chore(benchmarks): update known flaky benchmarks

[igoragoli]

Description

Updates known flaky benchmarks with httppropagationinject-ids_only, iastaspectsospath-ospathbasename_aspect, iastaspectssplit-(rsplit_aspect|rstrip_aspect), telemetryaddmetric-1-count-metric-1-times and tracer-small.

This prevents flaky benchmarks from breaching SLOs and polluting PR comments.

Testing

None.

Risks

None.

Additional Notes

I investigated 49 recent PRs that change code that shouldn't impact benchmarks: docs, release notes, tests, .riot/ lockfiles, CI config.

33 reported performance changes on scenarios outside our known flaky benchmarks. Adding the ones flagged on more than half of the 33:

• httppropagationinject-ids_only — 33/33, median +12.7%
• iastaspectsospath-ospathbasename_aspect — 33/33, median +31.2%
• telemetryaddmetric-1-count-metric-1-times — 33/33, median +16.3%
• tracer-small — 33/33, median +10.0%
• iastaspectssplit-rsplit_aspect — 23/33, median +14.6%

Some scenarios are already in the known flaky benchmarks regex, corroborated by this data:

• span-start -- 29/33
• iastaspects-ljust_noaspect -- 10/33
• iastaspects-casefold_noaspect -- 7/33
• iastaspects-title_noaspect -- 3/33

Flagged on half of the 33 or fewer, left out as weak evidence:

• iastaspects-add_aspect — 15/33
• iastaspects-title_aspect — 14/33
• iastaspects-join_aspect — 11/33
• httppropagationextract-tracecontext_headers — 10/33
• httppropagationextract-wsgi_valid_headers_all — 6/33
• iastaspects-upper_aspect — 4/33
• iastaspects-repr_aspect — 3/33
• iastaspects-modulo_noaspect — 2/33
• iastaspects-rstrip_aspect — 2/33
• iastaspects-stringio_noaspect — 2/33
• iastaspects-stringio_aspect — 2/33
• iastaspects-lstrip_aspect — 2/33
• iastaspects-ljust_aspect — 1/33
• iastaspects-lower_aspect — 1/33
• iastaspects-swapcase_aspect — 1/33
• iastaspects-translate_noaspect — 1/33
• samplingrules-high_match — 1/33

See investigated PRs (49)

- [#19545](https://github.com//pull/19545) — ci: itr rollout configuration - [#19566](https://github.com//pull/19566) — docs: add benchmarking CI documentation - [#19786](https://github.com//pull/19786) — chore: handle potential after-import exceptions - [#19826](https://github.com//pull/19826) — chore: update django latest version to 6.1 _(no benchmark report)_ - [#19828](https://github.com//pull/19828) — chore(tests): bump gevent to 26.8.0 to stop the flaky django IAST shutdown test - [#19831](https://github.com//pull/19831) — test(profiling): try to make uwsgi less flaky _(no benchmark report)_ - [#19835](https://github.com//pull/19835) — ci(hooks): block partially staged Python files from committing unformatted (PROF-15836) - [#19836](https://github.com//pull/19836) — ci: reuse test installs and lock documentation builds - [#19838](https://github.com//pull/19838) — ci(sg): fail commits on double backticks in ddtrace comments/docstrings (PROF-15837) - [#19839](https://github.com//pull/19839) — test(ci): harden Django and pytest subprocess tests - [#19842](https://github.com//pull/19842) — ci(commenter): safely serialize message from PR commenter bot - [#19843](https://github.com//pull/19843) — chore(py-315): raise SSI exclusive max to 3.16 and pre-stage 3.15 injection (PROF-15812) - [#19846](https://github.com//pull/19846) — test(appsec): bound and raise the test server startup budget - [#19847](https://github.com//pull/19847) — test(iast): flaky test_packages_patched pyjwt _(no benchmark report)_ - [#19853](https://github.com//pull/19853) — test(appsec): enable appsec_integrations_fastapi on Python 3.15 - [#19862](https://github.com//pull/19862) — test(opentracing): remove stale Riot environments - [#19866](https://github.com//pull/19866) — ci(appsec): isolate Django server from gevent cleanup _(no benchmark report)_ - [#19871](https://github.com//pull/19871) — test(aiguard): cover multi-turn sensitive data redaction scenarios - [#19872](https://github.com//pull/19872) — chore(kafka): remove stale _dd.svc_src from service override snapshot _(no benchmark report)_ - [#19874](https://github.com//pull/19874) — test(kafka): update service source snapshots [Part 2] _(no benchmark report)_ - [#19878](https://github.com//pull/19878) — refactor(tests): standardize test runner interface _(no benchmark report)_ - [#19879](https://github.com//pull/19879) — test(dramatiq): enable tests for Python 3.14 - [#19880](https://github.com//pull/19880) — ci: withhold cp315 wheels from PyPI and the prerelease index (PROF-15857) - [#19885](https://github.com//pull/19885) — ci: automatically retry os tests - [#19897](https://github.com//pull/19897) — chore(ci): djangosimple-tracer-dont-create-db-spans slo bump - [#19904](https://github.com//pull/19904) — ci(build_base_venvs): let riot install the dev package on Python 3.15 (PROF-14439) - [#19907](https://github.com//pull/19907) — ci(testrunner): rebuild 3.15-dev and pin Cython<3.3 on the 3.15 cache (PROF-14439) - [#19914](https://github.com//pull/19914) — test(profiling): unflake `gevent_greenlet_switch_not_blocked_by_profiler` - [#19923](https://github.com//pull/19923) — chore: exclude integration manifests yaml from source distributions _(no benchmark report)_ - [#19926](https://github.com//pull/19926) — ci: auto-retry `build windows` jobs - [#19936](https://github.com//pull/19936) — ci: bump wheel-builder IMAGE_TAGs after images (PROF-15844) - [#19951](https://github.com//pull/19951) — test(profiling): unflake gc tests - [#19954](https://github.com//pull/19954) — test(llmobs): fix flaky test _(no benchmark report)_ - [#19960](https://github.com//pull/19960) — ci: install ddtest 1.7.0 and riot 0.23.0 in testrunner image - [#19961](https://github.com//pull/19961) — test(iast): fix cloned venv poisoning [APPSEC-69907] _(no benchmark report)_ - [#19962](https://github.com//pull/19962) — test(iast): give appsec test servers their own port [APPSEC-69907] _(no benchmark report)_ - [#19964](https://github.com//pull/19964) — test(iast): assert MODULES_TO_UNPATCH deltas [APPSEC-69907] _(no benchmark report)_ - [#19965](https://github.com//pull/19965) — test(iast): unwrap weak hash before each test patches it [APPSEC-69907] _(no benchmark report)_ - [#19966](https://github.com//pull/19966) — test(iast): stop the global vulnerability limit leaking between tests [APPSEC-69907] _(no benchmark report)_ - [#19967](https://github.com//pull/19967) — chore(ci): pin numeric UID for testrunner image's bits user _(no benchmark report)_ - [#19968](https://github.com//pull/19968) — test(appsec): stop waiting 0.5s per api10 test teardown [APPSEC-69907] _(no benchmark report)_ - [#19973](https://github.com//pull/19973) — chore(tests): declare tracer environments in suitespec - [#19974](https://github.com//pull/19974) — test(telemetry): support PathResolutionError in wrapt 2.4.0 - [#19977](https://github.com//pull/19977) — ci(testrunner): install pip-tools for Python 3.15 - [#19984](https://github.com//pull/19984) — ci: update testrunner image - [#19985](https://github.com//pull/19985) — test(appsec): appsec_threats 55% faster, ~12h of CI saved per pipeline [APPSEC-69907] - [#19994](https://github.com//pull/19994) — ci(benchmarks): bump codeprovenancefork-fork-10 SLO - [#19997](https://github.com//pull/19997) — chore(tests): execute tracer suites with uv - [#20025](https://github.com//pull/20025) — test(appsec): run appsec_threats suites in parallel with pytest-xdist

[cit-pr-commenter-54b7da]

Dependency direction analysis

⚠️ Existing dependency direction violations

There are 230 dependency direction violations that already exist on the base branch and have not been changed by this PR.

Show existing violations (showing 5 of 230 highest severity)

ddtrace.internal.tracemethods -×-> ddtrace.trace  (internal-core -> product:tracing, score=134)
ddtrace.internal.ci_visibility.git_client -×-> ddtrace.trace  (product:ci_visibility -> product:tracing, score=132)
ddtrace.profiling.collector.pytorch -×-> ddtrace.trace  (product:profiling -> product:tracing, score=132)
ddtrace.internal.opentelemetry.span -×-> ddtrace.trace  (product:opentelemetry -> product:tracing, score=132)
ddtrace.internal.opentelemetry.trace -×-> ddtrace.trace  (product:opentelemetry -> product:tracing, score=132)

To see all violations, download the layers-base.json and layers-pr.json artifacts from this CI job and run:

uv run --script scripts/import-analysis/layers.py compare layers-base.json layers-pr.json

[cit-pr-commenter-54b7da]

Circular import analysis

⚠️ Existing circular imports

There are 3 circular imports that already exist on the base branch and have not been changed by this PR.

ddtrace.llmobs -> ddtrace.llmobs._evaluators -> ddtrace.llmobs._evaluators.format -> ddtrace.llmobs._experiment -> ddtrace.llmobs
ddtrace.errortracking._handled_exceptions.bytecode_injector -> ddtrace.errortracking._handled_exceptions.callbacks -> ddtrace.errortracking._handled_exceptions.collector -> ddtrace.errortracking._handled_exceptions.bytecode_reporting -> ddtrace.errortracking._handled_exceptions.bytecode_injector
ddtrace.appsec._asm_request_context -> ddtrace.appsec._iast._iast_request_context_base -> ddtrace.appsec._iast._iast_env -> ddtrace.appsec._iast.reporter -> ddtrace.appsec._exploit_prevention.stack_traces -> ddtrace.appsec._asm_request_context

[cit-pr-commenter-54b7da]

Codeowners resolved as

Resolved from the full PR diff against main using the target branch CODEOWNERS file.
CODEOWNERS team requests not listed below are not required by the current file set.

.gitlab/benchmarks/microbenchmarks.yml                                  @DataDog/apm-ecosystems-performance @DataDog/apm-core-python

[datadog-prod-us1-6]

 

✨ Unblock PR with BitsAI

⚠️ Warnings

❌ Your PR has failed checks. Please review the issues below and take necessary action before merging.

🚦 3 Pipeline jobs failed

DataDog/apm-reliability/dd-trace-py | Ubuntu_22_arm64.SID0: [test-app-python] — 🔧 Needs a code fix, caused by this PR

View more details · View in GitLab

DataDog/apm-reliability/dd-trace-py | Ubuntu_22_arm64.SIM: [test-app-python]

View more details · View in GitLab

DataDog/apm-reliability/dd-trace-py | download_dependency_wheels: [3.15.0rc1, 3.15]

View more details · View in GitLab

ℹ️ Info

No other issues found (see more)

🧪 All tests passed
❄️ No new flaky tests detected

🔄 Datadog retried 5 tests - 5 passed on retry

🚧 16 tests that failed were ignored due to quarantine

Useful? React with 👍 / 👎

_{This comment will be updated automatically if new data arrives.
🔗 Commit SHA: 74034ea | Docs | View more details | Give us feedback!}

[pr-commenter]

Benchmarks

Benchmark execution time: 2026-09-05 07:02:26

Comparing candidate commit 74034ea in PR branch augusto/update-known-flaky-benchmarks with baseline commit 6fd5a00 in branch main.

📊 Benchmarking dashboard

Found 0 performance improvements and 3 performance regressions! Performance is the same for 569 metrics, 10 unstable metrics, 7 known flaky benchmarks, 17 flaky benchmarks without significant changes.

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:iastaspects-join_aspect

• 🟥 execution_time [+39.846µs; +46.011µs] or [+18.071%; +20.866%]

scenario:otelspan-start

• 🟥 execution_time [+2.295ms; +3.331ms] or [+7.391%; +10.725%]

scenario:samplingrules-high_match

• 🟥 execution_time [+14.286µs; +17.292µs] or [+9.156%; +11.083%]

Unstable benchmarks

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

scenario:coreapiscenario-context_with_data_listeners

• unstable execution_time [-677.821ns; +794.008ns] or [-6.126%; +7.176%]

scenario:coreapiscenario-core_dispatch_1_listener

• unstable execution_time [-31.257ns; +34.380ns] or [-5.099%; +5.609%]

scenario:coreapiscenario-core_dispatch_50_listeners

• unstable execution_time [-1749.952ns; +1547.586ns] or [-10.245%; +9.060%]

scenario:coreapiscenario-core_dispatch_exception_listeners

• unstable execution_time [-1425.129ns; +1084.954ns] or [-10.770%; +8.199%]

scenario:coreapiscenario-core_dispatch_listeners

• unstable execution_time [-326.144ns; +321.964ns] or [-8.883%; +8.769%]

scenario:coreapiscenario-core_dispatch_no_args_listeners

• unstable execution_time [-242.919ns; +266.911ns] or [-8.359%; +9.184%]

scenario:coreapiscenario-core_dispatch_with_results_1_listener

• unstable execution_time [-56.852ns; +89.987ns] or [-5.007%; +7.924%]

scenario:coreapiscenario-core_dispatch_with_results_50_listeners

• unstable execution_time [-3656.644ns; +4379.960ns] or [-8.951%; +10.722%]

scenario:coreapiscenario-core_dispatch_with_results_listeners

• unstable execution_time [-655.462ns; +888.246ns] or [-8.184%; +11.091%]

scenario:packagesupdateimporteddependencies-import_many_stdlib_cached

• unstable execution_time [-43.457µs; +35.126µs] or [-7.437%; +6.011%]

Known flaky benchmarks

These benchmarks are marked as flaky and will not trigger a failure. Modify FLAKY_BENCHMARKS_REGEX to control which benchmarks are marked as flaky.

scenario:httppropagationinject-ids_only

• 🟥 execution_time [+1.825µs; +2.008µs] or [+10.748%; +11.826%]

scenario:iastaspects-casefold_noaspect

• 🟥 execution_time [+30.471µs; +35.120µs] or [+12.142%; +13.994%]

scenario:iastaspectsospath-ospathbasename_aspect

• 🟥 execution_time [+121.560µs; +128.425µs] or [+29.465%; +31.129%]

scenario:iastaspectssplit-rsplit_aspect

• 🟥 execution_time [+13.771µs; +17.558µs] or [+9.699%; +12.366%]

scenario:span-start

• 🟥 execution_time [+1.909ms; +2.152ms] or [+13.831%; +15.598%]

scenario:telemetryaddmetric-1-count-metric-1-times

• 🟥 execution_time [+360.775ns; +407.915ns] or [+12.375%; +13.992%]

scenario:tracer-small

• 🟥 execution_time [+40.902µs; +43.411µs] or [+12.669%; +13.447%]

Known flaky benchmarks without significant changes:

• scenario:errortrackingflasksqli-baseline
• scenario:flasksimple-iast-get
• scenario:iastaspects-casefold_aspect
• scenario:iastaspects-index_aspect
• scenario:iastaspects-ljust_noaspect
• scenario:iastaspects-lower_aspect
• scenario:iastaspects-replace_aspect
• scenario:iastaspects-rstrip_aspect
• scenario:iastaspects-swapcase_aspect
• scenario:iastaspects-title_noaspect
• scenario:iastaspects-translate_aspect
• scenario:iastaspects-translate_noaspect
• scenario:iastaspects-upper_noaspect
• scenario:packagespackageforrootmodulemapping-cache_off
• scenario:packagespackageforrootmodulemapping-cache_on
• scenario:sethttpmeta-all-enabled
• scenario:telemetryaddmetric-record-100-metrics