noCV
BRELEASE-106 · Exercise critical paths

Exercise partial provider failure without duplicating orders

Practice briefTaskAdvanced

The delivery provider accepts a booking but the checkout request times out.

Focused work estimate
4h + prerequisites
Priority in the scenario
High
Engineering practice
Failure testing · Idempotency

Estimated field mix

  • Quality engineering50%
  • Distributed systems30%
  • Integrations20%

Field percentages are editorial estimates of the ticket's engineering focus. They total 100%; they are not measured time, proficiency scores, or ownership evidence.

Your next step

Review it, then add it to your workspace.

The board opens an editable draft; nothing is saved until you confirm it. Sign-in and workspace permissions apply, and Demo boards remain ephemeral.

Project context

A fictional retailer is changing delivery options. Its regression suite is large, but no one can explain whether inventory, totals, or recovery behavior is covered.

Setup prerequisites

  • Build a minimal local checkout fixture with synthetic prices, inventory, and a mock delivery provider.

Preceding work

Complete these dependencies, or supply their agreed outputs before taking this ticket.

Acceptance criteria

  • Represent booking outcome as pending when confirmation is unknown.
  • Retry using the same operation identity.
  • Prevent a second order from the same confirmed submission.

Implementation constraints

  • Use a scripted provider double; send no external bookings.

Verification to include

  • Recover a timed-out accepted booking.
  • Replay duplicate callbacks and verify one final order.

Deliverables

  • Partial-failure acceptance suite.

Rollout and recovery

Release behind a limited local cohort switch; stop new bookings if reconciliation is unhealthy.

Value of the work

For the engineer: Practice risk analysis, exploratory testing, and evidence-based release decisions.

For the team: Provide a reviewable release assessment tied to customer-impacting invariants.

Evidence boundaries

Outcome Evidence: Tests, patches, and runbooks are requested deliverables. They become Outcome Evidence only through a qualified Mission and immutable Evidence IDs.

Ownership Evidence: Independent adaptation must be observed under a declared verification policy and cite immutable Evidence IDs. Completing a planning ticket establishes no Ownership Evidence.