Skip to content

feat(inst/langchain4j): add LangChain4j 1.0 phase attribution instrumentation#11530

Draft
jbachorik wants to merge 1 commit into
jb/llm-ddt-2-profilerfrom
jb/llm-ddt-3-lc4j
Draft

feat(inst/langchain4j): add LangChain4j 1.0 phase attribution instrumentation#11530
jbachorik wants to merge 1 commit into
jb/llm-ddt-2-profilerfrom
jb/llm-ddt-3-lc4j

Conversation

@jbachorik
Copy link
Copy Markdown
Contributor

@jbachorik jbachorik commented Jun 2, 2026 •
edited by atlassian Bot
Loading

What Does This Do

New instrumentation module dd-java-agent/instrumentation/langchain4j/langchain4j-1.0 with three InstrumenterModule.Profiling hooks tagging LangChain4j 1.0 pipeline boundaries:

Class Hook point Phase set
ChatModelInstrumentation ChatModel#chat(ChatRequest) — all implementations awaiting_inference
ToolExecutorInstrumentation ToolExecutor#execute(ToolExecutionRequest, Object) tool_execution
AiServicesInstrumentation DefaultAiServices$1#invoke(...) (anonymous InvocationHandler) context_build

Each advice sets llm.agent.phase on enter via Profiling.get().newScope().setContextValue(...) and clears on exit. Muzzle passes LC4j versions 1.0.0–1.15.1.

Motivation

LangChain4j uses its own HTTP client — the existing OpenAI Java SDK instrumentation does not fire on LC4j calls. Without this module, CPU flamegraphs from LC4j services show undifferentiated work.

Part of PROF-14858. Stacked on #11529 (DatadogProfiler slot). Next: [ddt-4-demo] (demo app).

Additional Notes

  • Activated only when TargetSystem.PROFILING is enabled (InstrumenterModule.Profiling).
  • ToolExecutor is at dev.langchain4j.service.tool (not agent.tool as the spec guessed).
  • AiServices handler is the anonymous class DefaultAiServices$1.
  • langchain4j-ollama 1.0.0 does not exist in Maven Central; earliest 1.x is 1.2.0.

Contributor Checklist

  • Format the title according to the contribution guidelines
  • Assign the type: and (comp: or inst:) labels in addition to any other useful labels
  • Avoid using close, fix, or any linking keywords when referencing an issue
    Use solves instead, and assign the PR milestone to the issue
  • Update the CODEOWNERS file on source file addition, migration, or deletion
  • Update public documentation with any new configuration flags or behaviors
  • Add your completed PR to the merge queue by commenting /merge. You can also:
    • Customize the commit message associated with the merge with /merge --commit-message "..."
    • Remove your PR from the merge queue with /merge -c
    • Skip all merge queue checks with /merge -f --reason "reason"; please use this judiciously, as some checks do not run at the PR-level
    • Get more information in this doc

Jira ticket: PROF-14863

@jbachorik jbachorik mentioned this pull request Jun 2, 2026
Draft
@jbachorik jbachorik added the AI label Jun 2, 2026
@datadog-datadog-prod-us1-2
Copy link
Copy Markdown
Contributor

datadog-datadog-prod-us1-2 Bot commented Jun 2, 2026 •
edited by datadog-datadog-prod-us1 Bot
Loading

Pipelines

✨ Fix all issues with BitsAI

⚠️ Warnings

🚦 15 Pipeline jobs failed

DataDog/apm-reliability/dd-trace-java | spotless   View in Datadog   GitLab

🔧 Fix in code (Fix with Cursor). Format violations in src/main/java/com/datadog/profiling/ddprof/DatadogProfiler.java. Run './gradlew spotlessApply' to fix.

Run system tests | main / End-to-end #4 / spring-boot-openliberty 4   View in Datadog   GitHub Actions

🔧 Fix in code (Fix with Cursor). 1 failed test. AssertionError: assert not ['[dd.trace 2026-06-02 13:33:56:311 +0000] [dd-jmx-collector] WARN org.datadog.jmxfetch.tasks.TaskProcessor - JMXFetch... com.ibm.ws.webcontainer.webapp.WebAppErrorReport: SRVE0295E: Error reported: 500\n']

Run system tests | main / End-to-end #5 / spring-boot 5   View in Datadog   GitHub Actions

🔧 Fix in code (Fix with Cursor). 1 failed test due to assertion error. AssertionError: Request failed: None at tests/ffe/test_dynamic_evaluation.py:237.

View all 15 failed jobs.

Useful? React with 👍 / 👎

This comment will be updated automatically if new data arrives.
🔗 Commit SHA: c8dab32 | Docs | Datadog PR Page | Give us feedback!

@PerfectSlayer PerfectSlayer added tag: ai generated Largely based on code generated by an AI or LLM and removed AI labels Jun 3, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

tag: ai generated Largely based on code generated by an AI or LLM

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

2 participants

@jbachorik @PerfectSlayer