The visible problem may look like one API or tuning value, but reliability is decided by state ownership, resource bounds, and recovery after failure. A single mocked page misses permission, lifecycle, and real WebRTC timing. Treat A and B as separate identities and storage with one scenarioId and two state traces.
Testing combines state models, fault injection, and real browser pairs. Deterministic cases protect known contracts, randomized timing finds races, and every failing seed plus endpoint trace becomes a permanent regression.
Make the implementation decisions explicit
Start from facts the data and protocol can guarantee, then decide what the interface may promise. Each rule below needs an owner, a bound, and a compatibility policy rather than an oral convention from one review.
- Each scenario declares identities, actions, assertions for both peers, cleanup, and timeout; isolate IndexedDB and permissions, using separate processes when needed.
- Separate protocol facts, user intent, and automatic recovery; automation may restore facts but never overturn an explicit choice.
- Retries need an idempotency key, backoff, and deadline; after the deadline create a new task instead of reviving old callbacks.
The delivery standard for A Two-Browser E2E Harness for Complete A-to-B Workflows is a usable normal path, convergent failures, bounded resources, and a state users can understand. The result is a production capability that can be explained, degraded safely, and rolled back—not a demo that works once.
Failure paths that are easy to miss
Prioritize faults that silently preserve false facts: the interface looks recovered while a queue, permission, or counter has diverged. The defect often appears only on the next action.
- Asserting only sent on A misses two notifications on B, while one shared context leaks storage and hides identity or refresh bugs.
- Refresh and network change start two recovery paths, and duplicate side effects look like two genuine user actions.
- Without backpressure or quota, a slow consumer raises memory, queue depth, and tail latency until unrelated users are affected.
How to verify it before release
Write the expected state trace before injecting faults. At every phase, reconcile user-visible outcome, both protocol endpoints, persistent records, and resource counts to prove the loop.
- For every feature run start, normal completion, one refresh, both refresh, cancel, network recovery, and close; endpoint state and resources must converge.
- Disconnect, change networks, and recover mid-operation; reconcile endpoint state, persistence, and resource counts.
- Allowlist log and analytics fields, proving payloads, secrets, full IP addresses, and identifying data never leave the device.
Completion is not one passing path. Every terminal state reconciles, automation stays below user intent, and every operational cost has an explicit ceiling.