v5.124.0 proposal

[dd-octo-sts]

Features

• AI Guard: Sensitive data redaction #9833
• Feature Flags: Send serial_id on exposure events #9929
• General: Support DD_TRACE_HTTP_CLIENT_ERROR_STATUSES #9910
• LLM Observability: Add experiment multirun concurrency #9686
• LLM Observability: Add records to datasets #10025
• LLM Observability: Support experiment project overrides #9849
• LLM Observability: Support opt-out for all integrations #9737
• Profiling: Use separate column field in pprof output #9948
• Test Optimization: Submit Bunyan logs agentlessly #9906
• Test Optimization: Submit Pino logs agentlessly #9978
• Test Optimization: Support logger submission from Jest static ESM imports #10020
• Test Optimization: Support Pino and Bunyan in Jest #10019
• tracer: Add recordException API #9863

Fixes

• AppSec: Clear stale blocking actions on RC ASM modify #9302
• Dependencies: Update native library dependencies #9924
• LLM Observability: Preserve in-flight dataset mutations #9903
• mariadb: Support 3.5.3 CommonJS bundles #9736
• Profiling: Reapply custom label keys when starting the profiler #9966
• Profiling: Resolve web tags for descendants cached before promotion #9805
• Serverless: Flush configured telemetry on Vercel #9735
• Test Optimization: Bound agentless final flush #9962
• Test Optimization: Flush logs before Playwright worker exit #9908
• Test Optimization: Flush logs before Vitest worker exit #9963
• Test Optimization: Flush logs before WebdriverIO worker exit #9964
• Test Optimization: Preserve late failure screenshots #9896
• Test Optimization: Release Playwright worker on flush failure #9983

Performance

• AppSec: Avoid loading rewriter when disabled #9942
• General: Lazy-load inactive startup modules #10022
• Test Optimization: Avoid loading startup modules outside test mode #10021
• Test Optimization: Calculate Jest coverage in one pass #9941

Internal (CI, Testing, Benchmarking)

• aws-sdk: Wait for EventBridge cold-start traces #9847
• ci: Add test-optimization files to CODEOWNERS #9974
• codeowners: Assign AGENTS.md owners #10030
• Dependencies: Bump @datadog/wasm-js-rewriter to 5.0.4 #9919
• Dynamic Instrumentation: Await probe diagnostics #9869
• Dynamic Instrumentation: Await probe runtime completion #9870
• eslint: Detect unnecessary array joins #9925
• General: Add dd-octo-sts as dependency code owner #9923
• General: Make skipped test intent explicit #9868
• General: Narrow the text map propagator surface #9636
• General: Reject Yarn build warnings from native dependencies #9494
• General: Require fake-agent teardown #9865
• General: Tolerate exporter socket resets #9928
• integration: Make suite completion explicit #9871
• LLM Observability: Await evaluation metric requests #9957
• mongodb: Await traced operations #9822
• pr-title: Reduce API calls and check out current revision #9823
• release: Warn about overdue releases #9955
• Serverless: Revert "Temporarily allow benchmark failures" #9992
• Serverless: Temporarily allow benchmark failures #9965
• skills: Replace flaky-test fixer workflow #9866
• Test Optimization: Cover Bunyan log submission in Mocha and Cucumber #9920
• Test Optimization: Cover Pino and Winston logs in WebdriverIO #10016
• Test Optimization: Cover Pino log submission in Playwright #9982
• Test Optimization: Cover Pino logs in Mocha, Cucumber, and Vitest #9981
• Test Optimization: Share Winston log submission #9922
• Test Optimization: Use dispatched workflow run id #9947
• undici: Await request completion #9874

Contributors

[dd-octo-sts]

Overall package size

Self size: 8.42 MB
Deduped: 9.09 MB
No deduping: 9.09 MB

Dependency sizes

| name | version | self size | total size | |------|---------|-----------|------------| | import-in-the-middle | 3.3.3 | 125.43 kB | 445.14 kB | | opentracing | 0.14.7 | 194.81 kB | 194.81 kB | | dc-polyfill | 0.1.11 | 25.74 kB | 25.74 kB |

_{🤖 This report was automatically generated by heaviest-objects-in-the-universe}

[datadog-prod-us1-5]

🎉 All green!

🧪 All tests passed
❄️ No new flaky tests detected

🎯 Code Coverage (details)
• Patch Coverage: 99.37%
• Overall Coverage: 98.52%

_{This comment will be updated automatically if new data arrives.
🔗 Commit SHA: 7c420ec | Docs | View more details | Give us feedback!}

[codecov]

Codecov Report

❌ Patch coverage is 99.37370% with 30 lines in your changes missing coverage. Please review.
✅ Project coverage is 98.52%. Comparing base (6b35e7d) to head (7c420ec).
⚠️ Report is 2188 commits behind head on v5.x.

Files with missing lines	Patch %	Lines
...ges/datadog-instrumentations/src/mariadb-bundle.js	98.68%	21 Missing ⚠️
packages/datadog-instrumentations/src/jest.js	97.14%	4 Missing ⚠️
...-trace/src/runtime_metrics/otlp_runtime_metrics.js	40.00%	3 Missing ⚠️
...instrumentations/src/vitest-main-no-worker-init.js	85.71%	1 Missing ⚠️
.../src/opentelemetry/otlp/otlp_http_exporter_base.js	98.36%	1 Missing ⚠️

Additional details and impacted files

@@             Coverage Diff             @@
##             v5.x    #9938       +/-   ##
===========================================
+ Coverage   83.19%   98.52%   +15.32%     
===========================================
  Files         476      990      +514     
  Lines       20153   149308   +129155     
  Branches        0    12769    +12769     
===========================================
+ Hits        16766   147102   +130336     
+ Misses       3387     2206     -1181     

Flag	Coverage Δ
aiguard	62.40% <44.10%> (?)
aiguard-integration	58.95% <37.09%> (?)
apm-bucket-0	61.83% <31.77%> (?)
apm-bucket-1	67.42% <48.47%> (?)
apm-bucket-2	66.98% <48.61%> (?)
apm-bucket-3	63.65% <42.56%> (?)
apm-capabilities-tracing	62.36% <68.62%> (?)
apm-integrations-aerospike	59.64% <30.07%> (?)
apm-integrations-confluentinc-kafka-javascript	65.96% <43.78%> (?)
apm-integrations-couchbase	60.19% <31.77%> (?)
apm-integrations-http	65.66% <48.47%> (?)
apm-integrations-kafkajs	66.56% <44.46%> (?)
apm-integrations-next	63.16% <44.34%> (?)
apm-integrations-prisma	61.27% <37.79%> (?)
appsec	76.48% <48.82%> (?)
appsec-express_fastify_graphql	72.94% <47.70%> (?)
appsec-integration	49.19% <38.68%> (?)
appsec-kafka_ldapjs_lodash	66.94% <47.81%> (?)
appsec-mongodb-core_mongoose_mysql	70.27% <48.65%> (?)
appsec-next	56.52% <40.10%> (?)
appsec-node-serialize_passport_postgres	69.72% <47.98%> (?)
appsec-sourcing_stripe_template	68.10% <47.94%> (?)
debugger	68.82% <42.98%> (?)
instrumentations-bucket-0	54.96% <31.77%> (?)
instrumentations-bucket-1	63.50% <42.77%> (?)
instrumentations-bucket-10	64.75% <42.72%> (?)
instrumentations-bucket-11	65.41% <42.76%> (?)
instrumentations-bucket-12	54.83% <31.77%> (?)
instrumentations-bucket-13	55.14% <35.32%> (?)
instrumentations-bucket-14	54.96% <31.77%> (?)
instrumentations-bucket-2	56.06% <35.32%> (?)
instrumentations-bucket-3	56.88% <35.32%> (?)
instrumentations-bucket-4	62.95% <42.77%> (?)
instrumentations-bucket-5	49.20% <30.96%> (?)
instrumentations-bucket-6	64.84% <47.79%> (?)
instrumentations-bucket-7	55.10% <31.77%> (?)
instrumentations-bucket-8	62.33% <42.85%> (?)
instrumentations-bucket-9	57.72% <31.00%> (?)
instrumentations-instrumentation-couchbase	53.50% <29.73%> (?)
instrumentations-integration-esbuild	34.41% <36.60%> (?)
llmobs-ai_anthropic_bedrock	66.23% <42.15%> (?)
llmobs-bucket-1	64.36% <41.62%> (?)
llmobs-openai	66.21% <42.00%> (?)
llmobs-openai-agents_vertex-ai	63.40% <39.58%> (?)
llmobs-sdk	75.63% <81.99%> (?)
openfeature	59.53% <35.40%> (?)
openfeature-unit	57.34% <35.67%> (?)
platform-core_esbuild_instrumentations-misc	40.50% <36.43%> (?)
platform-integration	64.31% <41.79%> (?)
platform-shimmer_unit-guardrails_webpack	38.06% <34.90%> (?)
plugins-browser-bunyan_bullmq_cassandra	65.63% <47.74%> (?)
plugins-bucket-0	60.26% <38.11%> (?)
plugins-bucket-1	57.47% <35.79%> (?)
plugins-bucket-11	66.14% <47.85%> (?)
plugins-bucket-18	65.66% <47.32%> (?)
plugins-bucket-19	64.24% <44.86%> (?)
plugins-bucket-20	66.13% <42.77%> (?)
plugins-bucket-4	60.42% <33.12%> (?)
plugins-cookie_cookie-parser_crypto	54.50% <31.77%> (?)
plugins-fastify_fetch_fs	64.82% <45.24%> (?)
plugins-generic-pool_google-cloud-pubsub_grpc	68.42% <48.40%> (?)
plugins-handlebars_hapi_hono	62.84% <44.34%> (?)
plugins-ioredis_knex_langgraph	60.91% <31.77%> (?)
plugins-ldapjs_light-my-request_limitd-client	62.56% <42.77%> (?)
plugins-lodash_mariadb_memcached	63.20% <66.66%> (?)
plugins-moleculer_mongodb_mongodb-core	65.50% <47.93%> (?)
plugins-mongoose_multer_mysql	63.30% <35.36%> (?)
plugins-mysql2_nats_node-serialize	65.48% <43.82%> (?)
plugins-opensearch_passport-http_pino	63.31% <38.65%> (?)
plugins-postgres_process_pug	62.40% <31.81%> (?)
plugins-redis_router_sequelize	66.08% <47.72%> (?)
plugins-test-and-upstream-rhea_undici_url	65.44% <47.87%> (?)
plugins-valkey_vm_winston	61.70% <35.58%> (?)
plugins-ws	63.71% <47.93%> (?)
profiling	65.75% <46.24%> (?)
serverless-aws-sdk-aws-sdk	54.71% <37.00%> (?)
serverless-aws-sdk-base-inject-field	54.08% <31.77%> (?)
serverless-aws-sdk-bedrockruntime	57.41% <35.40%> (?)
serverless-aws-sdk-client	59.25% <38.03%> (?)
serverless-aws-sdk-dynamodb	58.36% <35.40%> (?)
serverless-aws-sdk-eventbridge	56.96% <45.84%> (?)
serverless-aws-sdk-kinesis	62.45% <47.39%> (?)
serverless-aws-sdk-lambda	60.36% <47.39%> (?)
serverless-aws-sdk-s3	58.43% <35.40%> (?)
serverless-aws-sdk-serverless-peer-service	63.03% <38.11%> (?)
serverless-aws-sdk-sns	63.36% <47.39%> (?)
serverless-aws-sdk-sqs	63.83% <48.20%> (?)
serverless-aws-sdk-stepfunctions	58.23% <37.91%> (?)
serverless-aws-sdk-util	54.75% <31.77%> (?)
serverless-bucket-0	56.86% <38.06%> (?)
serverless-bucket-1	62.99% <42.32%> (?)
test-optimization-cucumber	70.41% <50.84%> (?)
test-optimization-cypress	64.31% <41.10%> (?)
test-optimization-jest	71.85% <57.90%> (?)
test-optimization-mocha	71.90% <50.83%> (?)
test-optimization-playwright-playwright-atr	59.37% <37.37%> (?)
test-optimization-playwright-playwright-efd	60.01% <38.01%> (?)
test-optimization-playwright-playwright-final-status	59.70% <37.34%> (?)
test-optimization-playwright-playwright-impacted-tests	59.81% <37.34%> (?)
test-optimization-playwright-playwright-reporting	60.88% <38.89%> (?)
test-optimization-playwright-playwright-test-management	60.89% <38.19%> (?)
test-optimization-playwright-playwright-test-span	59.42% <37.37%> (?)
test-optimization-selenium	58.61% <39.81%> (?)
test-optimization-testopt	61.69% <53.66%> (?)
test-optimization-vitest	72.72% <57.60%> (?)
test-optimization-vitest-browser	58.53% <36.23%> (?)
test-optimization-webdriverio	65.05% <47.93%> (?)

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 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.

[pr-commenter]

Benchmarks

Benchmark execution time: 2026-08-28 15:56:43

Comparing candidate commit 7c420ec in PR branch v5.124.0-proposal with baseline commit ea9005c in branch v5.x.

📊 Benchmarking dashboard

Found 20 performance improvements and 0 performance regressions! Performance is the same for 2278 metrics, 12 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 ----------------------------------'

scenario:net-with-tracer-20

• 🟩 instructions [-56.0M instructions; -32.4M instructions] or [-3.490%; -2.018%]

scenario:propagation-extract-datadog-24

• 🟩 cpu_user_time [-91.638ms; -32.010ms] or [-5.751%; -2.009%]

scenario:runtime-metrics-control-20

• 🟩 cpu_user_time [-15.040ms; -5.353ms] or [-6.371%; -2.268%]
• 🟩 execution_time [-15.200ms; -5.618ms] or [-5.445%; -2.013%]
• 🟩 instructions [-52.4M instructions; -23.9M instructions] or [-6.031%; -2.746%]

scenario:runtime-metrics-control-24

• 🟩 cpu_user_time [-14.953ms; -5.794ms] or [-6.377%; -2.471%]

scenario:runtime-metrics-control-26

• 🟩 cpu_user_time [-17.980ms; -6.220ms] or [-7.686%; -2.659%]
• 🟩 execution_time [-16.609ms; -6.876ms] or [-5.846%; -2.420%]
• 🟩 instructions [-45.5M instructions; -20.9M instructions] or [-5.694%; -2.613%]

scenario:runtime-metrics-with-runtime-metrics-20

• 🟩 cpu_user_time [-15.810ms; -6.061ms] or [-6.447%; -2.472%]
• 🟩 instructions [-55.1M instructions; -24.9M instructions] or [-6.156%; -2.780%]

scenario:runtime-metrics-with-runtime-metrics-26

• 🟩 instructions [-43.8M instructions; -18.0M instructions] or [-5.341%; -2.194%]

scenario:startup-with-tracer-20

• 🟩 cpu_user_time [-13.362ms; -6.922ms] or [-5.591%; -2.896%]
• 🟩 execution_time [-14.230ms; -8.530ms] or [-5.100%; -3.058%]
• 🟩 instructions [-48.1M instructions; -29.5M instructions] or [-5.515%; -3.384%]

scenario:startup-with-tracer-24

• 🟩 cpu_user_time [-12.457ms; -5.823ms] or [-5.182%; -2.422%]
• 🟩 execution_time [-14.236ms; -8.321ms] or [-4.967%; -2.904%]

scenario:startup-with-tracer-26

• 🟩 cpu_user_time [-13.035ms; -5.868ms] or [-5.449%; -2.453%]
• 🟩 execution_time [-13.421ms; -7.963ms] or [-4.677%; -2.775%]
• 🟩 instructions [-40.9M instructions; -24.7M instructions] or [-5.109%; -3.083%]

Unstable benchmarks

These benchmarks have a confidence interval too wide to call a change; treat them as noise rather than signal.

scenario:appsec-iast-with-vulnerability-iast-enabled-default-config-20

• unstable max_rss_usage [-20805.264KB; +20532.464KB] or [-5.445%; +5.374%]

scenario:debugger-line-probe-without-snapshot-24

• unstable max_rss_usage [-14638.686KB; +15922.686KB] or [-5.154%; +5.606%]

scenario:dogstatsd-with-tags-20

• unstable cpu_user_time [-267.558ms; +412.086ms] or [-5.493%; +8.461%]
• unstable execution_time [-260.951ms; +409.514ms] or [-5.276%; +8.280%]
• unstable throughput [-142481.196op/s; +88427.678op/s] or [-8.392%; +5.208%]

scenario:llmobs-encode-unicode-mixed-20

• unstable max_rss_usage [-7.475MB; +0.841MB] or [-10.727%; +1.206%]

scenario:plugin-claude-agent-sdk-compact-stream-scan-24

• unstable cpu_usage_percentage [-7.053%; +4.974%]

scenario:plugin-claude-agent-sdk-compact-stream-scan-26

• unstable cpu_usage_percentage [-1.708%; +9.381%]

scenario:plugin-graphql-long-with-depth-and-collapse-off-20

• unstable max_rss_usage [-23.325MB; +20.796MB] or [-5.966%; +5.320%]

scenario:plugin-graphql-long-with-depth-off-20

• unstable max_rss_usage [-7753.030KB; +7806.173KB] or [-5.950%; +5.991%]

scenario:plugin-graphql-long-with-depth-off-26

• unstable max_rss_usage [-2.007MB; +27.571MB] or [-0.948%; +13.026%]

scenario:test-optimization-large-suite-20

• unstable max_rss_usage [-6.084MB; +2.749MB] or [-7.594%; +3.431%]

[datadog-prod-us1-5]

More details

Source inspection found no concrete defect in the release proposal. The changed instrumentation contracts and Vercel serverless lifecycle keep their expected behavior.

Was this helpful? React 👍 or 👎

_{Open Bits AI session}

_{🤖 Datadog Autotest · Commit 7c420ec · What is Autotest? · @DataDog review to ask questions · Any feedback? Reach out in #autotest}

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 7c420ec7ce

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]

Restore array joining for batched trace encoding

When an agentless payload contains many traces, repeatedly appending each trace to tracesJson more than doubles encoding time because Buffer.from() must flatten the growing concatenation chain. In a fresh Node 24 process with 1,000 single-span traces, seven warmed trials of 500 encodes took 158–166 ms here versus 68–79 ms with the previous array-and-join implementation. This is a regression on the Test Optimization export path, so retain the prior join strategy unless a representative benchmark shows otherwise.

AGENTS.md reference: AGENTS.md:L213-L213

Useful? React with 👍 / 👎.

[chatgpt-codex-connector]

Include the run identity on external evaluation metrics

When submitEvaluationMetrics() is called with the span returned by submitSpan(), this toMetric() invocation omits its new IDs argument, so the metric lacks the run_id and run_iteration tags that were placed on the corresponding span. Internal row and summary metrics now include these tags, but externally submitted evaluations cannot be grouped or filtered by their run consistently; pass through the external experiment's run identity, including any per-span override.

Useful? React with 👍 / 👎.

[chatgpt-codex-connector]

Validate serial IDs as finite integers

When an exposure event contains a numeric but invalid serial_id, such as NaN, Infinity, or a fractional value, this check accepts it even though the intake contract requires an integer. Non-finite values serialize as null, while fractional values remain non-integers, causing the intake to reject the payload and drop otherwise valid exposures in the same batch; only include finite integer serial IDs.

Useful? React with 👍 / 👎.