fix(function-autoscaler): use unified invocation metric

[borao]

TL;DR

Use function_request_total as the single recent-invocation metric for both the invocation service and gRPC proxy.

This removes the unused function_request query and stops attributing combined invocation traffic to gRPC proxy.

Additional Details

• Removes the duplicate recent-invocation query.
• Renames the combined query source to invocation.
• Reduces discovery from 12 to 8 sharded queries.
• Reduces per-function checks from two queries to one.
• Preserves fail-closed behavior when the invocation query fails.
• Updates the affected query-generation and failure tests.

Managed deployments must expose function_request_total before rolling out this autoscaler version. Self-managed deployments already expose that name.

For the Reviewer

Please focus on the query construction and per-function fail-closed behavior in work/discovery.rs.

For QA

• bazel test //src/control-plane-services/function-autoscaler/crates/server:rs_autoscaler_test --nocache_test_results --runs_per_test=3
  • 135 passed and 10 ignored per run.
• Built the autoscaler library, server, and image layer successfully.
• The image-index build could not fetch the private nvcr.io base image without registry credentials.

Issues

Related to #15

Checklist

• I am familiar with the Contributing Guidelines.
• I have signed off my commits for Developer Certificate of Origin (DCO) compliance.
• New or existing tests cover these changes.
• The documentation is up to date with these changes.

Summary by CodeRabbit

• Improvements
  • Streamlined function invocation discovery by using a unified request metric and gateway data.
  • Reduced the number of queries required for sharded and per-version discovery, improving efficiency.
  • Sharded discovery now continues using successful partial results when some queries fail.
  • Per-function discovery now fails safely when its only discovery query is unsuccessful.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: bc42e5b6-c99e-45c8-a6d2-4b088ed6e596

📥 Commits

Reviewing files that changed from the base of the PR and between ed688b4 and a2ac622.

📒 Files selected for processing (1)

• src/control-plane-services/function-autoscaler/crates/server/src/work/discovery.rs

Included review availability: Your plan provides up to 12 included reviews per hour; 11 remain after this review.

---

📝 Walkthrough

Walkthrough

Invocation discovery now uses function_request_total and LLM Gateway queries instead of separate invocation-service and gRPC-proxy sources. Sharded and per-version query counts decrease. Per-function failure handling and related tests now use the unified query.

Changes

Invocation discovery query consolidation

Layer / File(s)	Summary
Unified query construction src/control-plane-services/function-autoscaler/crates/server/src/work/discovery.rs	Discovery uses one invocation source with LLM Gateway queries. Sharded discovery uses eight queries, and per-version discovery uses one query. Per-function discovery fails closed when the invocation query fails.
Discovery query and failure tests src/control-plane-services/function-autoscaler/crates/server/src/work/discovery.rs	Tests validate the reduced query counts, function_request_total, and failure results without a gRPC request.

Estimated code review effort: 2 (Simple) | ~10 minutes

Merge Risk: 🔵 Low · up to a2ac6

The autoscaler now relies on function_request_total for invocation discovery. This is mergeable provided managed deployments expose that metric before upgrade; otherwise autoscaling discovery may fail closed.

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 55.56% which is insufficient. The required threshold is 80.00%. Docstring coverage is scoped to functions touched by this diff. Analyzed 9 functions across 1 files.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title follows Conventional Commits format and accurately describes the primary change to use a unified invocation metric in the function autoscaler.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

• Fix all pre-merge checks with AI

✨ Finishing Touches 💡 1

📝 Generate docstrings 💡

• Create stacked PR
• Commit on current branch

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch fix/autoscaler/remove-legacy-invocation-query

Warning

Some tools did not complete. Review the errors below.

🔧 Clippy (1.97.1)

Clippy execution failed

---

_{Comment @coderabbitai help to get the list of available commands.}

[github-actions]

🛡️ CodeQL Analysis

🚨 Found 5 issue(s)

Severity Breakdown:

• 🔴 Errors: 0
• 🟡 Warnings: 0
• 🔵 Notes: 0

📋 Top Issues

🔗 View full details in Security tab

_{🕐 Last updated: 2026-09-05 20:26:07 UTC | Commit: a2ac622}

[dmikhaylovnv]

LGTM