --> [Metrics]
[Logs] --> [OpenSearch] --> [Log Analytics]
[Trace IDs] --> [Distributed Tracing]

Section 3 — Component Logic
Each stage emits structured logs with trace_id. Logs to OpenSearch; metrics to CloudWatch. Distributed tracing connects requests. Alerting on SLOs: consumer lag, job duration. Sampling reduces volume when ingest is slow. TTL policies control retention cost. In production, monitor consumer lag, checkpoint success rate, and sink write latency as primary SLOs. Partitioning strategies should align with query patterns; bucketing within partitions mitigates join skew. TTL policies on raw and intermediate data control storage cost while preserving replay capability for debugging and backfill. Data skew mitigation via salting or secondary hashing prevents single partitions from becoming bottlenecks. Exactly-once semantics require transactional commits at the sink; at-least-once delivery demands idempotent write logic to avoid duplicates. Fan-out patterns allow one source topic to feed multiple downstream consumers without re-ingestion. Backpressure handling ensures that slow processors do not cause unbounded buffer growth; Kafka consumer lag is a key metric. Schema evolution should follow additive-only rules where possible to avoid breaking consumer compatibility. The CAP trade-off should be documented per component: analytics typically favors AP, while financial reconciliation requires CP. Blast radius from component failure is bounded by replication and checkpointing; design for graceful degradation during partial outages. Cost optimization: use Spot instances for batch workloads and tier cold data to lower storage classes. Dead-letter queues preserve failed records for replay rather than dropping them.

Section 4 — The Trade-offs (The 'Senior' part)