fix(do): propagate the inbound sampled bit instead of hardcoding it by prisis · Pull Request #191 · anolilab/lunora

prisis · 2026-07-25T19:14:05Z

What

createTracedFetch set the sampled flag on every outbound traceparent to a
literal true:

request.headers.set("traceparent", buildTraceparent(anchor.traceId, spanId, true));

parseTraceparent already reads the inbound bit, but resolveTraceAnchor
discarded it — so a trace an upstream had sampled out was announced to every
callee as sampled in.

Why it matters

The callee records and exports spans for a trace nobody kept. The collector gets
a fragment with no root and no siblings, and the sampling decision the upstream
made is silently not the decision the system applied — the exact failure the
flag exists to prevent.

The runtime's own propagation (otel-trace.ts:197) already threaded
trace.sampled correctly. Only the shard path diverged, so worker-originated
and shard-originated hops of the same trace disagreed.

The fix

The verdict rides on the TraceAnchor, resolved once per dispatch, so every
ctx.fetch of one dispatch propagates one consistent answer rather than each
deriving its own.

A dispatch with no inbound traceparent — a subscription re-run, a
server-initiated call — still samples: it is the trace root, so there is no
upstream verdict to inherit.

Tests

traced-fetch.test.ts — both directions through ctx.fetch
trace-context.test.ts (new) — the anchor inheriting, defaulting, and
rejecting a malformed inbound header

Both halves are mutation-verified: restoring the hardcoded true, or making the
anchor ignore the inbound verdict, each fail a test. @lunora/do 1245 passing,
@lunora/runtime 741 passing.

Scope

Deliberately one fix. This was extracted by content from
feat/platform-abstraction-layer (#190); everything else OTel-related on that
branch either already exists on alpha (resource detection, tail-sampler
reporting, fail-closed redaction, the createOtelTracer bridge as
@lunora/server/otel) or is structural divergence with no behavioural benefit.
This was the only genuine bug that alpha still carries.

🤖 Generated with Claude Code

Summary by CodeRabbit

Bug Fixes
- Improved distributed tracing to preserve the inbound trace’s sampling decision across outbound requests.
- Prevented downstream spans from being recorded when the originating trace was not sampled.
- Reduced orphaned telemetry data and improved trace consistency across connected services.

`createTracedFetch` set every outbound `traceparent`'s sampled flag to `true`. `parseTraceparent` already reads the inbound bit and `resolveTraceAnchor` threw it away, so a trace an upstream had sampled OUT was announced to every callee as sampled in. The callee then records and exports spans for a trace nobody kept. The collector receives a fragment with no root and no siblings, and the sampling decision the upstream made is silently not the one the system applied — which is the exact failure the flag exists to prevent. The verdict now rides on the `TraceAnchor`, resolved once per dispatch, so every `ctx.fetch` of one dispatch propagates one consistent answer rather than each deriving its own. A dispatch with no inbound `traceparent` — a subscription re-run, a server-initiated call — still samples: it is the trace root, so there is no upstream verdict to inherit. The runtime's own propagation (`otel-trace.ts`) already threaded `trace.sampled` correctly; only the shard path diverged. Tests: both directions through `ctx.fetch`, plus `resolveTraceAnchor` inheriting, defaulting, and rejecting a malformed inbound header. Both halves are mutation-verified — restoring the hardcoded `true`, or making the anchor ignore the inbound verdict, each fail a test. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

netlify · 2026-07-25T19:14:09Z

✅ Deploy Preview for lunorash ready!

Name	Link
🔨 Latest commit	`1441bf4`
🔍 Latest deploy log	https://app.netlify.com/projects/lunorash/deploys/6a6510769ea6aa00089671b7
😎 Deploy Preview	https://deploy-preview-191--lunorash.netlify.app
📱 Preview on mobile	Toggle QR Code... Use your smartphone camera to open QR code link.
🤖 Make changes	Run an agent on this branch

To edit notification comments on pull requests, go to your Netlify project configuration.

coderabbitai · 2026-07-25T19:14:23Z

Important

Review skipped

Review was skipped due to path filters

⛔ Files ignored due to path filters (1)

api-snapshots/do.api.md is excluded by none and included by none

CodeRabbit blocks several paths by default. You can override this behavior by explicitly including those paths in the path filters. For example, including **/dist/** will override the default block on the dist directory, by removing the pattern from both the lists.

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro Plus

Run ID: 24036c7f-a467-4d6e-82c8-5c1fec8b2fff

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

🔍 Trigger review

Walkthrough

Tracing now preserves the inbound W3C traceparent sampling verdict in TraceAnchor and applies it to outbound ctx.fetch propagation. New trace contexts default to sampled.

Changes

Traceparent Sampling Propagation

Layer / File(s)	Summary
Resolve sampled trace context `packages/do/src/trace-context.ts`	`resolveTraceAnchor` returns `sampled`, preserving the inbound verdict or defaulting to `true` when no inbound traceparent exists.
Propagate sampled anchor state `packages/do/src/context-telemetry.ts`	`TraceAnchor` stores the optional sampling flag, and outbound `traceparent` headers use it instead of always setting sampling to `true`.

Estimated code review effort: 2 (Simple) | ~10 minutes

Possibly related PRs

anolilab/lunora#166: Implements related traceparent sampled-flag propagation through RPC dispatch and span export.
anolilab/lunora#189: Introduces earlier createTracedFetch and traceparent propagation work in the same tracing pipeline.
anolilab/lunora#143: Wires forwarded traceparent headers through worker, ShardDO, and container tracing.

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Description check	⚠️ Warning	The description is detailed, but it misses required template sections and the exact CLA statement.	Add the required Summary, Linked issues, Test plan, Checklist, Notes for reviewers, and the exact CLA statement from the template.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title is concise, conventional-commit style, and accurately describes the sampled-bit propagation change.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

✨ Finishing Touches

🧪 Generate unit tests (beta)

Create PR with unit tests
Commit unit tests in branch fix/trace-sampled-propagation

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

_{Comment @coderabbitai help to get the list of available commands.}

github-actions · 2026-07-25T19:14:24Z

Thank you for following the naming conventions! 🙏

github-actions · 2026-07-25T19:14:25Z

Thank you for confirming the Contributor License Agreement! 🙏

codspeed-hq · 2026-07-25T19:33:26Z

Merging this PR will regress 1 benchmark

⚠️

Different runtime environments detected

Some benchmarks with significant performance changes were compared across different runtime environments,
which may affect the accuracy of the results.

Open the report in CodSpeed to investigate

⚡ 2 improved benchmarks
❌ 1 regressed benchmark
✅ 236 untouched benchmarks
⏩ 2 skipped benchmarks¹

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Benchmark	`BASE`	`HEAD`	Efficiency
❌	`in-batch: single IN(...) query + id->doc re-projection`	816.3 µs	948.4 µs	-13.93%
⚡	`low fanout — 5 sockets × 1 sub each, table match`	73.2 µs	59.1 µs	+23.94%
⚡	`flat 3 primitives (the notify.send attribute shape)`	62.2 µs	55.5 µs	+12.1%

Tip

Investigate this regression by commenting @codspeedbot fix this regression on this PR, or directly use the CodSpeed MCP with your agent.

_{Comparing fix/trace-sampled-propagation (1441bf4) with alpha (49b10ad)²}

2 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩
No successful run was found on alpha (b66b371) during the generation of this report, so 49b10ad was used instead as the comparison base. There might be some changes unrelated to this pull request in this report. ↩

`TraceAnchor` is part of `@lunora/do`'s public surface, so adding the `sampled` verdict drifts the committed snapshot. Optional, so it is additive: an existing consumer constructing an anchor still compiles, and a consumer reading one gets `boolean | undefined` — which is why the propagation path treats an absent verdict as "sampled" rather than assuming the field is always present. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

prisis temporarily deployed to benchmarks July 25, 2026 19:14 — with GitHub Actions Inactive

github-actions Bot added the package: do label Jul 25, 2026

prisis temporarily deployed to benchmarks July 25, 2026 19:37 — with GitHub Actions Inactive

prisis merged commit 4ebb86a into alpha Jul 25, 2026
41 of 42 checks passed

prisis deleted the fix/trace-sampled-propagation branch July 25, 2026 19:56

This was referenced Jul 26, 2026

feat(platform): platform abstraction layer (plan 114) #190

Merged

feat(observability)!: sampling coherence, health, log-archive, DO telemetry + OTLP (Wave 15: 179,180,181,186,190) #198

Merged

Navigation Menu

Uh oh!

Uh oh!

fix(do): propagate the inbound sampled bit instead of hardcoding it - #191

fix(do): propagate the inbound sampled bit instead of hardcoding it#191
prisis merged 2 commits into
alphafrom
fix/trace-sampled-propagation

prisis commented Jul 25, 2026 •

edited by coderabbitai Bot

Loading

Uh oh!

netlify Bot commented Jul 25, 2026 •

edited

Loading

Uh oh!

coderabbitai Bot commented Jul 25, 2026 •

edited

Loading

Review skipped

❌ Failed checks (1 warning)

Uh oh!

github-actions Bot commented Jul 25, 2026

Uh oh!

github-actions Bot commented Jul 25, 2026

Uh oh!

codspeed-hq Bot commented Jul 25, 2026 •

edited

Loading

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

Uh oh!

Uh oh!

Conversation

prisis commented Jul 25, 2026 • edited by coderabbitai Bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

What

Why it matters

The fix

Tests

Scope

Summary by CodeRabbit

Uh oh!

netlify Bot commented Jul 25, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

✅ Deploy Preview for lunorash ready!

Uh oh!

coderabbitai Bot commented Jul 25, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Review skipped

Walkthrough

Changes

❌ Failed checks (1 warning)

Uh oh!

github-actions Bot commented Jul 25, 2026

Uh oh!

github-actions Bot commented Jul 25, 2026

Uh oh!

codspeed-hq Bot commented Jul 25, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Merging this PR will regress 1 benchmark

Performance Changes

Footnotes

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

prisis commented Jul 25, 2026 •

edited by coderabbitai Bot

Loading

netlify Bot commented Jul 25, 2026 •

edited

Loading

coderabbitai Bot commented Jul 25, 2026 •

edited

Loading

codspeed-hq Bot commented Jul 25, 2026 •

edited

Loading