Pick a focused question that fits your time, stack, and interview goal.
How much time do you have?
Show one-drill sessions you can finish now.
22 results across 1 active filter
Page 1 of 1
Explains partition key selection for ordering, load distribution, hot keys, and consumer scalability.
Compares point-to-point queues with publish-subscribe topics and explains when each model fits.
Explains useful message metadata for tracing, idempotency, ownership, versioning, and support diagnostics.
Explains why message handlers must tolerate duplicate delivery and how idempotency protects side effects.
Explains how systems behave when the broker is down or slow and how backpressure protects callers and dependencies.
Explains retry strategy for transient failures and poison-message handling for permanent failures.
Explains how to reprocess event streams safely for new projections, bug fixes, and data repair.
Explains command messages versus event messages and how ownership changes their design.
Explains how consumer groups share event-stream work and how offset commits affect duplicate or lost processing.
Explains why exactly-once claims do not remove the need for idempotent side effects and consistency boundaries.
Keeps long-running queue work safe when message ownership expires, renewal fails, settlement is uncertain, or another worker receives the same job.
Explains when to store large payloads outside the broker and pass references through messages.
Explains operational signals for queue-backed systems and how to respond when consumers fall behind.
Explains when backend work belongs in a queue instead of slow or unreliable inline request handling.
Preserves order only within the business key that requires it, while handling hot keys, gaps, stale events, replay, and poison messages explicitly.
Repairs a workflow when a remote operation may have succeeded but local state, messages, and callbacks disagree.
Explains how to inspect, fix, replay, skip, or archive dead-letter messages without repeating bad side effects.
Explains how multiple worker instances share queue work and what limits safe scaling.
Chooses event retention and compaction from consumer outage, replay, rebuild, audit, and disaster-recovery requirements rather than storage cost alone.
Explains delayed delivery, scheduled work, reminders, retry timing, cancellation, and clock-related trade-offs.
Makes consumer-side deduplication atomic with the business change while supporting concurrent delivery, retention, replay, and failed processing.
Explains event schema evolution, backward compatibility, additive changes, consumer rollout, and contract ownership.