fix(health): normalize prober global status_counts

[philluiz2323]

Summary

• The two GLOBAL summary counters in runHealthProber / persistToKv were the missed siblings of fix(health): normalize unrecognized probe statuses before rollup #1739: summarizeGroup already routes through normalizeProbeStatus, but the cron return counts and served KV summary.status_counts / meta.status_counts still bucketed raw row.status.

What Changed

• src/health-prober.mjs: both global counts loops now increment normalizeProbeStatus(row.status).
• tests/health-prober.test.mjs: regression test with status: \"throttled\" asserting it folds into unknown.

Fixes #1942

Registry Safety

• No secrets, PATs, wallet data, private dashboards, private URLs, or validator-local state.
• Generated artifacts were produced by repo scripts, not hand-edited.
• R2-only/high-churn detail artifacts are not committed.
• Public API/OpenAPI/schema changes are intentional and documented. (None.)

Validation

• npx vitest run tests/health-prober.test.mjs -t \"folds unrecognized probe status\"
• git diff --check

[superagent-security]

Superagent didn't find any vulnerabilities or security issues in this PR.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 93.23%. Comparing base (2bcd5c1) to head (2cd147a).
⚠️ Report is 6 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #2044   +/-   ##
=======================================
  Coverage   93.23%   93.23%           
=======================================
  Files          48       48           
  Lines        7626     7626           
  Branches     2804     2802    -2     
=======================================
  Hits         7110     7110           
  Misses         90       90           
  Partials      426      426           

Files with missing lines	Coverage Δ
src/health-prober.mjs	94.35% <100.00%> (ø)

... and 2 files with indirect coverage changes

🚀 New features to boost your workflow:

• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[gittensory-orb]

Tip

🟩🟩🟩🟩🟩🟩🟩🟩🟩🟩🟩🟩

✅ Gittensory review — safe to merge

2 files · 1 AI reviewers · no blockers · readiness 73/100 · CI green · clean

✅ Approved — safe to merge

Review summary
Two-line fix routing both global `counts` accumulator loops in `runHealthProber` and `persistToKv` through `normalizeProbeStatus` before keying into the counts object, so unrecognized statuses like `"throttled"` fold into `unknown` instead of silently creating out-of-schema keys. The switch from `(counts[row.status] || 0) + 1` to `counts[normalizeProbeStatus(row.status)] += 1` is safe: the counts object is pre-initialized with all four canonical keys to `0`, and `normalizeProbeStatus` — already in use for the same purpose in `health-serving.mjs::summarizeRows` — always returns one of those keys. The regression test is thorough, covering all three downstream representations (`result.counts`, `summary.status_counts`, `meta.status_counts`) and explicitly asserting the spurious key is absent.

Signal	Result	Evidence
Code review	✅ No blockers	1 reviewers, synthesized
Linked issue	✅ Linked	#1942
Related work	✅ No active overlap found	No same-issue or scoped active PR overlap found.
Review load	✅ 20/20	Readiness component derived from cached public PR metadata and labels.
Validation evidence	❌ 5/25	Cached preflight status is hold.
Open PR queue	❌ 3/10	14 open PR(s), 9 likely reviewable, 5 unlinked.
Contributor context	✅ Confirmed Gittensor contributor	philluiz2323; Gittensor profile; 740 PR(s), 171 issue(s).
Gate result	✅ Passing	No configured blocker found.

Nits — 3 non-blocking

• tests/health-prober.test.mjs: the final `assert.equal(current.summary.status_counts.throttled, undefined)` is redundant — the preceding `assert.deepEqual` with an exact four-key object already verifies no extra keys exist; harmless but worth trimming for clarity.
• tests/health-prober.test.mjs: consider adding a parallel assertion that a canonical status (e.g. `status: "ok"`) still passes through `normalizeProbeStatus` unchanged in this same test, making it a full round-trip regression rather than only testing the normalization-to-unknown path.
• Repository config was not parsed

Review context

• Author: philluiz2323
• Role context: outside_contributor
• Public audience mode: oss maintainer
• Lane context: Repository registration is not available in the local Gittensory cache.
• Public profile languages: JavaScript, MDX, TypeScript, Cuda, HTML, Kotlin, Rust
• Official Gittensor activity: 740 PR(s), 171 issue(s).
• PR-specific overlap: none found.

Contributor next steps

• Fix blocker.
• Expect slower review.
• Refresh registry data or choose a registered active repo.

Signal definitions

• Related work = same linked issue, overlapping active PRs, or title/path similarity.
• Review load = cached public PR metadata such as size labels, changed paths, and preflight status.
• Open PR queue = repo-wide review pressure; it is not a PR quality failure.
• Contributor context = public GitHub/Gittensor identity context; non-Gittensor status is not a blocker.

Review details

Generated from public PR metadata and the diff. Advisory only; deterministic signals remain authoritative.

Two-line fix routing both global `counts` accumulator loops in `runHealthProber` and `persistToKv` through `normalizeProbeStatus` before keying into the counts object, so unrecognized statuses like `"throttled"` fold into `unknown` instead of silently creating out-of-schema keys. The switch from `(counts[row.status] || 0) + 1` to `counts[normalizeProbeStatus(row.status)] += 1` is safe: the counts object is pre-initialized with all four canonical keys to `0`, and `normalizeProbeStatus` — already in use for the same purpose in `health-serving.mjs::summarizeRows` — always returns one of those keys. The regression test is thorough, covering all three downstream representations (`result.counts`, `summary.status_counts`, `meta.status_counts`) and explicitly asserting the spurious key is absent.

Nits (2)

• tests/health-prober.test.mjs: the final `assert.equal(current.summary.status_counts.throttled, undefined)` is redundant — the preceding `assert.deepEqual` with an exact four-key object already verifies no extra keys exist; harmless but worth trimming for clarity.
• tests/health-prober.test.mjs: consider adding a parallel assertion that a canonical status (e.g. `status: "ok"`) still passes through `normalizeProbeStatus` unchanged in this same test, making it a full round-trip regression rather than only testing the normalization-to-unknown path.

_{🟩 Safe / merged · 🟦 Advisory · 🟨 Held for review · 🟥 Blocked / closed}

---

💰 Earn for open-source contributions like this. Gittensor lets GitHub contributors earn for the work they already do — register to start earning →.

Checked by Gittensory, a quiet PR intelligence layer for OSS maintainers.

• Re-run Gittensory review