Enable sequential replays by default - #3229
Conversation
Signed-off-by: Peter Wielander <peter.wielander@vercel.com>
🦋 Changeset detectedLatest commit: f39f9be The changes in this PR will be included in the next version bump. This PR includes changesets to release 20 packages
Not sure what this means? Click here to learn what changesets are. Click here if you're a maintainer who wants to add another changeset to this PR |
📊 Workflow Benchmarkscommit Backend:
📈 STSO distribution vs main (inline / queue-hop histograms)1020 steps (inline) Cumulative STSO time: main 440353ms → this run 412969ms (Δ -27384ms, -6%) 📜 Previous results (1)ef49836Thu, 30 Jul 2026 21:40:56 GMT · run logs
ℹ️ Metric definitions & methodologyThe collapsed STSO distribution section above buckets every step gap of the sequential-steps run (not a sampled window), split by whether the step ending the gap ran inline — in the same warm process as the step before it, so the gap is pure framework overhead — or after a queue-hop — the first step of a fresh process, which pays queue dispatch, client reinit and event-log replay. Bars overlay the two runs: Best/P75/P90/P99 deltas compare against the most recent benchmark run on Metrics — TTFS: time to first step body (in-deployment start() → first step body, deployment clocks) · STSO: step-to-step overhead (gap between consecutive step bodies) · WO: workflow overhead (whole-run time outside step bodies, in-deployment anchored) · SL: stream latency (in-deployment write → read propagation, readAt - writtenAt) · SO: stream overhead (end-to-end write+consume time beyond the modelled generation window) Scenarios — step: one trivial no-op step, no stream; no hooks, so the run stays in turbo mode (in-process fast path) · stream: one streaming step; no hooks, so the run stays in turbo mode (in-process fast path) · hook + stream: registers a hook before one step, which exits turbo mode (dispatch path) · 1020 steps: 1020 trivial sequential steps; STSO is measured between consecutive steps in the given step ranges, and WO is the whole-run overhead outside step bodies · stream latency: parallel reader/writer steps on a dedicated stream; SL is the in-deployment write->read propagation (readAt - writtenAt) · stream overhead (text): writer streams 300 variable-length text token deltas paced at 100/s for 3s (a haiku-size LLM's token throughput) while a parallel reader drains the whole stream; SO is the end-to-end write+consume time beyond the 3s generation window (overhead/backpressure) · stream overhead (structured): same workload as stream overhead (text), but each delta is an AI-SDK-style structured object ({ type: 'text-delta', id, text }) instead of a raw string, so the SO gap vs the text scenario is the added serialization cost 🔴 marks a percentile over its target (within target is left unmarked). Targets (p75/p90/p99, ms) — TTFS 200/300/600 · SL 50/60/125 · SO 250/500/1000 All metrics are measured from deployment-side timestamps only. Runs are triggered by an in-deployment route that stamps the anchor ( Cold starts are kept in the numbers on purpose — they are part of real bursty-workload latency. The workbench deployment cold-starts the |
🧪 E2E Test Results✅ All tests passed E2E Test SummarySummary
Details by Category✅ ▲ Vercel Production
✅ 💻 Local Development
✅ 📦 Local Production
✅ 🐘 Local Postgres
✅ 🪟 Windows
✅ 📋 Other
✅ vercel-multi-region
|
VaguelySerious
added
the
event-log-race-repro
Event Log Race Repro559 of 1400 latest repro runs hit event-log regressions. 94 runs hit harness-side Run History
Latest Scenario Breakdown
Latest Non-Completed Runs
Showing 20 of 653 non-completed runs. |
![vercel[bot]](https://cdn.statically.io/img/avatars.githubusercontent.com/in/8329?s=60&v=4){kind=small}
With sequential replays on, a run's continuations share one `maxConcurrency: 1` topic. An inline step body holds the invocation's queue message — and therefore the run's only slot — until it finishes, so a wait continuation, hook resume or `wakeUpRun` for that run cannot be delivered: a timer can never preempt a step, and `Promise.race(step, sleep())` always observes the step winning. Worlds now declare `capabilities.serializedRunContinuations`; when a run continuation suspends with a wait or hook armed (or one still open in the log), the runtime passes `allowInlineSteps: false` so every step gets its eager `step_created` and is queued instead. Pure step chains, turbo's hot path, and per-step-topic deliveries keep inline execution. Also point @workflow/nitro at `getWorkflowQueueTrigger()` so its flow trigger carries the matching `maxConcurrency`. Signed-off-by: Peter Wielander <peter.wielander@vercel.com>
VaguelySerious
removed
the
event-log-race-repro
1 participant
No description provided.