Prevent duplicate Feature Flagging runtimes

[leoromanovsky]

Motivation

Java Feature Flags is gaining two supported installation surfaces: a standalone dd-openfeature runtime and an agent-integrated runtime used by SSI. Without explicit process-wide ownership, both surfaces can start configuration polling and event delivery in the same JVM, producing duplicate requests and duplicate telemetry for customers who combine the library with the Java agent.

Changes

The Feature Flagging gateway now records which runtime owns configuration and event delivery. Claims are exclusive and idempotent for the current owner. The agent system claims ownership immediately before initialization and releases it during shutdown, while declining to start when the standalone runtime already owns the process.

The focused gateway and agent-system tests cover exclusive claims, idempotency, release behavior, and the agentless activation path when standalone ownership already exists.

Decisions

Ownership lives in the bootstrap gateway because it is the one process-wide object shared by the provider and agent integration. This PR intentionally does not add the standalone runtime yet; it establishes the invariant needed by the next PR in the stack. A runtime may retry its own claim, but a different runtime cannot replace the active owner.

Validation: ./gradlew :products:feature-flagging:feature-flagging-bootstrap:test :products:feature-flagging:feature-flagging-agent:test :products:feature-flagging:feature-flagging-bootstrap:spotlessCheck :products:feature-flagging:feature-flagging-agent:spotlessCheck

[datadog-datadog-prod-us1-2]

🎯 Code Coverage (details)
• Patch Coverage: 100.00%
• Overall Coverage: 78.42% (+19.82%)

_{This comment will be updated automatically if new data arrives.
🔗 Commit SHA: 9a0f9fb | Docs | View more details | Give us feedback!}

[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	14.00 s	13.86 s	[+0.3%; +1.8%] (maybe worse)
startup:insecure-bank:tracing:Agent	12.87 s	13.04 s	[-2.1%; -0.6%] (maybe better)
startup:petclinic:appsec:Agent	16.42 s	16.80 s	[-6.4%; +1.9%] (no difference)
startup:petclinic:iast:Agent	16.96 s	17.06 s	[-1.5%; +0.3%] (no difference)
startup:petclinic:profiling:Agent	16.70 s	16.90 s	[-2.2%; -0.1%] (maybe better)
startup:petclinic:sca:Agent	16.69 s	16.60 s	[-0.5%; +1.6%] (no difference)
startup:petclinic:tracing:Agent	16.22 s	16.11 s	[-0.4%; +1.6%] (no difference)

Commit: 9a0f9fbe · 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.