▸case-01 We are building an e-commerce backend where placing an order involves charging a card, reserving inventory, and creating a shipment. If any step fails, previous actions must be rolled back. Can you provide an architectural breakdown and a code outline for a centralized coordinator that manages this sequence and handles rollbacks? | fail→fail | 29,775 | 23,466 | -21% | 1 | 1 | 0% | 4,805 | 4,333 | -10% | 0 | 0 | — |
▸case-02 Our enterprise platform processes multi-step vendor onboardings that require approvals across finance, legal, and HR. Some approvals take days and might time out or be rejected halfway through. Please give me a detailed workflow plan specifying how to manage long-running state transitions, timeout mechanisms, and cleanup actions upon failure. | fail→fail | 29,938 | 28,406 | -5% | 1 | 1 | 0% | 4,938 | 5,041 | +2% | 0 | 0 | — |
▸case-03 I'm designing a travel reservation system where booking a trip requires reserving flights, hotels, and rental cars across distinct microservices. I need a comprehensive blueprint detailing how to track execution states across services and implement compensating logic when downstream reservations fail. | fail→fail | 34,627 | 24,571 | -29% | 1 | 1 | 0% | 6,109 | 4,501 | -26% | 0 | 0 | — |
▸case-04 We are writing a SQL transaction for a single PostgreSQL database to transfer funds between two user balance records. Write the PL/pgSQL function using explicit `BEGIN`, `COMMIT`, and `ROLLBACK` blocks. | pass→fail | 13,729 | 11,721 | -15% | 1 | 1 | 0% | 2,759 | 2,401 | -13% | 0 | 0 | — |
▸case-05 We need a frontend React component using Axios to fetch user profile details from a GET /api/v1/user/profile REST endpoint and display a loading state. Provide the React component implementation. | pass→pass | 15,380 | 6,858 | -55% | 1 | 1 | 0% | 2,623 | 1,731 | -34% | 0 | 0 | — |
▸case-06 We have daily log files stored in Amazon S3 and need a PySpark batch job to aggregate log counts by hour and output Parquet files. Provide the PySpark script. | pass→pass | 15,862 | 10,479 | -34% | 1 | 1 | 0% | 3,239 | 2,321 | -28% | 0 | 0 | — |
▸case-07 We are building an asynchronous event-driven payment processing pipeline across 4 microservices using Apache Kafka. Should we publish domain events where each microservice listens and reacts directly to other services' events, or should we use a central coordinator service? Provide an architectural analysis and recommend when to choose choreography versus orchestration. | fail→fail | 30,145 | 24,719 | -18% | 1 | 1 | 0% | 4,782 | 3,817 | -20% | 0 | 0 | — |
▸case-08 When executing compensating steps in a distributed order cancellation workflow, network retries might deliver the refund request multiple times to the payment gateway microservice. How should the compensating payment action be designed to safely handle duplicate failure events? | fail→fail | 16,050 | 15,453 | -4% | 1 | 1 | 0% | 2,710 | 3,022 | +12% | 0 | 0 | — |
▸case-09 We are designing an orchestration engine that tracks long-running multi-day business processes. If the orchestrator node crashes mid-execution, how should the execution state be managed so the workflow can safely resume? | fail→fail | 19,445 | 22,407 | +15% | 1 | 1 | 0% | 3,282 | 3,636 | +11% | 0 | 0 | — |
▸case-10 In an automated subscription upgrade pipeline, step 1 generates a PDF invoice, step 2 charges the customer credit card via Stripe, and step 3 sends an SMS notification via Twilio. Sending an SMS cannot be undone once delivered. How should the workflow sequence be structured around this non-compensable step? | pass→fail | 13,205 | 16,606 | +26% | 1 | 1 | 0% | 2,195 | 3,258 | +48% | 0 | 0 | — |
▸case-11 In a supply chain fulfillment workflow, an external logistics service takes up to 48 hours to confirm carrier assignment. If no confirmation arrives within 48 hours, what workflow state transition and recovery actions should trigger? | fail→pass | 12,818 | 17,137 | +34% | 1 | 1 | 0% | 2,070 | 3,471 | +68% | 0 | 0 | — |
▸case-12 In a microservice architecture, a service needs to update its local PostgreSQL database and publish an event to RabbitMQ. If the event publish fails after the database commit, the system becomes inconsistent. What pattern addresses this dual-write problem in choreography sagas? | pass→pass | 7,599 | 8,381 | +10% | 1 | 1 | 0% | 1,237 | 1,770 | +43% | 0 | 0 | — |
▸case-13 In an automated deployment pipeline across multiple cloud regions, when region 2 deployment fails due to a transient rate limit, we want to retry or fallback rather than immediately rolling back region 1. How does forward recovery differ from backward compensation in saga design? | pass→pass | 15,095 | 14,043 | -7% | 1 | 1 | 0% | 2,458 | 2,768 | +13% | 0 | 0 | — |
▸case-14 In an event-driven food delivery system, a 'PaymentCompleted' event might arrive at the Restaurant service before the 'OrderCreated' event due to network latency. How should microservices handle out-of-order event execution in choreography sagas? | fail→fail | 17,286 | 17,945 | +4% | 1 | 1 | 0% | 2,950 | 3,514 | +19% | 0 | 0 | — |
▸case-15 During a rollback phase of a distributed hotel and flight booking saga, the compensation API call to refund the hotel fails with HTTP 500. How should the orchestrator handle failure during compensation execution? | pass→pass | 13,238 | 18,148 | +37% | 1 | 1 | 0% | 2,313 | 3,128 | +35% | 0 | 0 | — |
▸case-16 Unlike local database transactions, sagas lack global isolation (the 'I' in ACID), allowing users to see intermediate saga states before completion. What strategy prevents dirty reads or business logic errors caused by lack of isolation? | fail→fail | 14,726 | 9,645 | -35% | 1 | 1 | 0% | 2,411 | 1,954 | -19% | 0 | 0 | — |
▸case-17 We need to design a SaaS subscription renewal saga that charges a recurring invoice, extends account entitlements, and issues an updated invoice PDF. If the charge fails, access should remain unchanged. Outline the saga steps, states, and compensation logic. | fail→fail | 18,415 | 20,573 | +12% | 1 | 1 | 0% | 3,451 | 4,147 | +20% | 0 | 0 | — |
▸case-18 What key metrics and trace headers should be captured across microservices to monitor health and track stuck executions in a saga-orchestrated system? | pass→pass | 19,086 | 13,337 | -30% | 1 | 1 | 0% | 3,468 | 2,820 | -19% | 0 | 0 | — |
▸case-19 Two concurrent saga workflows (Saga A and Saga B) are trying to reserve the same inventory item and credit balance in reverse order. How can we prevent deadlocks in distributed saga execution? | fail→fail | 15,888 | 14,854 | -7% | 1 | 1 | 0% | 2,571 | 2,961 | +15% | 0 | 0 | — |
▸case-20 When building an orchestrator service in Node.js, should the workflow logic be written as code using async/await state machines or defined in JSON/YAML configuration files? Evaluate both approaches. | fail→fail | 18,596 | 19,722 | +6% | 1 | 1 | 0% | 3,000 | 3,484 | +16% | 0 | 0 | — |
▸case-21 We need a testing strategy to verify that our inventory release compensating action is safe to run multiple times. What specific testing approach and assertions should be used? | pass→fail | 18,369 | 13,128 | -29% | 1 | 1 | 0% | 3,265 | 2,682 | -18% | 0 | 0 | — |
▸case-22 Can we store active saga state transitions in Redis using hash structures to track current status (PENDING, RUNNING, COMPENSATING, COMPLETED, FAILED)? Outline the state machine data model and transition verification. | fail→fail | 17,762 | 18,146 | +2% | 1 | 1 | 0% | 3,358 | 3,818 | +14% | 0 | 0 | — |
▸case-23 In a high-value wire transfer saga (> $100,000), an automated step executes compliance checks, followed by an asynchronous human manager approval step that takes up to 24 hours. How should the orchestrator manage the suspended state? | fail→fail | 16,408 | 15,710 | -4% | 1 | 1 | 0% | 2,824 | 3,277 | +16% | 0 | 0 | — |