Fix issue with duplicate prometheus HELP/TYPE lines

[cigamit]

Resolves #766

[Copilot]

🟢 Approval recommended

Pull request overview

Fixes invalid Prometheus text exposition for /api/v2/metrics by ensuring shared “operational” metrics are emitted with a single # HELP/# TYPE header across namespaces, while still including per-namespace samples via the subsystem label.

Changes:

• Refactors metric rendering to separate header emission from sample emission (include_header, prometheus_header) and aggregates operational metric samples across namespaces.
• Adds OPERATIONAL_FIELDS to centrally define which metrics must be grouped/deduped.
• Adds a unit test to verify Prometheus parsing and prevent duplicate HELP/TYPE output regressions.

File summaries

File	Description
awx/main/analytics/subsystem_metrics.py	Aggregates operational metric samples and emits a single header per metric to avoid duplicate HELP/TYPE lines.
awx/main/tests/unit/analytics/test_subsystem_metrics.py	Adds a regression test ensuring /api/v2/metrics output is parseable and does not duplicate declarations.

Review details

• Files reviewed: 2/2 changed files
• Comments generated: 1
• Review effort level: Lite

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.

[Copilot]

🔵 Needs a closer look

Review details

Suppressed comments (1)

Previously missed (1) — in code that hasn't changed since the last review.

awx/main/analytics/subsystem_metrics.py:186

• HistogramM.to_prometheus() assumes instance_data[instance][self.field] is always present. In mixed-version clusters or after upgrades, stale instances can lack newer metrics; BaseM.to_prometheus() already guards against this, but HistogramM will raise a KeyError and break the entire /api/v2/metrics response.

        for instance in instance_data:
            # Build label string
            node_label = f'node="{instance}"'
            subsystem_label = f',subsystem="{namespace}"' if namespace else ''
            for i, b in enumerate(self.buckets):

• Files reviewed: 2/2 changed files
• Comments generated: 0 new
• Review effort level: Lite