Uh oh!
There was an error while loading. Please reload this page.
[world-local] Scope untagged startup recovery to untagged runs - #2667
Merged
Conversation
An untagged world (a normal local dev server) shares the default
`.workflow-data` directory with the vitest harness, which writes
tag-suffixed files (`{id}.{tag}.json`). On startup, `reenqueueActiveRuns`
lists active runs and re-enqueues them; `runs.list` enumerates every file
in the directory and keys each by the `runId` inside it, so an untagged
recovery would pick up tagged runs left behind by tests. Those re-enqueued
messages carry no `runInput`, and `readJSONWithFallback` can only resolve
untagged (or own-tag) files, so `run_started` returns no run entity and the
runtime fails with "did not return the run entity" — a 500 storm on the
flow route until the message exhausts its deliveries.
Tagged worlds already scoped recovery via `fileIdFilter: hasTag(...)`; give
the untagged path the symmetric guard with a new `isUntagged` predicate so
each world only recovers files it can actually read back.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>Contributor
🦋 Changeset detectedLatest commit: 41ca2ca The changes in this PR will be included in the next version bump. This PR includes changesets to release 18 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 |
Contributor
📊 Benchmark Results
workflow with no steps💻 Local Development
▲ Production (Vercel)
🔍 Observability: Nitro | Next.js (Turbopack) workflow with 1 step💻 Local Development
▲ Production (Vercel)
🔍 Observability: Nitro | Next.js (Turbopack) workflow with 10 sequential steps💻 Local Development
▲ Production (Vercel)
🔍 Observability: Nitro | Next.js (Turbopack) workflow with 25 sequential steps💻 Local Development
▲ Production (Vercel)
🔍 Observability: Next.js (Turbopack) | Nitro workflow with 50 sequential steps💻 Local Development
▲ Production (Vercel)
🔍 Observability: Nitro | Next.js (Turbopack) Promise.all with 10 concurrent steps💻 Local Development
▲ Production (Vercel)
🔍 Observability: Nitro | Next.js (Turbopack) Promise.all with 25 concurrent steps💻 Local Development
▲ Production (Vercel)
🔍 Observability: Nitro | Next.js (Turbopack) Promise.all with 50 concurrent steps💻 Local Development
▲ Production (Vercel)
🔍 Observability: Nitro | Next.js (Turbopack) Promise.race with 10 concurrent steps💻 Local Development
▲ Production (Vercel)
🔍 Observability: Nitro | Next.js (Turbopack) Promise.race with 25 concurrent steps💻 Local Development
▲ Production (Vercel)
🔍 Observability: Nitro | Next.js (Turbopack) Promise.race with 50 concurrent steps💻 Local Development
▲ Production (Vercel)
🔍 Observability: Nitro | Next.js (Turbopack) workflow with 10 sequential data payload steps (10KB)💻 Local Development
▲ Production (Vercel)
🔍 Observability: Nitro | Next.js (Turbopack) workflow with 25 sequential data payload steps (10KB)💻 Local Development
▲ Production (Vercel)
🔍 Observability: Nitro | Next.js (Turbopack) workflow with 50 sequential data payload steps (10KB)💻 Local Development
▲ Production (Vercel)
🔍 Observability: Nitro | Next.js (Turbopack) workflow with 10 concurrent data payload steps (10KB)💻 Local Development
▲ Production (Vercel)
🔍 Observability: Nitro | Next.js (Turbopack) workflow with 25 concurrent data payload steps (10KB)💻 Local Development
▲ Production (Vercel)
🔍 Observability: Nitro | Next.js (Turbopack) workflow with 50 concurrent data payload steps (10KB)💻 Local Development
▲ Production (Vercel)
🔍 Observability: Nitro | Next.js (Turbopack) Stream Benchmarks(includes TTFB metrics)workflow with stream💻 Local Development
▲ Production (Vercel)
🔍 Observability: Nitro | Next.js (Turbopack) stream pipeline with 5 transform steps (1MB)💻 Local Development
▲ Production (Vercel)
🔍 Observability: Nitro | Next.js (Turbopack) 10 parallel streams (1MB each)💻 Local Development
▲ Production (Vercel)
🔍 Observability: Nitro | Next.js (Turbopack) fan-out fan-in 10 streams (1MB each)💻 Local Development
▲ Production (Vercel)
🔍 Observability: Nitro | Next.js (Turbopack) SummaryFastest Framework by WorldWinner determined by most benchmark wins
Fastest World by FrameworkWinner determined by most benchmark wins
Column Definitions
Worlds:
❌ Some benchmark jobs failed:
Check the workflow run for details. |
Contributor
🧪 E2E Test Results✅ All tests passed Summary
Details by Category✅ ▲ Vercel Production
✅ 💻 Local Development
✅ 📦 Local Production
✅ 🐘 Local Postgres
✅ 🪟 Windows
✅ 📋 Other
|
pranaygp
approved these changes
Jun 26, 2026
VaguelySerious
enabled auto-merge (squash)
June 26, 2026 20:25
Uh oh!
There was an error while loading. Please reload this page.
github-actionsBot
added a commit
that referenced
this pull request
Jun 26, 2026
Signed-off-by: Peter Wielander <mittgfu@gmail.com>
Contributor
Backport PR opened against |
VaguelySerious added a commit
that referenced
this pull request
Jun 26, 2026
#2668) Co-authored-by: Peter Wielander <mittgfu@gmail.com>
pranaygp added a commit
that referenced
this pull request
Jun 26, 2026
…edrive * origin/main: (90 commits) [ai] Fix "missing text part" stream crash by repairing UI message framing (#2537) [ci] Fix release "Premature close" by bounding changelog GraphQL batch size (#2669) [ai] Drop orphan UI chunks after negative-index reconnect (#2082) Fix trace detail panel Input/Output flicker (selection-driven state machine) (#2637) [world-local] Scope untagged startup recovery to untagged runs (#2667) fix(world-postgres): defer loopback worker startup (#2657) docs: deprecate DurableAgent and WorkflowChatTransport, point to AI SDK WorkflowAgent (#2616) [web] Fix HTTP/2 in bundled server build so observability reads work (#2632) [core] Fix turbo-mode step-body writes racing run_started (#2629) fix(world-postgres): rename setup command (#2644) chore: ignore workflow swc caches (#2640) fix(ai): ignore intentional stream aborts (#2635) fix(nitro): reload steps during Vite HMR (#2572) [world] Make @workflow/world own its Zod dependency (#2622) feat(web-shared): extend cn with custom tailwind-merge class groups (#2619) [world-vercel] Enable HTTP/2 for the events API and stream writes (#2573) Remove shadow from off-screen marker indicator button (#2614) Fix discovery of dotted JS workflow imports (#2594) Version Packages (beta) (#2597) Thread occurredAt into web-shared entities (#2608) ... # Conflicts: # packages/core/src/runtime/step-executor.ts # packages/world-vercel/src/http-client.ts
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for freeto join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Problem
Using Workflow SDK locally with
world-local, users can hit a startup error storm:Root cause
createLocalWorlddefaultsrecoverActiveRuns: true, soworld.start()(on every dev-server boot) callsreenqueueActiveRuns, which lists active runs and re-enqueues them with norunInput.runs.listenumerates every*.jsonfile in therunsdirectory and keys each by therunIdinside the file — including tag-suffixed files likewrun_X.vitest-0.json. But reads go throughreadJSONWithFallback, which only resolves{id}.{tag}.json(own tag) then untagged{id}.json.So an untagged world (a normal
pnpm dev) sharing the default.workflow-datadirectory with the vitest harness — which writes tagged files there "so runs are visible to the observability dashboard" — recovers leftover tagged runs it cannot read back. Since the re-enqueued message carries norunInput, the resilient-start path can't recreate the run either, andrun_startedreturns no run entity →RUNTIME_ERROR500, retried until the message exhausts its deliveries (hence "multiple times").Tagged worlds already guarded against this with
fileIdFilter: (fileId) => hasTag(fileId, tag). The untagged path had no symmetric guard.Fix
Give the untagged recovery path the symmetric filter via a new
isUntaggedpredicate, so each world only recovers files it can actually read back:Test
Added
untagged recovery skips tagged runs sharing the data dirtoreenqueue.test.ts: a tagged (vitest-style) active run + an untagged active run share a data dir; an untagged dev boot must recover only the untagged run. Verified the test fails on the old code (the tagged run gets re-enqueued) and passes with the fix. Full@workflow/world-localsuite (412 tests) passes.