feat(llmobs): add internal FlushSync function by rarguelloF · Pull Request #4695 · DataDog/dd-trace-go

rarguelloF · 2026-04-24T16:32:11Z

What does this PR do?

Adds a FlushSync function that does the same as the existing Flush but does not return until payloads have been actually sent.

This is a requirement for #4512

Motivation

Reviewer's Checklist

Changed code has unit tests for its functionality at or near 100% coverage.
System-Tests covering this feature have been added and enabled with the va.b.c-dev version tag.
There is a benchmark for any new code, or changes to existing code.
If this interacts with the agent in a new way, a system test has been added.
New code is free of linting errors. You can check this by running make lint locally.
New code doesn't break existing tests. You can check this by running make test locally.
Add an appropriate team label so this PR gets put in the right place for the release notes.
All generated files are up to date. You can check this by running make generate locally.
Non-trivial go.mod changes, e.g. adding new modules, are reviewed by @DataDog/dd-trace-go-guild. Make sure all nested modules are up to date by running make fix-modules locally.

Unsure? Have a question? Request a review!

codecov · 2026-04-24T16:37:39Z

Codecov Report

❌ Patch coverage is 89.28571% with 3 lines in your changes missing coverage. Please review.
✅ Project coverage is 62.09%. Comparing base (ffc33dc) to head (6812c31).

Files with missing lines	Patch %	Lines
internal/llmobs/llmobs.go	89.28%	2 Missing and 1 partial ⚠️

Additional details and impacted files

Files with missing lines	Coverage Δ
internal/llmobs/llmobs.go	`80.07% <89.28%> (-4.80%)`	⬇️

... and 277 files with indirect coverage changes

🚀 New features to boost your workflow:

❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

datadog-datadog-prod-us1-2 · 2026-04-24T16:37:49Z

🎉 All green!

❄️ No new flaky tests detected
🧪 All tests passed

🎯 Code Coverage (details)
• Patch Coverage: 92.31%
• Overall Coverage: 61.37% (-0.04%)

_{This comment will be updated automatically if new data arrives.

🔗 Commit SHA: 6812c31 | Docs | Datadog PR Page | Give us feedback!}

darccio · 2026-04-24T17:31:21Z

@codex review

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 66532d582f

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

chatgpt-codex-connector · 2026-04-24T17:36:29Z

+	case l.flushNowCh <- done:
+		<-done


Return from FlushSync when the worker is stopped

When FlushSync is called on an LLMObs pointer after Stop (or races with Stop), stopCh is closed but the send to the buffered flushNowCh can also be selected when the buffer is empty. In that branch there is no worker left to receive the done channel, so the following <-done blocks forever instead of returning; check the stopped state before enqueueing the sync flush.

Useful? React with 👍 / 👎.

pr-commenter · 2026-04-24T18:11:38Z

Benchmarks

Benchmark execution time: 2026-05-05 13:58:43

Comparing candidate commit 6812c31 in PR branch rarguelloF/llmobs-flush-sync with baseline commit ffc33dc in branch main.

Found 0 performance improvements and 0 performance regressions! Performance is the same for 176 metrics, 0 unstable metrics.

Explanation

This is an A/B test comparing a candidate commit's performance against that of a baseline commit. Performance changes are noted in the tables below as:

🟩 = significantly better candidate vs. baseline
🟥 = significantly worse candidate vs. baseline

We compute a confidence interval (CI) over the relative difference of means between metrics from the candidate and baseline commits, considering the baseline as the reference.

If the CI is entirely outside the configured SIGNIFICANT_IMPACT_THRESHOLD (or the deprecated UNCONFIDENCE_THRESHOLD), the change is considered significant.

Feel free to reach out to #apm-benchmarking-platform on Slack if you have any questions.

More details about the CI and significant changes

You can imagine this CI as a range of values that is likely to contain the true difference of means between the candidate and baseline commits.

CIs of the difference of means are often centered around 0%, because often changes are not that big:

---------------------------------(------|---^--------)-------------------------------->
                              -0.6%    0%  0.3%     +1.2%
                                 |          |        |
         lower bound of the CI --'          |        |
sample mean (center of the CI) -------------'        |
         upper bound of the CI ----------------------'

As described above, a change is considered significant if the CI is entirely outside the configured SIGNIFICANT_IMPACT_THRESHOLD (or the deprecated UNCONFIDENCE_THRESHOLD).

For instance, for an execution time metric, this confidence interval indicates a significantly worse performance:

----------------------------------------|---------|---(---------^---------)---------->
                                       0%        1%  1.3%      2.2%      3.1%
                                                  |   |         |         |
       significant impact threshold --------------'   |         |         |
                      lower bound of CI --------------'         |         |
       sample mean (center of the CI) --------------------------'         |
                      upper bound of CI ----------------------------------'

FlushSync correctly blocks until batchSend completes, but the testtracer mock pushes parsed spans through an unbuffered payloadChan that is drained by collectPayloads in a separate goroutine. The unbuffered send synchronizes only the value transfer; the recorder's lock+append can lag behind FlushSync's return, leaving SentPayloads empty. Use the existing WaitForLLMObsSpans helper, which is the pattern used by the other tests in this file that read recorded payloads. Co-Authored-By: Claude Opus 4.7 (1M context) <[email protected]>

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 8c6b506335

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

kakkoyun

LGTM.

I have raised only two minor issues.

kakkoyun · 2026-04-29T12:56:39Z

 				params := l.clearBuffersNonLocked()
 				l.wg.Go(func() {
 					l.batchSend(params)
+					if done != nil {


nit: We can call this wrap with defer, in case batchSend panics.

Without recover() the defer is not gonna help much in that case here as the program would still crash in case of panic. batchSend is not expected to panic, so if it does I would prefer not to silence it here (in general I think that would look like the try catch pattern from other languages that I would like to avoid if possible 🙏 )

Even without a recover, defers would be executed. It's a guaranteed way to making sure the channel is closed. So it is not about being silent but about making sure we unblock other goroutine wait on this channel if something goes wrong with batchSend. We can have it defensively or as you said if we don't expect batchSend call-chain to never fail, we can accept the state as is.

darccio · 2026-05-04T15:03:30Z

@rarguelloF @kakkoyun Should we admin merge this?

feat(llmobs): add internal FlushSync function

66532d5

darccio mentioned this pull request Apr 24, 2026

feat(internal/llmobs): add FlushSync for synchronous flush in tests #4689

Closed

4 tasks

chatgpt-codex-connector Bot reviewed Apr 24, 2026

View reviewed changes

kakkoyun self-requested a review April 27, 2026 09:06

rarguelloF marked this pull request as ready for review April 28, 2026 07:45

rarguelloF requested review from a team as code owners April 28, 2026 07:45

Merge branch 'main' into rarguelloF/llmobs-flush-sync

4f14984

chatgpt-codex-connector Bot reviewed Apr 28, 2026

View reviewed changes

Comment thread internal/llmobs/llmobs.go Outdated

darccio approved these changes Apr 28, 2026

View reviewed changes

Merge branch 'main' into rarguelloF/llmobs-flush-sync

ae62577

kakkoyun reviewed Apr 29, 2026

View reviewed changes

rarguelloF and others added 3 commits April 30, 2026 12:58

fix review issues

c81204c

fix lint and review comments

2c6c455

Merge branch 'main' into rarguelloF/llmobs-flush-sync

ea94778

Merge branch 'main' into rarguelloF/llmobs-flush-sync

6812c31

darccio merged commit e2961fd into main May 5, 2026
170 of 188 checks passed

darccio deleted the rarguelloF/llmobs-flush-sync branch May 5, 2026 12:30

Uh oh!

Conversation

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

Uh oh!

What does this PR do?

Motivation

Reviewer's Checklist

Uh oh!

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

Uh oh!

Codecov Report

Uh oh!

datadog-datadog-prod-us1-2 Bot commented Apr 24, 2026 • edited by datadog-official Bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

darccio commented Apr 24, 2026

Uh oh!

chatgpt-codex-connector Bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

chatgpt-codex-connector Bot Apr 24, 2026

Choose a reason for hiding this comment

Uh oh!

Uh oh!

pr-commenter Bot commented Apr 24, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Benchmarks

Explanation

More details about the CI and significant changes

Uh oh!

chatgpt-codex-connector Bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

Uh oh!

kakkoyun left a comment

Choose a reason for hiding this comment

Uh oh!

kakkoyun Apr 29, 2026

Choose a reason for hiding this comment

Uh oh!

rarguelloF Apr 30, 2026

Choose a reason for hiding this comment

Uh oh!

kakkoyun Apr 30, 2026

Choose a reason for hiding this comment

Uh oh!

Uh oh!

darccio commented May 4, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

rarguelloF commented Apr 24, 2026 •

edited

Loading

codecov Bot commented Apr 24, 2026 •

edited

Loading

datadog-datadog-prod-us1-2 Bot commented Apr 24, 2026 •

edited by datadog-official Bot

Loading

pr-commenter Bot commented Apr 24, 2026 •

edited

Loading