[world-vercel] Attest the executor's spec version on run_started by VaguelySerious · Pull Request #4366 · vercel/workflow

VaguelySerious · 2026-09-24T17:02:20Z

Companion to #4327 for #4251.

Problem

A run's spec version is stamped by whoever calls start(), and on world-vercel it decides the run's event-id scheme for life. #4327 makes a v5 caller stamp the target deployment's version. It cannot help when the caller is an older SDK that is already deployed: a v4 deployment calling start(wf, [state], { deploymentId: 'latest' }) after the project upgrades to v5 stamps spec 3. The backend then numbers the run's events by ULID, and the v5 runtime executing it fails every delivery (Event id is not slot-numbered) until the run ends with MAX_DELIVERIES_EXCEEDED. This is the handoff pattern the upgrading-workflows cookbook recommends, and the calling code runs on the old deployment, so no v4 release can fix it for runs already in flight.

Change

run_started now carries executorSpecVersion: the spec version this World declared when createWorld ran (mintedSpecVersion(), read once, so it follows WORKFLOW_SEALED_LOG at creation). It is read once so the attestation always matches what the runtime validated; the formal model on #4406 (E1) showed that a per-request env read could over-claim. specVersion on run_started is unchanged and still repeats the caller's stamp, which resilient start relies on.

The world-vercel backend uses it to raise a run stamped lower to this version: before any other event exists when the raise changes the event-id scheme (for example a v4-stamped run moving onto slot ids), and at any point before the run finishes when it only enables newer capabilities. The field is a new optional frame-meta key: a backend that does not know it drops it, so this can ship before or after the backend change, and without it the backend keeps its current behavior.

Only world-vercel is affected at runtime. world-local and world-postgres mint their own event ids and are unchanged.

@workflow/world also gets the classification the backend's raise depends on, next to the constants: CAPABILITY_ONLY_SPEC_VERSIONS (3, 4, 5, 8), STRUCTURAL_SPEC_VERSIONS (2, 6, 7) and crossesStructuralSpecVersion(from, to). A running run may only be raised across capability-only versions. The check is an allow-list, so an unclassified future version is structural by default, and a test fails when a SPEC_VERSION_SUPPORTS_* constant is added without a classification. vercel/workflow-server#1044 mirrors it in version-utils.ts.

Ordering. This has to ship with v5 GA for v4 → v5 handoffs to be repaired from day one. If it slips, the backend could fall back to the client's User-Agent (it already parses @workflow/world-vercel/<version>, and "world-vercel ≥ 5 needs slots" is fixed), but that is only worth building if this misses GA.

Tests

events.test.ts: run_started sends executorSpecVersion equal to mintedSpecVersion() next to the caller's specVersion, and follows the sealed-log kill switch.
spec-version.test.ts: every SPEC_VERSION_SUPPORTS_* constant is in exactly one set; slot identity and the sealed log are structural; capability-only versions can be crossed; an unclassified future version cannot.
pnpm --filter @workflow/world-vercel test: 753 passed. Typecheck clean.

🤖 Generated with Claude Code

run_started repeats the spec version the caller of start() stamped, which is older when the run was started from another deployment. Send the version this SDK mints alongside it, so the backend can move such a run onto the event-id scheme this runtime reads before any event is numbered. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>

changeset-bot · 2026-09-24T17:02:25Z

🦋 Changeset detected

Latest commit: 03e6e67

The changes in this PR will be included in the next version bump.

This PR includes changesets to release 21 packages

Name	Type
@workflow/world-vercel	Patch
@workflow/world	Patch
@workflow/cli	Patch
@workflow/core	Patch
@workflow/web	Patch
@workflow/errors	Patch
@workflow/vitest	Patch
@workflow/web-shared	Patch
@workflow/world-local	Patch
@workflow/world-postgres	Patch
@workflow/world-testing	Patch
workflow	Patch
@workflow/builders	Patch
@workflow/next	Patch
@workflow/nitro	Patch
@workflow/astro	Patch
@workflow/nest	Patch
@workflow/rollup	Patch
@workflow/sveltekit	Patch
@workflow/vite	Patch
@workflow/nuxt	Patch

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

github-actions · 2026-09-24T17:02:41Z

🧪 E2E Test Results

❌ Some tests failed

❌ Failed E2E Tests

▲ Vercel Production (62 failed)

python-node (62 failed):

promiseAllWorkflow | wrun_41M3BB1B2R0GYJGN16R625X5CJ | 🔍 observability
promiseRaceWorkflow | wrun_41M3BB1B2R0GYJGN16R625X5CK | 🔍 observability
promiseAnyWorkflow | wrun_41M3BB1B2S0GQC8GDCY4Q2RS5K | 🔍 observability
retainedInterleavingWorkflow | wrun_41M3BB1J4C0GTYQS7ZE88T6HGS | 🔍 observability
hookWorkflow | wrun_41M3BB1QY20GRCG431R34R73QM | 🔍 observability
hookWorkflow is not resumable via public webhook endpoint | wrun_41M3BB1RG60GHKHGRT9X9X0Q36 | 🔍 observability
sleepingWorkflow | wrun_41M3BB20V20GNX1GQ35TDWAY7E | 🔍 observability
parallelSleepWorkflow | wrun_41M3BB20YJ0GN3WT3XAM02SJ1W | 🔍 observability
sleepWinsRaceWorkflow | wrun_41M3BB22K80GS5P7PCDCYA3Z58 | 🔍 observability
stepWinsRaceWorkflow | wrun_41M3BB25PC0GNWQBRQ9DQME5A1 | 🔍 observability
nullByteWorkflow | wrun_41M3BB27RP0GWRC7KYNB7AD6WA | 🔍 observability
outputStreamWorkflow no startIndex (reads all chunks)
outputStreamWorkflow positive startIndex (skips first chunk)
outputStreamWorkflow negative startIndex (reads from end)
outputStreamWorkflow - getTailIndex and getChunks getTailIndex returns correct index after stream completes
outputStreamWorkflow - getTailIndex and getChunks getChunks returns same content as reading the stream
outputStreamInsideStepWorkflow - getWritable() called inside step functions | wrun_41M3BB2T370GQ81ZFTBT412N7T | 🔍 observability
utf8StreamWorkflow | wrun_41M3BB2WEC0GX6BS5SRH5Z2ERS | 🔍 observability
writableForwardedFromWorkflowWorkflow | wrun_41M3BB2YB60GNCDG74Q4NKXWCX | 🔍 observability
writableForwardedFromStepWorkflow | wrun_41M3BB2Z8B0GNCQKTCD34K2ZCR | 🔍 observability
promiseRaceStressTestWorkflow | wrun_41M3BB317S0GJDY5NQ225VVE1D | 🔍 observability
error handling retry behavior regular Error retries until success
error handling retry behavior FatalError fails immediately without retries
error handling retry behavior RetryableError respects custom retryAfter delay
error handling retry behavior maxRetries=0 disables retries
error handling catchability FatalError can be caught and detected with FatalError.is()
error handling catchability step throw round-trips FatalError with cause chain to workflow catch
error handling catchability workflow throw round-trips FatalError + cause through run_failed event
error handling not registered StepNotRegisteredError fails the step but workflow can catch it
error handling not registered StepNotRegisteredError fails the run when not caught in workflow
hookCleanupTestWorkflow - hook token reuse after workflow completion | wrun_41M3BB4J4M0GGDP8426R6F6FFG | 🔍 observability
concurrent hook token conflict - two workflows cannot use the same hook token simultaneously | wrun_41M3BB4F7F0GR6SBCSDA8CPEAT | 🔍 observability
hookGetConflictWorkflow - awaiting hook.getConflict() registers hook without payload | wrun_41M3BB4ETZ0GKYTEQMQN60X02H | 🔍 observability
'hookGetConflictWithPriorStepWorkflow' - hook.getConflict() does not block step execution | wrun_41M3BB4GS60GV117ZR9QJT7350 | 🔍 observability
'hookGetConflictWithParallelStepWorkfl…' - hook.getConflict() does not block step execution | wrun_41M3BB4M1G0GGY4JZB3V6YN401 | 🔍 observability
hookGetConflictThenStepParallelWorkflow - hook.getConflict() continuation step runs alongside other steps | wrun_41M3BB4M5Y0GY2MXH9S4TBCZXT | 🔍 observability
hookGetConflictWorkflow - hook.getConflict() resolves with the conflicting run when token is already registered | wrun_41M3BB4RJK0GT8AKJNBA9YXMC1 | 🔍 observability
hookClaimOnlyMutexWorkflow - hook works as a pure run mutex without payload data | wrun_41M3BB5JMM0GVV46R5DZCBC5YD | 🔍 observability
hookAdoptOwnerResultWorkflow - duplicate adopts the owner result via conflict.returnValue | wrun_41M3BB57H10GXYRQ0FDC9H41EW | 🔍 observability
hookSignalOwnerWorkflow - duplicate forwards its payload to the owner via resumeHook | wrun_41M3BB57JD0GHEP7NG6N1FS5EQ | 🔍 observability
resume-or-start route pattern - resumeHook retried after start() reaches the new run | wrun_41M3BB5XYD0GQBMHVH00FKQ21K | 🔍 observability
hookDisposeTestWorkflow - hook token reuse after explicit disposal while workflow still running | wrun_41M3BB629T0GTBYQ4FVAC26RB7 | 🔍 observability
hookTokenReuseLoopWorkflow - same run recreates a hook with the same token after dispose() | wrun_41M3BB601X0GRFA8YP7RV6CWCB | 🔍 observability
spawnWorkflowFromStepWorkflow - spawning a child workflow using start() inside a step | wrun_41M3BB6MW20GYFJG3XWEYQJX4N | 🔍 observability
cancelRun - cancelling a running workflow | wrun_41M3BB7HK90GZJBWSY2NDF0MZK | 🔍 observability
cancelRun via CLI - cancelling a running workflow | wrun_41M3BB7KYQ0GNRAFMJHCFGGK78 | 🔍 observability
hookWithSleepWorkflow - hook payloads delivered correctly with concurrent sleep | wrun_41M3BB7VFY0GMWBCT35Q6H0C43 | 🔍 observability
hookWithSleepFinalStepWorkflow - step only on final payload | wrun_41M3BB7W5V0GGNHJQEVQRC50HS | 🔍 observability
sleepInLoopWorkflow - sleep inside loop with steps actually delays each iteration | wrun_41M3BB7YGD0GZDKA54J7KEBKVT | 🔍 observability
sleepWithSequentialStepsWorkflow - sequential steps work with concurrent sleep (control) | wrun_41M3BB7Z7X0GVZ766DMBYSGAZK | 🔍 observability
AbortController abortTimeoutWorkflow: timeout cancels long-running step
AbortController abortParallelWorkflow: abort cancels all parallel steps
AbortController abortReasonWorkflow: abort reason preserved across boundaries
AbortController abortViaHookWorkflow: external hook triggers abort on in-flight step
metadataFromHelperWorkflow - getWorkflowMetadata/getStepMetadata work from module-level helper (#1577) | wrun_41M3BB8GJE0GW08SA6CFDFS7KG | 🔍 observability
resilient start: addTenWorkflow completes when run_created returns 500 | wrun_41M3BB8HV20GRTAR6MCKZJ67Q6 | 🔍 observability
setAttributes start: initial attributes are seeded on run creation
setAttributes start: reserved-prefix initial attributes are seeded with allowReservedAttributes
setAttributes setAttributesWorkflow: workflow-body calls append native attr_set events and merge correctly
setAttributes setAttributesInsideStepWorkflow: step-body calls append attributed native events
setAttributes Promise.all of disjoint-key writes: every key lands
setAttributes workflow throws after awaited setAttributes: attribute still persists on the failed run

⚠️ Flaky E2E Tests (passed on retry)

These tests failed at least once and passed on a retry. A recurring entry here is a real race worth investigating.

negative startIndex (reads from end) (express · vercel-prod / vercel / node / production)

🛠 Infra Events (absorbed by the harness)

Platform anomalies the e2e harness detected and worked around (e.g. a run the queue never picked up, replaced by a fresh run). Clustered timestamps indicate a backend blip; a steady drip indicates a platform issue worth escalating.

124 infra events

cold-start-warmup · suite warmup (python) · at 03:53:04Z · abandoned wrun_41M3BB0AHY0GJ0Y6C0NY6P1EQS · (+7 more)
run-pickup-stall · promiseAllWorkflow (python) · at 03:53:20Z · abandoned wrun_41M3BB3ZBB0GPTTGFSM6CMFJ1S
run-pickup-stall · promiseAnyWorkflow (python) · at 03:53:20Z · abandoned wrun_41M3BB3ZBB0GPTTGFSM6CMFJ1V
run-pickup-stall · promiseRaceWorkflow (python) · at 03:53:20Z · abandoned wrun_41M3BB3ZBB0GPTTGFSM6CMFJ1T
run-pickup-stall · retainedInterleavingWorkflow (python) · at 03:53:21Z · abandoned wrun_41M3BB3ZBB0GPTTGFSM6CMFJ1W
run-pickup-stall · hookWorkflow (python) · at 03:53:21Z · abandoned wrun_41M3BB3ZBG0GVJMDYBC0EHSZTG
run-pickup-stall · hookWorkflow (python) · at 03:54:08Z · abandoned wrun_41M3BB5ESV0GZRN4JBJR2CJPR4
run-pickup-stall · retainedInterleavingWorkflow (python) · at 03:54:09Z · abandoned wrun_41M3BB5F1P0GJZTYAKGN3DWRXN
run-pickup-stall · promiseAllWorkflow (python) · at 03:54:20Z · abandoned wrun_41M3BB5TP50GMP7X6WNSWA6R8R
run-pickup-stall · promiseRaceWorkflow (python) · at 03:54:21Z · abandoned wrun_41M3BB5TJY0GGZVWXE0GYBB0TG
run-pickup-stall · promiseAnyWorkflow (python) · at 03:54:21Z · abandoned wrun_41M3BB5TKY0GVC7D3K444A0SBV
run-pickup-stall · hookWorkflow is not resumable via public webhook endpoint (python) · at 03:54:55Z · abandoned wrun_41M3BB6WH30GHN8BY3XH2CZWXM
run-pickup-stall · sleepingWorkflow (python) · at 03:54:56Z · abandoned wrun_41M3BB6X150GVKZQGDNRDQVB1J
cold-start-warmup · suite warmup (tanstack-start) · at 03:55:06Z · abandoned wrun_01M3BB77VHQ437913S4XANFD0A
run-pickup-stall · parallelSleepWorkflow (python) · at 03:55:22Z · abandoned wrun_41M3BB7PA30GNCZHKX1AVY0MNQ
run-pickup-stall · sleepWinsRaceWorkflow (python) · at 03:55:23Z · abandoned wrun_41M3BB7PGM0GR5YA1PQEYJ8Q5X
run-pickup-stall · stepWinsRaceWorkflow (python) · at 03:55:23Z · abandoned wrun_41M3BB7PN20GWXB0XM60MN3WBA
run-pickup-stall · sleepingWorkflow (python) · at 03:56:51Z · abandoned wrun_41M3BB9DTJ0GXD16E1670CNA5W
run-pickup-stall · hookWorkflow is not resumable via public webhook endpoint (python) · at 03:56:51Z · abandoned wrun_41M3BB9E1F0GV7CS53NB0J2Q15
run-pickup-stall · sleepWinsRaceWorkflow (python) · at 03:56:58Z · abandoned wrun_41M3BBAED50GTT1MPM1GAFFK2V
run-pickup-stall · parallelSleepWorkflow (python) · at 03:56:58Z · abandoned wrun_41M3BBAE6M0GWTVXYQRT9PP3QH
run-pickup-stall · stepWinsRaceWorkflow (python) · at 03:56:58Z · abandoned wrun_41M3BBAEHM0GPD3458PJARNZ4E
run-pickup-stall · nullByteWorkflow (python) · at 03:57:19Z · abandoned wrun_41M3BBB90P0GZ2JE7JFY801RK7
run-pickup-stall · no startIndex (reads all chunks) (python) · at 03:57:19Z · abandoned wrun_41M3BBB9960GGT37PGV9HGHCNC
run-pickup-stall · positive startIndex (skips first chunk) (python) · at 03:57:51Z · abandoned wrun_41M3BBC8SK0GXRJS8V7DCP15SV
run-pickup-stall · negative startIndex (reads from end) (python) · at 03:57:52Z · abandoned wrun_41M3BBC9050GZSJ0EXBD17XZF9
run-pickup-stall · getTailIndex returns correct index after stream completes (python) · at 03:57:54Z · abandoned wrun_41M3BBCB3P0GTWPYAJGRPYEZ4G
run-pickup-stall · getTailIndex returns -1 before any chunks are written (python) · at 03:58:19Z · abandoned wrun_41M3BBD3KQ0GYFWQP0G2H1K27H
run-pickup-stall · getChunks returns same content as reading the stream (python) · at 03:58:19Z · abandoned wrun_41M3BBD3W60GH0YCT6FPEZDY7R
run-pickup-stall · outputStreamInsideStepWorkflow - getWritable() called inside step functions (python) · at 03:58:35Z · abandoned wrun_41M3BBDKAR0GREKS1BQ8WCQ2PJ
run-pickup-stall · utf8StreamWorkflow (python) · at 03:58:51Z · abandoned wrun_41M3BBE3CM0GV6A1HMZ60K2NT1
run-pickup-stall · no startIndex (reads all chunks) (python) · at 03:58:54Z · abandoned wrun_41M3BBE5VG0GPMNWBDK8ZWVX0F
run-pickup-stall · nullByteWorkflow (python) · at 03:58:54Z · abandoned wrun_41M3BBE5PN0GS8ETYQZDBDK4YJ
run-pickup-stall · positive startIndex (skips first chunk) (python) · at 03:59:19Z · abandoned wrun_41M3BBEYF60GT86CBW4JHYJ60A
run-pickup-stall · negative startIndex (reads from end) (python) · at 03:59:36Z · abandoned wrun_41M3BBFEGP0GZZ3MNRBF8NMSKE
run-pickup-stall · getTailIndex returns correct index after stream completes (python) · at 03:59:54Z · abandoned wrun_41M3BBG0VB0GSKKZ4M5DX3E1GC
run-pickup-stall · getChunks returns same content as reading the stream (python) · at 03:59:56Z · abandoned wrun_41M3BBG25Y0GVF9Z26Q0Q6R4GS
run-pickup-stall · hookCleanupTestWorkflow - hook token reuse after workflow completion (nextjs-webpack) · at 04:00:17Z · abandoned wrun_01M3BBGQPN4J1DEJTSXSXXXRX7
run-pickup-stall · outputStreamInsideStepWorkflow - getWritable() called inside step functions (python) · at 04:00:19Z · abandoned wrun_41M3BBGS260GTMXA4KFCCPBD96
run-pickup-stall · writableForwardedFromWorkflowWorkflow (python) · at 04:00:36Z · abandoned wrun_41M3BBH93Q0GGGQEEEFH0VTWR6
run-pickup-stall · writableForwardedFromStepWorkflow (python) · at 04:00:53Z · abandoned wrun_41M3BBHTD80GQC49E60QJV5TE1
run-pickup-stall · utf8StreamWorkflow (python) · at 04:00:55Z · abandoned wrun_41M3BBHVEB0GXFQF67DBCJE2EX
run-pickup-stall · promiseRaceStressTestWorkflow (python) · at 04:00:57Z · abandoned wrun_41M3BBHXXP0GR3FZS66GC62EGW
run-pickup-stall · regular Error retries until success (python) · at 04:01:20Z · abandoned wrun_41M3BBJM9B0GKF3W6R62N48EJW
run-pickup-stall · FatalError fails immediately without retries (python) · at 04:01:57Z · abandoned wrun_41M3BBKRGQ0GY5DFQJV0364QFN
run-pickup-stall · RetryableError respects custom retryAfter delay (python) · at 04:02:20Z · abandoned wrun_41M3BBMEWC0GG1R8V3GA67CD55
run-pickup-stall · maxRetries=0 disables retries (python) · at 04:02:36Z · abandoned wrun_41M3BBMY9R0GZRRWASSGY266MX
run-pickup-stall · FatalError can be caught and detected with FatalError.is() (python) · at 04:02:53Z · abandoned wrun_41M3BBNFK90GMZ8ZEEF9Y8TW0J
run-pickup-stall · step throw round-trips FatalError with cause chain to workflow catch (python) · at 04:02:54Z · abandoned wrun_41M3BBNGMB0GSNCQSZC371A88W
run-pickup-stall · workflow throw round-trips FatalError + cause through run_failed event (python) · at 04:02:57Z · abandoned wrun_41M3BBNK3Q0GTAHQ4VP4NVRXKQ
run-pickup-stall · StepNotRegisteredError fails the step but workflow can catch it (python) · at 04:03:20Z · abandoned wrun_41M3BBP9FC0GVHZYHWQVKTBVAT
run-pickup-stall · StepNotRegisteredError fails the run when not caught in workflow (python) · at 04:03:36Z · abandoned wrun_41M3BBPRWR0GQ0ERTPP1E55VJ0
run-pickup-stall · promiseRaceStressTestWorkflow (python) · at 04:05:07Z · abandoned wrun_41M3BBRJPT0GJ8FVFJ01M5BP2K
run-pickup-stall · regular Error retries until success (python) · at 04:05:08Z · abandoned wrun_41M3BBRK6Q0GJ1QERVPRS5XYJS
run-pickup-stall · writableForwardedFromWorkflowWorkflow (python) · at 04:05:08Z · abandoned wrun_41M3BBRKFR0GK4TNW121MH1VF4
run-pickup-stall · writableForwardedFromStepWorkflow (python) · at 04:05:10Z · abandoned wrun_41M3BBS5RJ0GM657QX0KQXH1JA
run-pickup-stall · hookCleanupTestWorkflow - hook token reuse after workflow completion (python) · at 04:05:31Z · abandoned wrun_41M3BBS8NW0GRDHQYRB5HV72T9
run-pickup-stall · FatalError fails immediately without retries (python) · at 04:06:11Z · abandoned wrun_41M3BBTD9T0GRF4T4MVCNFE1S9
run-pickup-stall · RetryableError respects custom retryAfter delay (python) · at 04:06:11Z · abandoned wrun_41M3BBTDSQ0GKD3JXPFR6AEDBV
run-pickup-stall · maxRetries=0 disables retries (python) · at 04:06:37Z · abandoned wrun_41M3BBV38W0GQXGRRE340TWH3W
run-pickup-stall · FatalError can be caught and detected with FatalError.is() (python) · at 04:06:49Z · abandoned wrun_41M3BBW7WT0GZR8B7TEZ7GMF60
run-pickup-stall · step throw round-trips FatalError with cause chain to workflow catch (python) · at 04:06:49Z · abandoned wrun_41M3BBW8NR0GYJ7W5VB4AJSF5B
run-pickup-stall · workflow throw round-trips FatalError + cause through run_failed event (python) · at 04:06:56Z · abandoned wrun_41M3BBWSQF0GPJCT541NJAZM64
run-pickup-stall · StepNotRegisteredError fails the step but workflow can catch it (python) · at 04:06:57Z · abandoned wrun_41M3BBWTYJ0GHYNJV51TRMEXQD
run-pickup-stall · StepNotRegisteredError fails the run when not caught in workflow (python) · at 04:07:07Z · abandoned wrun_41M3BBX7P80GHDBF9TJ3FEY7YB
run-pickup-stall · hookCleanupTestWorkflow - hook token reuse after workflow completion (python) · at 04:07:36Z · abandoned wrun_41M3BBY3RJ0GS2630GX7F8QKX3
run-pickup-stall · hookGetConflictWorkflow - awaiting hook.getConflict() registers hook without payload (python) · at 04:07:53Z · abandoned wrun_41M3BBYMAF0GN2ADG619B6DQ66
run-pickup-stall · 'hookGetConflictWithPriorStepWorkflow' - hook.getConflict() does not block step execution (python) · at 04:07:55Z · abandoned wrun_41M3BBYP9G0GRAYXPQT1J2JFN1
run-pickup-stall · 'hookGetConflictWithParallelStepWorkfl…' - hook.getConflict() does not block step execution (python) · at 04:08:08Z · abandoned wrun_41M3BBZ37C0GZDEFMGF730C2AA
run-pickup-stall · hookGetConflictThenStepParallelWorkflow - hook.getConflict() continuation step runs alongside other steps (python) · at 04:08:23Z · abandoned wrun_41M3BBZHED0GM90Q9VX9S6KTK0
run-pickup-stall · hookGetConflictWorkflow - hook.getConflict() resolves with the conflicting run when token is already registered (python) · at 04:08:39Z · abandoned wrun_41M3BC019F0GY148C00E6TGF48
run-pickup-stall · hookGetConflictWorkflow - awaiting hook.getConflict() registers hook without payload (python) · at 04:08:54Z · abandoned wrun_41M3BC0FMD0GNBR36W7T5XZ237
run-pickup-stall · 'hookGetConflictWithPriorStepWorkflow' - hook.getConflict() does not block step execution (python) · at 04:08:56Z · abandoned wrun_41M3BC0HDC0GTSJHYEY6Z1R7FP
run-pickup-stall · 'hookGetConflictWithParallelStepWorkfl…' - hook.getConflict() does not block step execution (python) · at 04:09:09Z · abandoned wrun_41M3BC0YCV0GQ6BKVRXMSV9B2Q
run-pickup-stall · hookGetConflictWorkflow - hook.getConflict() resolves with the conflicting run when token is already registered (python) · at 04:09:26Z · abandoned wrun_41M3BC1EXT0GQADDP3V2Z57VV4
run-pickup-stall · hookGetConflictThenStepParallelWorkflow - hook.getConflict() continuation step runs alongside other steps (python) · at 04:09:53Z · abandoned wrun_41M3BC29Y10GSD0RJXPFB41Y1A
run-pickup-stall · hookClaimOnlyMutexWorkflow - hook works as a pure run mutex without payload data (python) · at 04:09:55Z · abandoned wrun_41M3BC2ATT0GTFZJW527GKHVKQ
run-pickup-stall · hookAdoptOwnerResultWorkflow - duplicate adopts the owner result via conflict.returnValue (python) · at 04:09:57Z · abandoned wrun_41M3BC2CMH0GRTPQPTHQPJAN2F
run-pickup-stall · hookSignalOwnerWorkflow - duplicate forwards its payload to the owner via resumeHook (python) · at 04:10:10Z · abandoned wrun_41M3BC2SPF0GJA2H92W5QNDE34
run-pickup-stall · resume-or-start route pattern - resumeHook retried after start() reaches the new run (python) · at 04:10:13Z · abandoned wrun_41M3BC2X1R0GNKM85H72VH123M
run-pickup-stall · hookClaimOnlyMutexWorkflow - hook works as a pure run mutex without payload data (python) · at 04:10:41Z · abandoned wrun_41M3BC3RM60GQ26224YFMPQY27
run-pickup-stall · hookAdoptOwnerResultWorkflow - duplicate adopts the owner result via conflict.returnValue (python) · at 04:10:43Z · abandoned wrun_41M3BC3THZ0GN4SVW8TA6FV5CZ
run-pickup-stall · hookSignalOwnerWorkflow - duplicate forwards its payload to the owner via resumeHook (python) · at 04:10:57Z · abandoned wrun_41M3BC47G90GYEQ28RHGVPCKBP
run-pickup-stall · resume-or-start route pattern - resumeHook retried after start() reaches the new run (python) · at 04:11:00Z · abandoned wrun_41M3BC4BE00GN2VGTVNEHH0XEZ
run-pickup-stall · hookDisposeTestWorkflow - hook token reuse after explicit disposal while workflow still running (python) · at 04:11:24Z · abandoned wrun_41M3BC52E30GR8FFY6P9NRJT02
run-pickup-stall · hookTokenReuseLoopWorkflow - same run recreates a hook with the same token after dispose() (python) · at 04:11:28Z · abandoned wrun_41M3BC567S0GPQ881Q0PXV3N5P
run-pickup-stall · spawnWorkflowFromStepWorkflow - spawning a child workflow using start() inside a step (python) · at 04:11:30Z · abandoned wrun_41M3BC582D0GKTF24S02GBWG1Q
run-pickup-stall · cancelRun - cancelling a running workflow (python) · at 04:11:47Z · abandoned wrun_41M3BC5S7Y0GXVXG7GH17MZTVM
run-pickup-stall · cancelRun via CLI - cancelling a running workflow (python) · at 04:11:49Z · abandoned wrun_41M3BC5TZE0GT0Z36JNB97G4W9
run-pickup-stall · hookDisposeTestWorkflow - hook token reuse after explicit disposal while workflow still running (python) · at 04:12:11Z · abandoned wrun_41M3BC6G4R0GN3TMVJAFN7J2C9
run-pickup-stall · hookTokenReuseLoopWorkflow - same run recreates a hook with the same token after dispose() (python) · at 04:12:14Z · abandoned wrun_41M3BC6KNB0GHGJQ4BK2B1Q4YY
run-pickup-stall · cancelRun - cancelling a running workflow (python) · at 04:12:19Z · abandoned wrun_41M3BC6R3W0GXEYKENDCBWSN8Q
run-pickup-stall · cancelRun via CLI - cancelling a running workflow (python) · at 04:12:21Z · abandoned wrun_41M3BC6T050GJ7CFSVHZ6YV0MA
run-pickup-stall · hookWithSleepWorkflow - hook payloads delivered correctly with concurrent sleep (python) · at 04:12:51Z · abandoned wrun_41M3BC7QA10GK61J5C57CXBSMA
run-pickup-stall · hookWithSleepFinalStepWorkflow - step only on final payload (python) · at 04:12:53Z · abandoned wrun_41M3BC7RZV0GG0G52Y661VCVCN
run-pickup-stall · sleepInLoopWorkflow - sleep inside loop with steps actually delays each iteration (python) · at 04:12:58Z · abandoned wrun_41M3BC7XRA0GH919RCC031GQGM
run-pickup-stall · sleepWithSequentialStepsWorkflow - sequential steps work with concurrent sleep (control) (python) · at 04:13:01Z · abandoned wrun_41M3BC81G00GNNMXM7EH40Z973
run-pickup-stall · spawnWorkflowFromStepWorkflow - spawning a child workflow using start() inside a step (python) · at 04:13:31Z · abandoned wrun_41M3BC8XRM0GXASW8QR6T912PJ
run-pickup-stall · hookWithSleepWorkflow - hook payloads delivered correctly with concurrent sleep (python) · at 04:13:38Z · abandoned wrun_41M3BC952T0GN1FFCEQDCTTTMX
run-pickup-stall · hookWithSleepFinalStepWorkflow - step only on final payload (python) · at 04:13:40Z · abandoned wrun_41M3BC96WX0GZBHRWXBKY1K6AQ
run-pickup-stall · sleepInLoopWorkflow - sleep inside loop with steps actually delays each iteration (python) · at 04:13:59Z · abandoned wrun_41M3BC9RXN0GP032HSP6QQPRW4
run-pickup-stall · sleepWithSequentialStepsWorkflow - sequential steps work with concurrent sleep (control) (python) · at 04:14:03Z · abandoned wrun_41M3BC9WMY0GYRDE5Y9AAHACZW
run-pickup-stall · abortTimeoutWorkflow: timeout cancels long-running step (python) · at 04:14:25Z · abandoned wrun_41M3BCAK2K0GT2HDG7BP2T7K3V
run-pickup-stall · abortParallelWorkflow: abort cancels all parallel steps (python) · at 04:14:27Z · abandoned wrun_41M3BCAM200GY93WXBR7KMRAPM
run-pickup-stall · abortReasonWorkflow: abort reason preserved across boundaries (python) · at 04:15:12Z · abandoned wrun_41M3BCBKGN0GXJ7THYSBYE8V7Q
run-pickup-stall · abortViaHookWorkflow: external hook triggers abort on in-flight step (python) · at 04:15:45Z · abandoned wrun_41M3BCCDNK0GKCXEGCEG7X2SXA
run-pickup-stall · metadataFromHelperWorkflow - getWorkflowMetadata/getStepMetadata work from module-level helper (Vercel World Regression: getWorkflowMetadata() fully breaking runs #1577) (python) · at 04:15:48Z · abandoned wrun_41M3BCCFD10GV7ES9XC836Q1GW
run-pickup-stall · abortTimeoutWorkflow: timeout cancels long-running step (python) · at 04:16:46Z · abandoned wrun_41M3BCDJV80GXZ5VA5SERK4NJH
run-pickup-stall · abortParallelWorkflow: abort cancels all parallel steps (python) · at 04:17:09Z · abandoned wrun_41M3BCEA010GSX3SEK3TNNDC08
run-pickup-stall · start: initial attributes are seeded on run creation (python) · at 04:17:21Z · abandoned wrun_41M3BCF2C80GVADJ8D3AA101SB
run-pickup-stall · start: reserved-prefix initial attributes are seeded with allowReservedAttributes (python) · at 04:17:22Z · abandoned wrun_41M3BCF7KA0GJSR6ESP4JM711Q
run-pickup-stall · setAttributesWorkflow: workflow-body calls append native attr_set events and merge correctly (python) · at 04:17:33Z · abandoned wrun_41M3BCFDE80GRS3N4J01G3RDK0
run-pickup-stall · setAttributesInsideStepWorkflow: step-body calls append attributed native events (python) · at 04:17:36Z · abandoned wrun_41M3BCFY9D0GSV07FNRCTGEAYF
run-pickup-stall · abortReasonWorkflow: abort reason preserved across boundaries (python) · at 04:17:40Z · abandoned wrun_41M3BCG4K10GVDXCMECJ9DWG63
run-pickup-stall · abortViaHookWorkflow: external hook triggers abort on in-flight step (python) · at 04:17:41Z · abandoned wrun_41M3BCG4WT0GGRWTFYBGVYHQGN
run-pickup-stall · metadataFromHelperWorkflow - getWorkflowMetadata/getStepMetadata work from module-level helper (Vercel World Regression: getWorkflowMetadata() fully breaking runs #1577) (python) · at 04:17:44Z · abandoned wrun_41M3BCGC3D0GGQ9WYWBT9DYAFX
run-pickup-stall · Promise.all of disjoint-key writes: every key lands (python) · at 04:17:52Z · abandoned wrun_41M3BCGVJY0GZ576SDHKSR43FC
run-pickup-stall · start: initial attributes are seeded on run creation (python) · at 04:18:09Z · abandoned wrun_41M3BCHDN90GJ7F3X35GKE5FX1
run-pickup-stall · start: reserved-prefix initial attributes are seeded with allowReservedAttributes (python) · at 04:18:20Z · abandoned wrun_41M3BCHRWF0GPM777GX1D8EVCS
run-pickup-stall · setAttributesWorkflow: workflow-body calls append native attr_set events and merge correctly (python) · at 04:18:26Z · abandoned wrun_41M3BCHYYE0GVGCV4QMJRR0FHN
run-pickup-stall · setAttributesInsideStepWorkflow: step-body calls append attributed native events (python) · at 04:18:27Z · abandoned wrun_41M3BCHZCY0GSQ2VVTS0BQDD5H
run-pickup-stall · Promise.all of disjoint-key writes: every key lands (python) · at 04:18:37Z · abandoned wrun_41M3BCJ8W70GYQGHW3X3F65P8P
run-pickup-stall · workflow throws after awaited setAttributes: attribute still persists on the failed run (python) · at 04:18:39Z · abandoned wrun_41M3BCJBJ40GGD0V730TH8ZYED
run-pickup-stall · workflow throws after awaited setAttributes: attribute still persists on the failed run (python) · at 04:19:10Z · abandoned wrun_41M3BCK9CM0GMMHGGYXDV3F4M7

E2E Test Summary

Summary

	Passed	Failed	Skipped	Total
❌ ▲ Vercel Production	3816	62	739	4617
✅ 💻 Local Development	4238	0	550	4788
✅ 📦 Local Production	4238	0	550	4788
✅ 🐘 Local Postgres	4238	0	550	4788
✅ 🪟 Windows	340	0	2	342
✅ vercel-http-transport	873	0	153	1026
✅ vercel-multi-region	27	0	0	27
✅ vercel-ws-transport	591	0	93	684
Total	18361	62	2637	21060

Details by Category

❌ ▲ Vercel Production

App	Passed	Failed	Skipped
✅ astro-node	141	0	30
✅ astro-quickjs	141	0	30
✅ example-node	141	0	30
✅ example-quickjs	141	0	30
✅ express-node	141	0	30
✅ express-quickjs	141	0	30
✅ fastify-node	141	0	30
✅ fastify-quickjs	141	0	30
✅ hono-node	141	0	30
✅ hono-quickjs	141	0	30
✅ nest-node	141	0	30
✅ nest-quickjs	141	0	30
✅ nextjs-turbopack-node	168	0	3
✅ nextjs-turbopack-quickjs	168	0	3
✅ nextjs-webpack-node	168	0	3
✅ nextjs-webpack-quickjs	168	0	3
✅ nitro-node	141	0	30
✅ nitro-quickjs	141	0	30
✅ nuxt-node	141	0	30
✅ nuxt-quickjs	141	0	30
❌ python-node	4	62	105
✅ sveltekit-node	160	0	11
✅ sveltekit-quickjs	160	0	11
✅ tanstack-start-node	141	0	30
✅ tanstack-start-quickjs	141	0	30
✅ vite-node	141	0	30
✅ vite-quickjs	141	0	30

✅ 💻 Local Development

App	Passed	Skipped
✅ astro-stable-node	142	29
✅ astro-stable-quickjs	142	29
✅ express-stable-node	142	29
✅ express-stable-quickjs	142	29
✅ fastify-stable-node	142	29
✅ fastify-stable-quickjs	142	29
✅ hono-stable-node	142	29
✅ hono-stable-quickjs	142	29
✅ nest-stable-node	142	29
✅ nest-stable-quickjs	142	29
✅ nextjs-turbopack-canary-node	170	1
✅ nextjs-turbopack-canary-quickjs	170	1
✅ nextjs-turbopack-stable-node	170	1
✅ nextjs-turbopack-stable-quickjs	170	1
✅ nextjs-webpack-canary-node	170	1
✅ nextjs-webpack-canary-quickjs	170	1
✅ nextjs-webpack-stable-node	170	1
✅ nextjs-webpack-stable-quickjs	170	1
✅ nitro-stable-node	142	29
✅ nitro-stable-quickjs	142	29
✅ nuxt-stable-node	142	29
✅ nuxt-stable-quickjs	142	29
✅ sveltekit-stable-node	161	10
✅ sveltekit-stable-quickjs	161	10
✅ tanstack-start-node	142	29
✅ tanstack-start-quickjs	142	29
✅ vite-stable-node	142	29
✅ vite-stable-quickjs	142	29

✅ 📦 Local Production

App	Passed	Skipped
✅ astro-stable-node	142	29
✅ astro-stable-quickjs	142	29
✅ express-stable-node	142	29
✅ express-stable-quickjs	142	29
✅ fastify-stable-node	142	29
✅ fastify-stable-quickjs	142	29
✅ hono-stable-node	142	29
✅ hono-stable-quickjs	142	29
✅ nest-stable-node	142	29
✅ nest-stable-quickjs	142	29
✅ nextjs-turbopack-canary-node	170	1
✅ nextjs-turbopack-canary-quickjs	170	1
✅ nextjs-turbopack-stable-node	170	1
✅ nextjs-turbopack-stable-quickjs	170	1
✅ nextjs-webpack-canary-node	170	1
✅ nextjs-webpack-canary-quickjs	170	1
✅ nextjs-webpack-stable-node	170	1
✅ nextjs-webpack-stable-quickjs	170	1
✅ nitro-stable-node	142	29
✅ nitro-stable-quickjs	142	29
✅ nuxt-stable-node	142	29
✅ nuxt-stable-quickjs	142	29
✅ sveltekit-stable-node	161	10
✅ sveltekit-stable-quickjs	161	10
✅ tanstack-start-node	142	29
✅ tanstack-start-quickjs	142	29
✅ vite-stable-node	142	29
✅ vite-stable-quickjs	142	29

✅ 🐘 Local Postgres

App	Passed	Skipped
✅ astro-stable-node	142	29
✅ astro-stable-quickjs	142	29
✅ express-stable-node	142	29
✅ express-stable-quickjs	142	29
✅ fastify-stable-node	142	29
✅ fastify-stable-quickjs	142	29
✅ hono-stable-node	142	29
✅ hono-stable-quickjs	142	29
✅ nest-stable-node	142	29
✅ nest-stable-quickjs	142	29
✅ nextjs-turbopack-canary-node	170	1
✅ nextjs-turbopack-canary-quickjs	170	1
✅ nextjs-turbopack-stable-node	170	1
✅ nextjs-turbopack-stable-quickjs	170	1
✅ nextjs-webpack-canary-node	170	1
✅ nextjs-webpack-canary-quickjs	170	1
✅ nextjs-webpack-stable-node	170	1
✅ nextjs-webpack-stable-quickjs	170	1
✅ nitro-stable-node	142	29
✅ nitro-stable-quickjs	142	29
✅ nuxt-stable-node	142	29
✅ nuxt-stable-quickjs	142	29
✅ sveltekit-stable-node	161	10
✅ sveltekit-stable-quickjs	161	10
✅ tanstack-start-node	142	29
✅ tanstack-start-quickjs	142	29
✅ vite-stable-node	142	29
✅ vite-stable-quickjs	142	29

✅ 🪟 Windows

App	Passed	Failed	Skipped
✅ nextjs-turbopack-node	170	0	1
✅ nextjs-turbopack-quickjs	170	0	1

✅ vercel-http-transport

App	Passed	Skipped
✅ example	141	30
✅ express	141	30
✅ hono	141	30
✅ nextjs-turbopack	168	3
✅ nitro	141	30
✅ vite	141	30

✅ vercel-multi-region

App	Passed	Failed	Skipped
✅ nextjs-turbopack	27	0	0

✅ vercel-ws-transport

App	Passed	Skipped
✅ example	141	30
✅ express	141	30
✅ nextjs-turbopack	168	3
✅ vite	141	30

📋 View full workflow run

github-actions · 2026-09-24T17:02:42Z

📊 Workflow Benchmarks

commit 03e6e67 · Fri, 25 Sep 2026 04:09:12 GMT · run logs

Backend: vercel · app: nextjs-turbopack

Metric	Scenario	Best (ms)	P75 (ms)	P90 (ms)	P99 (ms)	Samples
TTFS	step	2078 (+46%) 🔻	2497 🔴 (+37%) 🔻	2570 🔴 (+37%) 🔻	2708 🔴 (+36%) 🔻	30
TTFS	stream	178 (-20%) 💚	2431 🔴 (+17%) 🔻	2459 🔴 (+16%) 🔻	2487 🔴 (+7.8%)	30
TTFS	hook + stream	538 (+45%) 🔻	2704 🔴 (+35%) 🔻	2767 🔴 (+32%) 🔻	2922 🔴 (+26%) 🔻	30
Fan-out TTFS	Promise.all(100 steps)	494 (-1.2%)	749 (+0.8%)	914 (+22%) 🔻	2617 (+22%) 🔻	10
Fan-out TTLS	Promise.all(100 steps)	2945 (+39%) 🔻	4181 (-17%) 💚	5583 (-5.5%)	10027 (+26%) 🔻	10
STSO	1020 steps (inline)	96 (-19%) 💚	140 (-11%)	159 (-12%)	248 (-31%) 💚	1019
WO	1020 steps	140715 (-10%)	140715 (-10%)	140715 (-10%)	140715 (-10%)	1
CRTT	first chunk (pooled)	63 (-17%) 💚	96 (-19%) 💚	126 (-11%)	185 (-21%) 💚	28

Streams

Scenario	CRTT 1st	p75	p90	p99	CDV max	iters
paced control (100/s, 60B)	83.5 (-22%)	118 (-45%)	138 (-66%)	196 (-83%)	89 (-66%)	10
size sweep (100/s, 160B-12KB)	91.5 (-18%)	122 (-35%)	159 (-38%)	269 (-37%)	98 (-38%)	10
replay gateway-gpt-5.4-nano-2000t (1x)	81 (-28%)	118 (-38%)	134 (-47%)	366 (-39%)	354 (+9%)	3
replay eve-gpt-5.6-sol-2000t (1x)	97.5 (-16%)	125 (-31%)	154 (-33%)	216 (-45%)	269 (-26%)	2
replay eve-gpt-5.6-sol-2000t (2x)	81 (-30%)	821 (+244%)	2284 (+637%)	4307 (+772%)	406 (+10%)	3

📈 STSO distribution vs main (inline / queue-hop histograms)

1020 steps (inline)

Cumulative STSO time: main 156783ms → this run 140582ms (Δ -16201ms, -10%)

 50-100 ms  ┃                         main   0  this   1    +1
100-150 ms  ███████████████████░░░░┃  main 675  this 863  +188
150-200 ms  ██┃█████                  main 283  this 123  -160
200-250 ms  ┃                         main  27  this  22    -5
250-300 ms  ┃                         main  12  this   5    -7
300-350 ms  ┃                         main  11  this   2    -9
350-400 ms  ┃                         main   3  this   0    -3
400-450 ms  ┃                         main   0  this   2    +2
450-500 ms  ┃                         main   3  this   1    -2
500-550 ms  ┃                         main   2  this   0    -2
550-600 ms  ┃                         main   1  this   0    -1
600-650 ms  ┃                         main   1  this   0    -1
700-750 ms  ┃                         main   1  this   0    -1

📈 CRTT drill-down vs main (RTT distributions & profiles)

variant  RTT 1ms→5s+             avg         p50           p90           p99     n
control  ······██·····  101.4 (-44%)  100 (-31%)    138 (-66%)    196 (-83%)  3000
sweep    ······██▁····  104.5 (-32%)   99 (-32%)    159 (-38%)    269 (-37%)  3000
gw 1x    ·····▁██▁····  105.3 (-33%)  101 (-27%)    134 (-47%)    366 (-39%)  5295
eve 1x   ·····▁▇█▁····  106.2 (-32%)  100 (-29%)    154 (-33%)    216 (-45%)  5186
eve 2x   ·····▁▄█▂▂▁▁·    347 (+79%)  116 (-33%)  2284 (+637%)  4307 (+772%)  7779

RTT over stream progress (avg per tenth of stream, bars scaled min→max):

control  █▅▅▃▁▄▂▂▁▄  97–111ms
sweep    █▂▄▄▃▁▃▂▅▂  99–116ms
gw 1x    █▃▁▁▂▄▅▂▂▄  97–124ms
eve 1x   █▁▂▄▄▃█▇▃▂  100–115ms
eve 2x   ▂█▃▂▁▁▁▁▂▁  164–1101ms

RTT by chunk size (avg per log size bin, ~160B → ~12KB serialized, bars scaled min→max):

sweep  ▆█▇▅▃▁▁  101–109ms

Delivery jitter over stream progress (avg positive CDV per tenth of stream, bars scaled min→max):

control  ▁▅▃▁█▅▇▃▁▃  29–33ms
sweep    ▁▂▃▄▁█▄▅▅▃  38–48ms
gw 1x    █▂▄▁▄█▄▄▁▆  27–36ms
eve 1x   ▅▆▁▅▆▄█▅▅▄  19–25ms
eve 2x   █▁▁▄▄▃▃▂▄▄  18–48ms

ℹ️ Metric definitions & methodology

_{Streams: first-chunk RTT (the stream-open path, before any buffering/backpressure), CRTT percentiles, and worst delivery stall (CDV max). Cells are medians across iterations; per-run values in the artifacts. No 🔴/🟢 marks until targets attach.}

_{The collapsed STSO distribution section above buckets every step gap, split inline (same warm process — pure framework overhead) vs queue-hop (fresh process — dispatch, reinit, replay). █ = main, ┃ = this run, ░ = fill.}

_{The collapsed CRTT drill-down: per-variant RTT histograms (fixed log bins, · = empty) and mean RTT/positive-CDV profile lines over stream progress and chunk size. Histograms, avgs, and profiles merge exactly across runs; p50–p99 are percentile-of-percentiles. Per-index rows live in the artifacts.}

_{Best/P75/P90/P99 deltas compare against the most recent benchmark run on main at the time of this run. 🔻 flags a delta worse than +15%, 💚 one better than −15%.}

_{Metrics — TTFS: time to first step body (in-deployment start() → first step body) · Fan-out TTFS: fan-out time to first step (in-deployment start() → first of the parallel step bodies to complete) · Fan-out TTLS: fan-out time to last step (in-deployment start() → last of the parallel step bodies to complete, i.e. when the Promise.all resolves) · STSO: step-to-step overhead (gap between consecutive step bodies) · WO: workflow overhead (whole-run time outside step bodies, in-deployment anchored) · CRTT: chunk round-trip time (per-chunk write → read latency, one clock domain: deployment → stream backend → same deployment) · CDV: chunk delay variation / delivery jitter (inter-arrival gap minus inter-write gap per seq-adjacent pair; skew-free; the row is each run's MAX positive value, so one stall moves it)}

_{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 · Promise.all(100 steps): 100 trivial no-op steps started together in a single Promise.all; Fan-out TTFS is the first of them to complete and Fan-out TTLS the last, both from the in-deployment clientStart, so their gap is the spread the runtime adds across the fan-out · paced control (100/s, 60B): the control: 300 tiny (~60B) deltas metronome-paced at 100/s — zero workload structure, so it reads the transport floor and flush cadence, and disambiguates transport-wide vs workload-specific when a replay row moves · size sweep (100/s, 160B-12KB): same pacing as the control with deltas padded in rotation across seven log-spaced sizes (~160B–12KB) — rotation decouples size from stream position, so it isolates whether chunk size causes latency · replay gateway-gpt-5.4-nano-2000t (1x): raw provider SSE cadence captured at the AI gateway boundary (gpt-5.4-nano, the most popular gateway model; per-token deltas p50 208B = the modal production chunk size), replayed exactly as measured — the typical customer's workload; its CDV is the typical customer's real delivery jitter · replay eve-gpt-5.6-sol-2000t (1x): a captured eve turn (gpt-5.6-sol, the most-used demanding eve model; ~2000 output tokens = production p50 turn length) replayed exactly as measured — eve's envelope protocol re-ships the cumulative message so sizes ramp 142B→13KB; the demanding outlier tenant's reality · replay eve-gpt-5.6-sol-2000t (2x): the same eve capture at 2x — the headroom/stress row; real fast-tier models emit the same chunk sizes at proportionally higher rate, so time compression is a faithful speed model · first chunk (pooled): every run's seq-0 RTT pooled across all stream scenarios — the first chunk precedes any workload differentiation, so pooling samples one shared stream-open path with exact percentiles}

_{Replay cadences (semantic sha256) — eve-gpt-5.6-sol-2000t eaf22f5946e7c61f3c65c7006d550df180cfabd4e706254a09f22aec0cfb420d · gateway-gpt-5.4-nano-2000t 6f24ac518b6b83ff1d0e85a5fe78230db192716d66a7fc6b2fe022752001d041}

_{🔴 marks a percentile over its target (within target is left unmarked). Targets (p75/p90/p99, ms) — TTFS 200/300/600}

_{All timestamps are deployment-side; runs are triggered in-deployment, so the CI runner and api.vercel.com sit outside every measured window. TTFS = start() → first step body (includes dispatch + any cold start); Fan-out TTFS/TTLS = first/last step completion of one Promise.all from the same anchor (the gap is the runtime’s fan-out spread); STSO/WO between step bodies; CRTT inside the workflow (excludes the api.vercel.com read path).}

_{Cold starts stay in the numbers (real bursty-workload latency, inflates P75+); Best is the warm floor.}

github-actions · 2026-09-24T17:06:25Z

Sim World

Simulated world deterministic testing for races. Traces

🟠 world-sim scenario book — 1 fail of 42 total

fence=per-spec

scenario	outcome	events	virt	replay
✅ `smoke-no-steps`	completed	3	0ms	ok
✅ `smoke-one-step`	completed	6	0ms	ok
✅ `hook-at-step-started`	completed	12	0ms	ok
✅ `hook-at-step-completed`	completed	12	0ms	ok
✅ `hook-at-hook-created`	completed	12	0ms	ok
✅ `deadline-hook-wins`	completed	7	1.0h	ok
✅ `deadline-expires`	completed	7	1.0h	ok
✅ `step-vs-timer-early-settlement`	completed	8	1.0h	ok
✅ `long-sleep`	completed	11	30.0d	ok
✅ `hook-never-arrives`	stalled	3	0ms	skipped
✅ `step-retries-twice`	completed	10	2.0s	ok
✅ `parallel-steps`	completed	9	0ms	ok
✅ `hook-on-execution-state`	completed	12	0ms	ok
✅ `peek-hook-before-branch`	completed	12	0ms	ok
✅ `peek-hook-after-branch`	completed	12	0ms	ok
✅ `peek-hook-at-registration`	completed	12	0ms	ok
✅ `race-hook-before-probe`	completed	12	0ms	ok
✅ `race-hook-after-probe`	completed	12	0ms	ok
✅ `race-duplicate-delivery`	completed	13	0ms	ok
✅ `attr-hook-before-step`	completed	11	0ms	ok
✅ `attr-hook-after-step`	completed	11	0ms	ok
✅ `attr-from-step-body`	completed	13	0ms	ok
✅ `fork-hook-after-timeout`	completed	14	1.0m	ok
✅ `fork-hook-before-timeout`	completed	14	1.0m	ok
✅ `count-hook-after-timeout`	completed	17	1.0m	ok
✅ `count-hook-before-timeout`	completed	20	1.0m	ok
✅ `stale-read-step-count-fork`	completed	20	1.0m	ok
✅ `stale-read-equal-step-counts`	completed	14	1.0m	ok
✅ `step-vs-step-fork`	completed	12	0ms	ok
✅ `step-vs-step-fork-fenced`	completed	12	0ms	ok
✅ `fence-catches-benign-direction`	completed	12	5ms	ok
✅ `in-flight-before-decision`	completed	17	1.0m	ok
❌ `in-flight-before-decision-counted`	completed	17	1.0m	ok
✅ `in-flight-after-decision`	completed	19	2.0m	ok
✅ `stale-read-step-count-fork-fenced`	completed	20	1.0m	ok
✅ `fork-hook-wins`	completed	13	1.0m	ok
✅ `fork-timeout-wins`	completed	13	1.0m	ok
✅ `unclaimed-payload-under-fork`	completed	17	1.0m	ok
✅ `claimed-payload-under-fork`	completed	17	1.0m	ok
✅ `writers-independent-step-bodies`	completed	12	0ms	ok
✅ `writers-scripted-tempo`	completed	12	0ms	ok
✅ `cancel-mid-step`	cancelled	7	0ms	skipped

Full trace: world-sim.txt

github-actions · 2026-09-24T17:07:06Z

Framework	Flow route	Step reg.	Framework output
hono	252.0 KiB (±0)	94.9 KiB (+39 B)	1.91 MiB (+82 B)
nextjs-turbopack	259.3 KiB (±0)	426 B (±0)	908.3 KiB (+78 B)

About these numbers

Sizes are gzip; parentheses show the change against main.
Flow route and Step reg. gate this job, on raw bytes rather than the gzip shown, at max(2%, 50.0 KiB). Framework output is informational.

03e6e67 · run

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>

vercel · 2026-09-24T17:34:00Z

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated
example-nextjs-workflow-turbopack	Ready	Preview, v0	Sep 25, 2026 3:53am UTC
example-nextjs-workflow-webpack	Ready	Preview, v0	Sep 25, 2026 3:53am UTC
example-workflow	Ready	Preview, v0	Sep 25, 2026 3:53am UTC
workbench-astro-workflow	Ready	Preview, v0	Sep 25, 2026 3:53am UTC
workbench-express-workflow	Ready	Preview, v0	Sep 25, 2026 3:53am UTC
workbench-fastify-workflow	Ready	Preview, v0	Sep 25, 2026 3:53am UTC
workbench-hono-workflow	Ready	Preview, v0	Sep 25, 2026 3:53am UTC
workbench-nestjs-workflow	Ready	Preview, v0	Sep 25, 2026 3:53am UTC
workbench-nitro-workflow	Ready	Preview, v0	Sep 25, 2026 3:53am UTC
workbench-nuxt-workflow	Ready	Preview, v0	Sep 25, 2026 3:53am UTC
workbench-python-workflow	Ready	Preview, v0	Sep 25, 2026 3:53am UTC
workbench-sveltekit-workflow	Ready	Preview, v0	Sep 25, 2026 3:53am UTC
workbench-tanstack-start-workflow	Ready	Preview, v0	Sep 25, 2026 3:53am UTC
workbench-vite-workflow	Ready	Preview, v0	Sep 25, 2026 3:53am UTC
workflow-docs	Ready	Preview, v0	Sep 25, 2026 3:53am UTC
workflow-swc-playground	Ready	Preview, v0	Sep 25, 2026 3:53am UTC
workflow-tarballs	Ready	Preview, v0	Sep 25, 2026 3:53am UTC
workflow-web	Ready	Preview, v0	Sep 25, 2026 3:53am UTC

…mped spec version Address review on #4327: - Raise the capability probe budget from 2s to 10s. healthCheck() returns on the first answer, so only a miss pays; a cold but healthy older-major target (e.g. stable, spec 3) is now read from the probe instead of being stamped with the spec-6 guess it would reject. - Be honest about the miss floor: spec 6 is the lowest a v5 runtime executes, and is not safe for older-major targets. Log a once-per-process warning on a miss, and TODO lowering the floor once #4366 and workflow-server#1044 ship. - Record workflow.run.spec_version / spec_version_source and the probe's latency or error on the start() span. - Cap `wf inspect` replay's probed spec version at the CLI World's. - Tests: cold spec-3 target, end-to-end probe timeout, 'latest' to another deployment and to self, retention error naming the target, compression off below spec 5, explicit-version attribution. - Docs and changeset updates. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Co-Authored-By: Peter Wielander <29887157+VaguelySerious@users.noreply.github.com>

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>

run_started sent mintedSpecVersion() per request, which re-reads WORKFLOW_SEALED_LOG, while the runtime validated the value createWorld captured. Flipping the env in-process could make an executor attest a version it was never validated for, and the backend would raise runs to it. createWorld now reads it once and threads it to event writes (internal APIConfig.mintedSpecVersion); direct callers without it fall back to a fresh read. From the formal-model review on #4406 (E1). Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Co-Authored-By: Peter Wielander <29887157+VaguelySerious@users.noreply.github.com>

alangenfeld · 2026-09-25T15:08:03Z

+export const CAPABILITY_ONLY_SPEC_VERSIONS: ReadonlySet<number> = new Set([
+  SPEC_VERSION_SUPPORTS_CBOR_QUEUE_TRANSPORT,
+  SPEC_VERSION_SUPPORTS_ATTRIBUTES,
+  SPEC_VERSION_SUPPORTS_COMPRESSION,


Non-blocking, doc only. The definition says a capability-only version affects only "the runtime executing it". Compression (5) also affects anyone else who reads the run's payloads: a caller awaiting returnValue, a parent run, or an older dashboard/CLI. It can't be reached today, because only v5 executors attest their version and they're at 6 or above, so raising a run past 5 always crosses slot identity and needs a fresh log. Could the doc say that "capability-only" also means "no reader outside the executor is affected"? Then whoever classifies the next version checks that too.

_{Local agent review (`vercel-ai-gateway/anthropic/claude-opus-5.5`)}

pranaygp

Approving. The attested version is now the World's declared specVersion, fixed when the World is created, and it is sent only on run_started. The capability/structural classification treats any unclassified version as structural, and the test that enumerates every SPEC_VERSION_SUPPORTS_* constant guards that. world 21 and world-vercel events 65 pass locally. The two failing CI jobs (Python Conformance and python-node) are also cancelled on every recent main run. Shipping this before workflow-server#1044 is safe: older servers ignore the unknown meta key.

pranaygp · 2026-09-25T18:15:41Z

Qualifying my approval note: this is safe to ship before vercel/workflow-server#1044 only while #1044 is not partly rolled out with run-spec-version-upgrade on. If the server goes out with the flag on from day one, release this after the server rollout has finished in every region. See https://github.com/vercel/workflow-server/pull/1044 for the rollback caveat.

github-actions · 2026-09-25T18:59:28Z

No backport to stable for 54b48ef (AI decision).

The change depends entirely on main-only APIs: origin/stable's packages/world/src/spec-version.ts stops at spec version 3 and has no mintedSpecVersion(), slot identity, or sealed log, so neither the executorSpecVersion attestation nor the new capability-only/structural classification has anything to attach to there. It is also a fix that by construction only the newer executor can apply — the PR itself notes no v4 release can repair these runs — and it adds new exported API surface (CAPABILITY_ONLY_SPEC_VERSIONS, STRUCTURAL_SPEC_VERSIONS, crossesStructuralSpecVersion) plus a new protocol field.

To override, re-run the Backport to stable workflow manually via workflow_dispatch and paste this commit SHA into the ref input:

54b48ef4deffb8333de5d8c1dff8e8ad8a791efd

Retrigger preview deployments

0fddb58

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>

vercel Bot temporarily deployed to Preview – workflow-docs September 24, 2026 17:34 Inactive

vercel Bot deployed to Preview – workbench-python-workflow September 24, 2026 17:34 View deployment

vercel Bot deployed to Preview – workflow-web September 24, 2026 17:34 View deployment

vercel Bot deployed to Preview – example-nextjs-workflow-turbopack September 24, 2026 17:34 View deployment

vercel Bot deployed to Preview – workflow-tarballs September 24, 2026 17:34 View deployment

vercel Bot deployed to Preview – example-workflow September 24, 2026 17:34 View deployment

vercel Bot deployed to Preview – workbench-nitro-workflow September 24, 2026 17:34 View deployment

vercel Bot deployed to Preview – workbench-express-workflow September 24, 2026 17:35 View deployment

vercel Bot deployed to Preview – workbench-fastify-workflow September 24, 2026 17:35 View deployment

vercel Bot deployed to Preview – workbench-nestjs-workflow September 24, 2026 17:35 View deployment

vercel Bot deployed to Preview – workbench-astro-workflow September 24, 2026 17:35 View deployment

vercel Bot deployed to Preview – workbench-hono-workflow September 24, 2026 17:35 View deployment

vercel Bot deployed to Preview – workbench-vite-workflow September 24, 2026 17:35 View deployment

vercel Bot deployed to Preview – workbench-tanstack-start-workflow September 24, 2026 17:35 View deployment

vercel Bot deployed to Preview – workbench-sveltekit-workflow September 24, 2026 17:35 View deployment

vercel Bot deployed to Preview – example-nextjs-workflow-webpack September 24, 2026 17:35 View deployment

vercel Bot deployed to Preview – workbench-nuxt-workflow September 24, 2026 17:35 View deployment

vercel Bot deployed to Preview – workflow-swc-playground September 24, 2026 17:37 View deployment

VaguelySerious mentioned this pull request Sep 24, 2026

fix(core): stamp cross-deployment runs with the target deployment's spec version #4327

Merged

[world-vercel] Describe executorSpecVersion as a general spec raise

49eda6f

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>

vercel Bot deployed to Preview – workbench-python-workflow September 24, 2026 20:48 View deployment

vercel Bot deployed to Preview – workflow-tarballs September 25, 2026 00:47 View deployment

vercel Bot deployed to Preview – example-workflow September 25, 2026 00:47 View deployment

vercel Bot deployed to Preview – example-nextjs-workflow-turbopack September 25, 2026 00:47 View deployment

vercel Bot deployed to Preview – workbench-sveltekit-workflow September 25, 2026 00:47 View deployment

vercel Bot deployed to Preview – example-nextjs-workflow-webpack September 25, 2026 00:48 View deployment

vercel Bot deployed to Preview – workflow-docs September 25, 2026 00:48 View deployment

vercel Bot deployed to Preview – workflow-swc-playground September 25, 2026 00:50 View deployment

Merge branch 'main' into peter/executor-spec-version

d2b8375

vercel Bot deployed to Preview – workbench-python-workflow September 25, 2026 00:58 View deployment

vercel Bot deployed to Preview – workflow-web September 25, 2026 01:00 View deployment

vercel Bot deployed to Preview – workbench-express-workflow September 25, 2026 01:00 View deployment

vercel Bot deployed to Preview – workbench-nitro-workflow September 25, 2026 01:00 View deployment

vercel Bot deployed to Preview – example-workflow September 25, 2026 01:00 View deployment

vercel Bot deployed to Preview – workbench-hono-workflow September 25, 2026 01:00 View deployment

vercel Bot deployed to Preview – workflow-tarballs September 25, 2026 01:00 View deployment

vercel Bot deployed to Preview – example-nextjs-workflow-webpack September 25, 2026 01:00 View deployment

vercel Bot deployed to Preview – workbench-astro-workflow September 25, 2026 01:00 View deployment

vercel Bot deployed to Preview – workbench-tanstack-start-workflow September 25, 2026 01:00 View deployment

vercel Bot deployed to Preview – workbench-fastify-workflow September 25, 2026 01:00 View deployment

pranaygp mentioned this pull request Sep 25, 2026

test(formal): TLA+/Lean models of cross-deployment spec-version stamping #4406

Open

VaguelySerious mentioned this pull request Sep 25, 2026

Refuse to replay a run stamped above core's SPEC_VERSION_MAX_SUPPORTED on every World #4407

Open

alangenfeld approved these changes Sep 25, 2026

View reviewed changes

pranaygp approved these changes Sep 25, 2026

View reviewed changes

This was referenced Sep 25, 2026

Version Packages (beta) #4345

Merged

Version Packages (beta) ben-vargas/workflow#104

Open

Version Packages (beta) Jokacar10/jkworkflow#548

Open

pranaygp mentioned this pull request Sep 28, 2026

test(e2e): expect a takeover when the World raises a force-claim victim to its executor's spec version #4462

Merged

Navigation Menu

[world-vercel] Attest the executor's spec version on run_started - #4366

[world-vercel] Attest the executor's spec version on run_started#4366
VaguelySerious merged 7 commits into
mainfrom
peter/executor-spec-version

VaguelySerious commented Sep 24, 2026 •

edited

Loading

Uh oh!

changeset-bot Bot commented Sep 24, 2026 •

edited

Loading

Uh oh!

github-actions Bot commented Sep 24, 2026 •

edited

Loading

Uh oh!

github-actions Bot commented Sep 24, 2026 •

edited

Loading

Uh oh!

github-actions Bot commented Sep 24, 2026 •

edited

Loading

Uh oh!

github-actions Bot commented Sep 24, 2026 •

edited

Loading

Uh oh!

vercel Bot commented Sep 24, 2026 •

edited

Loading

Uh oh!

alangenfeld Sep 25, 2026

Uh oh!

pranaygp left a comment

Uh oh!

pranaygp commented Sep 25, 2026

Uh oh!

github-actions Bot commented Sep 25, 2026

Uh oh!

This branch was successfully deployed

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Conversation

VaguelySerious commented Sep 24, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Problem

Change

Tests

Uh oh!

changeset-bot Bot commented Sep 24, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

🦋 Changeset detected

Uh oh!

github-actions Bot commented Sep 24, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

🧪 E2E Test Results

❌ Failed E2E Tests

⚠️ Flaky E2E Tests (passed on retry)

🛠 Infra Events (absorbed by the harness)

E2E Test Summary

Uh oh!

github-actions Bot commented Sep 24, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

📊 Workflow Benchmarks

Uh oh!

github-actions Bot commented Sep 24, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Sim World

Uh oh!

github-actions Bot commented Sep 24, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

vercel Bot commented Sep 24, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

alangenfeld Sep 25, 2026

Choose a reason for hiding this comment

Uh oh!

pranaygp left a comment

Choose a reason for hiding this comment

Uh oh!

pranaygp commented Sep 25, 2026

Uh oh!

github-actions Bot commented Sep 25, 2026

Uh oh!

This branch was successfully deployed

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

VaguelySerious commented Sep 24, 2026 •

edited

Loading

changeset-bot Bot commented Sep 24, 2026 •

edited

Loading

github-actions Bot commented Sep 24, 2026 •

edited

Loading

github-actions Bot commented Sep 24, 2026 •

edited

Loading

github-actions Bot commented Sep 24, 2026 •

edited

Loading

github-actions Bot commented Sep 24, 2026 •

edited

Loading

vercel Bot commented Sep 24, 2026 •

edited

Loading