[dougqh]

What Does This Do

When Kafka producers inject trace context, each header is now removed first only if it's already on the record. Before, every injected header went through headers.remove(key), which iterates with an iterator it allocates, though on an outgoing record the header is usually not there yet. Now lastHeader(key), which loops by index and doesn't allocate, probes first.

Behavior doesn't change:

• Duplicate keys still collapse. The Datadog and W3C codecs both write ot-baggage- headers (span baggage, the end-to-end start time). A second write still replaces the first, leaving one header.
• Read-only headers still throw. add throws IllegalStateException as remove did, so the producer's retry with a writable copy of the record still runs.

Motivation

A JFR profile of kafka-producer-perf-test under the agent showed ArrayList.iterator() at 4.5% of allocation, most of it from this setter, which runs for every injected header on every send.

KafkaInjectBenchmark (new) injects a span into new RecordHeaders (JDK 17, 3 forks, per inject):

spanBaggage	before ns	after ns	before B	after B
false	358.1 ± 2.1	342.3 ± 4.2	1728	1536
true	504.8 ± 8.9	508.0 ± 9.7	2400	2144

That's 192 B less per send, and 256 B with span baggage, where the replacing write still removes. The machine was busy (load 7–15), so treat the times as approximate; they're flat or slightly better.

Additional Notes

• Why not add-only: dropping the removal on records that start with no headers would have saved the scan too, but it would leave duplicate ot-baggage- headers, since two codecs write them.
• The value copy stays. value.getBytes(UTF_8) still copies each value. Removing it needs a way for codecs to write header values straight into bytes.
• Tests: TextMapInjectAdapterTest, in both client modules, covers adding, replacing, setting the same key twice and read-only headers. It passes against the old setter too.

🤖 Generated with Claude Code

[dd-octo-sts]

🟢 Java Benchmark SLOs — All performance SLOs passed

Suite	Status
Startup	🟢 pass

SLO thresholds are defined here based on automatically generated metrics. A warning is raised when results are within 5% of the threshold.

PR vs. master results

Scenario	Candidate	master	Δ (95% CI of mean)
startup:insecure-bank:iast:Agent	14.82 s	14.74 s	[-0.1%; +1.2%] (no difference)
startup:insecure-bank:tracing:Agent	13.56 s	13.69 s	[-1.6%; -0.3%] (maybe better)
startup:petclinic:appsec:Agent	17.25 s	17.12 s	[-0.0%; +1.5%] (no difference)
startup:petclinic:iast:Agent	17.08 s	17.00 s	[-0.4%; +1.4%] (no difference)
startup:petclinic:profiling:Agent	16.76 s	16.96 s	[-2.3%; -0.1%] (maybe better)
startup:petclinic:sca:Agent	17.12 s	16.96 s	[+0.1%; +1.8%] (maybe worse)
startup:petclinic:tracing:Agent	16.21 s	16.19 s	[-0.9%; +1.1%] (no difference)

Commit: 722d664e · CI Pipeline · Benchmarking Platform UI

---

Load and DaCapo benchmarks can be triggered manually in the GitLab pipeline. Results will appear in the Benchmarking Platform UI after completion.

[datadog-datadog-prod-us1]

🎯 Code Coverage (details)
• Patch Coverage: 100.00%
• Overall Coverage: 68.82% (+9.05%)

_{This comment will be updated automatically if new data arrives.
🔗 Commit SHA: 722d664 | Docs | Give us feedback!}

[chatgpt-codex-connector]

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-11-03T16:27:34.929069Z	722d664	Draft marked ready

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" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

[datadog-datadog-prod-us1]

More details

Both Kafka adapters retain duplicate-header replacement and read-only exceptions, preserving the producer’s retry with writable headers.

Was this helpful? React 👍 or 👎

_{Open Bits AI session}

_{🤖 Bits Code Review · Commit 722d664}

[pr-commenter]

Kafka / consumer-benchmark

Parameters

	Baseline	Candidate
baseline_or_candidate	baseline	candidate
git_branch	master	dougqh/kafka-inject-dedup-probe
git_commit_date	1793717677	1793720031
git_commit_sha	82b7ad5	722d664

See matching parameters

	Baseline	Candidate
ci_job_date	1793722111	1793722111
ci_job_id	2207145043	2207145043
ci_pipeline_id	149823280	149823280
cpu_model	Intel(R) Xeon(R) Platinum 8175M CPU @ 2.50GHz	Intel(R) Xeon(R) Platinum 8175M CPU @ 2.50GHz
jdkVersion	11.0.31	11.0.31
jmhVersion	1.36	1.36
jvm	/usr/lib/jvm/java-11-openjdk-amd64/bin/java	/usr/lib/jvm/java-11-openjdk-amd64/bin/java
jvmArgs	-Dhttp.proxyHost=127.0.0.1 -Dhttp.proxyPort=15002 -Dhttps.proxyHost=127.0.0.1 -Dhttps.proxyPort=15002 -Dhttp.nonProxyHosts=localhost	*.localhost
kernel_version	Linux runner-zfyrx7zua-project-304-concurrent-0-pic8vicl 6.8.0-1031-aws #33~22.04.1-Ubuntu SMP Thu Jun 26 14:22:30 UTC 2025 x86_64 x86_64 x86_64 GNU/Linux	Linux runner-zfyrx7zua-project-304-concurrent-0-pic8vicl 6.8.0-1031-aws #33~22.04.1-Ubuntu SMP Thu Jun 26 14:22:30 UTC 2025 x86_64 x86_64 x86_64 GNU/Linux
vmName	OpenJDK 64-Bit Server VM	OpenJDK 64-Bit Server VM
vmVersion	11.0.31+11-post-1ubuntu1-22.04.2-Ubuntu	11.0.31+11-post-1ubuntu1-22.04.2-Ubuntu

Summary

Found 0 performance improvements and 0 performance regressions! Performance is the same for 3 metrics, 0 unstable metrics.

See unchanged results

scenario	Δ mean throughput
scenario:not-instrumented/KafkaConsumerBenchmark.benchConsume	unsure [+1273.221op/s; +14019.943op/s] or [+0.543%; +5.982%]
scenario:only-tracing-dsm-disabled-benchmarks/KafkaConsumerBenchmark.benchConsume	same
scenario:only-tracing-dsm-enabled-benchmarks/KafkaConsumerBenchmark.benchConsume	same