Lesson 01
Event-Driven Architecture Decisions Explained
Decision: choose asynchronous boundaries from durable facts, ownership, delay, duplicates, ordering, and recovery.
Event-Driven Systems topic cluster
Event-Driven Systems
A production learning path for event architecture, Kafka, contracts, reliable publication, duplicate-safe consumption, ordering, recovery, and operations.
17 ordered lessons ยท Interactive failure lab
Trace each event boundary from design to recovery.
Move from architecture and contracts through publication, consumption, convergence, coordination, production operations, and deterministic crash experiments.
Lesson 02
Message Queues Explained
Decision: separate broker delivery from application processing and business completion.
Lesson 03
Kafka Topics, Partitions, Consumer Groups, and Ordering
Decision: align partition keys, group parallelism, offsets, and ordering with business invariants.
Lesson 04
Message Delivery Semantics Explained
Decision: place acknowledgment after the durable effect while planning for redelivery.
Lesson 05
Event Design with CloudEvents and AsyncAPI
Decision: separate event envelopes, payload schemas, channels, operations, and transport bindings.
Lesson 06
Event Schema Evolution and Compatibility Explained
Decision: test structural and semantic compatibility across retained events and supported consumers.
Lesson 07
Transactional Outbox Pattern Explained
Decision: commit business state and publication intent atomically inside one local database.
Lesson 08
Reliable Event Publishing with Spring Boot and an Outbox
Decision: operate recoverable claims, broker acceptance, stable identity, and reconciliation.
Lesson 09
Idempotent Consumer Pattern Explained
Decision: commit duplicate detection with the business update under a uniqueness invariant.
Lesson 10
Reliable Spring Kafka Consumers: ACKs, Retries, and DLTs
Decision: match ACK, retry, DLT, and transaction settings to the actual durable effect.
Lesson 11
Dead Letter Queue Explained
Decision: stop bounded automatic work and make remediation and replay an operated state.
Lesson 12
Event Ordering, Deduplication, and State Convergence
Decision: combine stable event identity with entity versions and atomic conditional updates.
Lesson 13
Saga Pattern and Compensating Transactions Explained
Decision: persist long-running transitions, compensation, and manual intervention.
Lesson 14
Background Jobs Explained
Decision: bound worker admission, retry, expiry, duplicate safety, and operational ownership.
Lesson 15
Webhooks Explained
Decision: accept signed delivery, respond deliberately, and process downstream work safely.
Lesson 16
Production Kafka and Event-Driven Systems Troubleshooting
Decision: trace user impact through producer, broker, consumer, retry, and business evidence.
Lesson 17
Message Delivery Semantics Lab: Crash, Retry, and Idempotency Examples
Interactive failure lab: predict loss, replay, duplicate effects, idempotent convergence, and retry exhaustion from a deterministic timeline.