chore(components): migrate DogStatsD source to full supervision

[tobz]

Summary

This PR is the third of many (hopefully just a few) for converting components to spawn all of their child tasks under their dedicated supervisor to fully unify task execution in Saluki under the supervision tree model, specifically handling the DogStatsD source.

We skipped the DogStatsD source in the initial sources/relay migration PR because it's very beefy, and deserves its own standalone PR.

Change Type

• Bug fix
• New feature
• Non-functional (chore, refactoring, docs)
• Performance

How did you test this PR?

Existing tests.

References

DADP-2

[tobz]

• chore(components): migrate DogStatsD source to full supervision #2385 👈 (View in Graphite)
• chore(components): migrate encoders to full supervision #2380
• chore(app, components, core, io): migrate sources/relays to full supervision #2342
• main

This stack of pull requests is managed by Graphite. Learn more about stacking.

[pr-commenter]

Binary Size Analysis (Agent Data Plane)

Baseline: 0384c33 · Comparison: 5d45df0 · diff
Analysis Configuration: stripped binaries · Pass/Fail Threshold: +5%
Sizes: 41.25 MiB (baseline) vs 41.19 MiB (comparison)
Size Change: -56.95 KiB (-0.13%)

✅ Binary size difference within threshold

Changes by Module

Module	File Size	Symbols
figment	-57.24 KiB	54
tracing	-53.05 KiB	31
saluki_core::runtime::workers	+44.30 KiB	23
saluki_components::sources::dogstatsd	+42.55 KiB	59
tokio	-38.41 KiB	736
otlp_protos::otlp_include::opentelemetry	-26.85 KiB	113
tonic	+23.12 KiB	41
http_body_util	-17.54 KiB	26
anon.5acabfac6d79e86a87fa9c42e93ba25f.983.llvm.18022061730132125650	+17.45 KiB	1
anon.5acabfac6d79e86a87fa9c42e93ba25f.983.llvm.8585410731666136712	-17.45 KiB	1
anyhow	+15.23 KiB	154
axum	-13.11 KiB	51
http	-13.01 KiB	27
core	+12.45 KiB	1726
anon.5acabfac6d79e86a87fa9c42e93ba25f.988.llvm.18022061730132125650	+12.37 KiB	1
anon.5acabfac6d79e86a87fa9c42e93ba25f.988.llvm.8585410731666136712	-12.28 KiB	1
saluki_components::encoders::datadog	+10.34 KiB	48
saluki_context::resolver::ContextResolver	-10.33 KiB	4
saluki_common::cache::CacheBuilder<K,V,W,H>	+10.24 KiB	4
rmp	+9.71 KiB	14

Detailed Symbol Changes

    FILE SIZE        VM SIZE    
 --------------  -------------- 
  [NEW] +34.8Ki  [NEW] +34.5Ki    _<saluki_components::sources::dogstatsd::_::<impl serde_core::de::Deserialize for saluki_components::sources::dogstatsd::DogStatsDConfiguration>::deserialize::__Visitor as serde_core::de::Visitor>::visit_map::h71b8b14f8fe676df
  [NEW] +20.1Ki  [NEW] +19.9Ki    saluki_core::runtime::workers::noninterruptible_worker::_{{closure}}::_{{closure}}::hc919fbe93619ca0a
  [NEW] +19.2Ki  [NEW] +19.1Ki    h2::proto::connection::Connection<T,P,B>::poll::h41693c50a5e2e6a5
  +301% +18.2Ki  +307% +18.2Ki    h2::proto::connection::DynConnection<B>::recv_frame::h5569cd28715d21f3
  [NEW] +17.5Ki  [NEW]     +76    anon.5acabfac6d79e86a87fa9c42e93ba25f.983.llvm.18022061730132125650
  [NEW] +17.2Ki  [NEW] +17.1Ki    h2::proto::connection::Connection<T,P,B>::poll::h2e3c8ca5e1a0a93e
  [NEW] +15.4Ki  [NEW] +15.2Ki    saluki_core::runtime::workers::noninterruptible_worker::_{{closure}}::_{{closure}}::h83b3da3366e34314
  [NEW] +12.4Ki  [NEW]     +81    anon.5acabfac6d79e86a87fa9c42e93ba25f.988.llvm.18022061730132125650
  [NEW] +9.96Ki  [NEW] +9.83Ki    saluki_components::transforms::apm_stats::span_concentrator::SpanConcentrator::flush::hb47f5dc465a13918
 +12e2% +9.46Ki +13e2% +9.46Ki    prost::message::Message::decode::h69d79adb9bd80dcf
  [NEW] +9.12Ki  [NEW] +8.99Ki    _<core::future::poll_fn::PollFn<F> as core::future::future::Future>::poll::h8104c67e547e4217
  [DEL] -10.6Ki  [DEL] -10.5Ki    _<tracing::instrument::Instrumented<T> as core::future::future::Future>::poll::h80c2eda1571612c8
  [DEL] -11.8Ki  [DEL] -11.7Ki    h2::server::Connection<T,B>::poll_closed::hcdb28a9fd1dd53ad
  [DEL] -12.3Ki  [DEL]     -81    anon.5acabfac6d79e86a87fa9c42e93ba25f.988.llvm.8585410731666136712
  [DEL] -12.9Ki  [DEL] -12.8Ki    h2::server::Connection<T,B>::poll_closed::h11fd495c293c3043
 -40.5% -13.0Ki -40.7% -13.0Ki    _<saluki_components::transforms::apm_stats::ApmStats as saluki_core::components::transforms::Transform>::run::_{{closure}}::hf2e1067e9305fcf4
  [DEL] -16.0Ki  [DEL] -15.8Ki    saluki_components::transforms::apm_stats::ApmStats::process_trace::hd369f1bfa3073d0a
  [DEL] -17.4Ki  [DEL]     -76    anon.5acabfac6d79e86a87fa9c42e93ba25f.983.llvm.8585410731666136712
  [DEL] -18.0Ki  [DEL] -17.9Ki    _<tracing::instrument::Instrumented<T> as core::future::future::Future>::poll::h7466185b0d228024
  [DEL] -22.7Ki  [DEL] -22.6Ki    _<figment::value::de::ConfiguredValueDe<I> as serde_core::de::Deserializer>::deserialize_any::h8e9b8330c1d94c85
  -1.4%  -105Ki  -1.5% -92.0Ki    [8280 Others]
  -0.1% -56.9Ki  -0.1% -44.0Ki    TOTAL

[pr-commenter]

Regression Detector (Agent Data Plane)

Run ID: 330e634e-3f43-45e3-ba2e-809c86ccc849
Baseline: 0384c331 · Comparison: 5d45df06 · diff

Optimization Goals: ✅ No significant changes detected

Fine details of change detection per experiment (5)

Experiments configured erratic: true are tagged (ignored) and skipped when determining which experiments regressed or improved. Experiments which are detected as erratic at runtime are tagged (erratic) to flag that the run's sample dispersion was high, but their regression / improvement signal still counts.

experiment	goal	Δ mean %	links
quality_gates_rss_dsd_heavy	memory	⚪ +0.35	metrics profiles logs
quality_gates_rss_dsd_medium	memory	⚪ +0.13	metrics profiles logs
quality_gates_rss_idle	memory	⚪ +0.13	metrics profiles logs
quality_gates_rss_dsd_ultraheavy	memory	⚪ -0.39	metrics profiles logs
quality_gates_rss_dsd_low	memory	⚪ -0.53	metrics profiles logs

Bounds Checks: ✅ Passed (5)

experiment	check	replicates	observed	links
quality_gates_rss_dsd_heavy	memory_usage	10/10	✅ 232 MiB ≤ 250 MiB	metrics profiles logs
quality_gates_rss_dsd_low	memory_usage	10/10	✅ 52.9 MiB ≤ 60 MiB	metrics profiles logs
quality_gates_rss_dsd_medium	memory_usage	10/10	✅ 91.6 MiB ≤ 100 MiB	metrics profiles logs
quality_gates_rss_dsd_ultraheavy	memory_usage	10/10	✅ 397 MiB ≤ 420 MiB	metrics profiles logs
quality_gates_rss_idle	memory_usage	10/10	✅ 34.3 MiB ≤ 40 MiB	metrics profiles logs

Explanation

A change is flagged as a regression when |Δ mean %| > 5.00% in the regressing direction for its optimization goal AND SMP marks the experiment as a regression (is_regression: true). Improvements use the matching criteria for the improving direction. Experiments configured erratic: true (tagged (ignored)) are skipped outright; experiments detected as erratic at runtime (tagged (erratic)) still count, since that flag describes sample dispersion rather than directional certainty. The Δ mean % cell is colored accordingly: 🟢 = improvement, 🔴 = regression, ⚪ = neutral. Reduction in CPU or memory is an improvement; reduction in ingress throughput is a regression. Experiments tagged (no analysis) show ⚠️ n/a: SMP ran them but produced no analysis, usually because a replicate failed and exhausted its retries. Check the SMP report for that experiment's replicate failures.

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 5d45df06eb

ℹ️ About Codex in GitHub

Codex has been enabled to automatically 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 👍.

When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".

[chatgpt-codex-connector]

Drain queued forwarding packets before supervisor exit

When the component supervisor begins its concurrent shutdown, spawn_interruptible drops ConnectedPacketForwarder::run at its current await point, so any DogStatsD messages already accepted into packets_rx are discarded instead of being sent. The previous detached loop remained alive until the listener/handler-held senders closed and then drained the channel; use a non-interruptible child that finishes after channel closure, or explicitly coordinate this queue's drain before shutdown completes.

Useful? React with 👍 / 👎.

[datadog-datadog-prod-us1]

Shutdown can stop the packet forwarder before DogStatsD decoders finish their lossless drain. This can lose queued metrics.

_{Open Bits AI session}

_{🤖 Datadog Autotest · Commit 5d45df0 · What is Autotest? · @DataDog review to ask questions · Any feedback? Reach out in #autotest}

[datadog-datadog-prod-us1]

Keep the forwarder active during queue drain

DogStatsD packet forwarding can lose metrics during shutdown.

Assertion details

• Input: Packet forwarding is enabled. Shutdown starts while a decoder or the forwarder queue still has packets.
• Expected: The forwarder stays active until all packet senders close. It sends all queued packets before it stops.
• Actual: The supervisor stops the interruptible forwarder at once. The decoders continue their queue drain. Later sends fail, and packets already in the forwarder queue are lost. Make the forward loop non-interruptible until all senders close. Update the forwarder shutdown tests to release the sender before supervisor shutdown. Add coverage for queue drain during shutdown.

Was this helpful? React 👍 or 👎
_{🤖 Datadog Autotest · What is Autotest? · @DataDog review to ask questions · Any feedback? Reach out in #autotest}