chore(ci): drop DogStatsD-disable workaround from OTLP tests - #1907
Conversation
#1570 set DD_DATA_PLANE_DOGSTATSD_ENABLED=false in the OTLP correctness tests to work around pre-7.80 misalignment with Core Agent behavior (#1567). Now that we run Agent 7.80, ADP's DogStatsD pipeline no longer perturbs these OTLP-only comparisons, so the override and its TODO are removed from all six OTLP cases. Closes #1577 Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
|
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 0b90e57899
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".
| - DD_DATA_PLANE_STANDALONE_MODE=true | ||
| - DD_DATA_PLANE_OTLP_ENABLED=true |
There was a problem hiding this comment.
Keep DogStatsD disabled while in standalone mode
With these OTLP comparison runs still setting DD_DATA_PLANE_STANDALONE_MODE=true, ADP does not bootstrap the remote-agent config stream that would let Agent 7.80 supply its resolved data_plane.dogstatsd.enabled decision; in this path DataPlaneDogStatsDConfiguration falls back to its default of true. Removing the explicit DD_DATA_PLANE_DOGSTATSD_ENABLED=false therefore re-enables ADP's DogStatsD pipeline in all six OTLP-only cases, so any DogStatsD/internal traffic can again perturb the baseline-vs-comparison results despite the tests intending to validate only OTLP behavior.
Useful? React with 👍 / 👎.
There was a problem hiding this comment.
This is correct that ADP will own the dogstatsd socket now, but this is actually fine. The Core Agent will read DD_DATA_PLANE_ENABLED, see that is enabled, and that DD_DATA_PLANE_DOGSTATSD_ENABLED isn't set to false so it'll not start dogstatsd (and thus the ports won't conflict).
There could be an argument here to keep dogstatsd disabled in ADP for these specific OTLP tests, but given that the default is that it is enabled, I think it is better to run the OTLP tests with in enabled.
Binary Size Analysis (Agent Data Plane)Baseline: 2a95db2 · Comparison: 0b90e57 · diff ✅ Binary size difference within thresholdChanges by Module
Detailed Symbol Changes |
Regression Detector (Agent Data Plane)Run ID: Optimization Goals: ✅ No significant changes detectedFine details of change detection per experiment (35)Experiments configured
Bounds Checks: ✅ Passed (5)
ExplanationA 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 ( |
| - DD_DATA_PLANE_STANDALONE_MODE=true | ||
| - DD_DATA_PLANE_OTLP_ENABLED=true |
Summary
The OTLP correctness tests carried
DD_DATA_PLANE_DOGSTATSD_ENABLED=false(added in #1570) purely to work around pre-7.80 misalignment between ADP and Core Agent behavior (#1567) — without it, ADP's DogStatsD pipeline perturbed these OTLP-only baseline-vs-comparison runs. Now that we're on Agent 7.80 that misalignment is gone, so this removes the override (and itsTODO) from all six OTLP cases, letting them run with ADP's default pipeline configuration as intended. This resolves the cleanup tracked in #1577.Closes #1577
Test plan
dockerruntime, Agent 7.80 baseline) with the override removed — all pass:otlp-metrics,otlp-traces,otlp-traces-ets,otlp-traces-probabilistic,otlp-traces-ottl-filtering,otlp-traces-ottl-transform.🤖 Generated with Claude Code