chore(api): emit explicit zero for IPNS-entries gauge on empty table

[FSM1]

Summary

Two small chore cleanups, batched on this branch.

1. chore(api) — emit explicit zero for the IPNS-entries gauge on an empty table

MetricsService.collectGauges() calls this.ipnsEntriesTotal.reset() then repopulates cipherbox_ipns_entries_total from a GROUP BY record_type query over folder_ipns. When the table is empty that query returns zero rows, so reset() leaves the gauge with no series at all — Prometheus then serves the last non-zero sample through its staleness window and Grafana stat panels "stick" at it (a freshly-wiped environment mis-reporting as N IPNS entries; observed as a stale 43 after a staging reset).

Fix: seed every known record_type ('folder', 'file') to 0 after reset(), then overlay the real counts. users_total/files_total already behaved correctly because they always .set(0); this brings the label-partitioned IPNS gauge in line. Internal metrics only — no API surface change, no api-client regen.

2. chore(ci) — remove the stale release-as guard

Drops the "Check for stale release-as pins" CI step and its companion "Test stale release-as guard" step from ci.yml, and deletes .github/scripts/check-stale-release-as.js + its node:test suite. The guard failed CI whenever a release-as pin equalled its manifest version, adding friction on every release for little value. The underlying advice (don't leave an already-shipped release-as pin in place) stays documented in CLAUDE.md as a manual check — just no longer CI-enforced.

Testing

• metrics.service.spec.ts: added an ipnsEntriesTotal gauge zero-state (regression) block — empty table → 0 for both types; partial population overlays real counts while absent types stay 0. Full spec: 11 passed.
• eslint clean, tsc --noEmit clean on the API project.
• ci.yml / CLAUDE.md pass prettier --check; YAML structure verified (lint job intact, typecheck job follows).

🤖 Generated with Claude Code

[coderabbitai]

Walkthrough

The IPNS metrics collector now seeds known record_type labels with zero values before applying database counts. The spec adds regression coverage for empty input and partial overlay output for cipherbox_ipns_entries_total.

Changes

IPNS gauge zero seeding

Layer / File(s)	Summary
Gauge initialization and regression coverage apps/api/src/metrics/metrics.service.ts, apps/api/src/metrics/metrics.service.spec.ts	ipnsEntriesTotal now seeds folder and file at zero before database counts are applied, and the spec asserts zero-series emission plus overwrite behavior for present counts.

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~10 minutes

Possibly related PRs

• FSM1/cipher-box#141: Introduces the MetricsService/collectGauges() path and the ipnsEntriesTotal gauge that this PR extends.

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title clearly and accurately summarizes the main change: emitting explicit zeros for the IPNS entries gauge on empty tables.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
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.
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

📝 Generate docstrings

• Create stacked PR
• Commit on current branch

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch chore/ipns-entries-gauge-zero-state

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

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

[github-actions]

Release Preview

No version bumps detected. All changes are in unversioned paths or use exempt commit types.

[greptile-apps]

Greptile Summary

This PR fixes cipherbox_ipns_entries_total mis-reporting on an empty folder_ipns table: after reset() the gauge now seeds every known record_type ('folder', 'file') to 0 before overlaying real DB counts, so Prometheus serves an explicit zero instead of holding the last non-zero sample through its staleness window. It also includes regression tests and — unrelated to the metrics fix — removes the stale release-as CI guard entirely.

• Core fix (metrics.service.ts): adds IPNS_RECORD_TYPES constant and a seed loop after reset(), bringing ipnsEntriesTotal in line with the scalar gauges that already call .set(0).
• Regression tests (metrics.service.spec.ts): two new cases assert both types emit 0 on an empty table and that real counts overlay correctly while absent types remain 0.
• Unrelated cleanup (.github/): deletes check-stale-release-as.js, its test, and the two CI steps that ran them; downgrades an automated release-pin guard to a manual convention in CLAUDE.md.

Confidence Score: 5/5

The metrics fix is minimal and correct; the unrelated CI guard removal is deliberate but unexplained.

The core change is a small, well-tested seed loop that only affects internal Prometheus gauge reporting. The logic is sound: reset → seed known types to 0 → overlay real counts. Tests cover both the zero-state and partial-count paths. The unrelated removal of the stale release-as CI guard is worth clarifying but does not affect the correctness of any deployed code.

.github/workflows/ci.yml — removal of the release-as CI guard is unexplained and unrelated to the stated PR scope.

Important Files Changed

Filename	Overview
apps/api/src/metrics/metrics.service.ts	Adds IPNS_RECORD_TYPES constant and seeds known record types to 0 after reset() before overlaying DB counts — correctly fixes the stale Prometheus series on empty table.
apps/api/src/metrics/metrics.service.spec.ts	Adds two regression tests for the zero-state behavior: one for an empty table (both types → 0) and one for partial DB results (real count overlays, missing type → 0).
.github/workflows/ci.yml	Removes the stale release-as CI guard steps without explanation in the PR description, downgrading an automated safety check to a manual one.
.github/scripts/check-stale-release-as.js	Deleted — the stale release-as guard script is fully removed alongside its CI steps.
.github/scripts/check-stale-release-as.test.js	Deleted — unit tests for the stale release-as guard removed alongside the script itself.
CLAUDE.md	Updates the release-as pin note to acknowledge the check is no longer CI-enforced and must be done manually.

Sequence Diagram

%%{init: {'theme': 'neutral'}}%%
sequenceDiagram
    participant Scheduler as setInterval (30s)
    participant collectGauges
    participant DB as PostgreSQL
    participant gauge as ipnsEntriesTotal

    Scheduler->>collectGauges: tick
    collectGauges->>DB: GROUP BY record_type (folder_ipns)
    DB-->>collectGauges: "[] (empty) or [{recordType, count}]"

    collectGauges->>gauge: reset()
    Note over collectGauges,gauge: NEW: seed known types to 0
    collectGauges->>gauge: labels('folder').set(0)
    collectGauges->>gauge: labels('file').set(0)

    loop for each row returned
        collectGauges->>gauge: labels(row.recordType).set(parseInt(row.count))
    end

    Note over gauge: Prometheus now always has an explicit series,
    Note over gauge: no stale-window carry-over

Loading

%%{init: {'theme': 'base', 'themeVariables': {"darkMode": true, "background": "#0d1117", "primaryColor": "#21262d", "primaryTextColor": "#e6edf3", "primaryBorderColor": "#8b949e", "lineColor": "#8b949e", "textColor": "#e6edf3", "edgeLabelBackground": "#161b22", "actorBkg": "#21262d", "actorBorder": "#8b949e", "actorTextColor": "#e6edf3", "actorLineColor": "#8b949e", "signalColor": "#8b949e", "signalTextColor": "#e6edf3", "noteBkgColor": "#373320", "noteBorderColor": "#d4a72c", "noteTextColor": "#f0e6c0", "labelBoxBkgColor": "#21262d", "labelBoxBorderColor": "#8b949e", "labelTextColor": "#e6edf3", "loopTextColor": "#e6edf3", "activationBkgColor": "#30363d", "activationBorderColor": "#8b949e"}}}%%
sequenceDiagram
    participant Scheduler as setInterval (30s)
    participant collectGauges
    participant DB as PostgreSQL
    participant gauge as ipnsEntriesTotal

    Scheduler->>collectGauges: tick
    collectGauges->>DB: GROUP BY record_type (folder_ipns)
    DB-->>collectGauges: "[] (empty) or [{recordType, count}]"

    collectGauges->>gauge: reset()
    Note over collectGauges,gauge: NEW: seed known types to 0
    collectGauges->>gauge: labels('folder').set(0)
    collectGauges->>gauge: labels('file').set(0)

    loop for each row returned
        collectGauges->>gauge: labels(row.recordType).set(parseInt(row.count))
    end

    Note over gauge: Prometheus now always has an explicit series,
    Note over gauge: no stale-window carry-over

Loading

_{Reviews (2): Last reviewed commit: "chore(ci): remove stale release-as guard" | Re-trigger Greptile}

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 70.01%. Comparing base (4b2e81c) to head (ea8260a).
⚠️ Report is 1 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #561      +/-   ##
==========================================
+ Coverage   65.96%   70.01%   +4.05%     
==========================================
  Files         149      177      +28     
  Lines       11503    21151    +9648     
  Branches     1305     1305              
==========================================
+ Hits         7588    14809    +7221     
- Misses       3673     6100    +2427     
  Partials      242      242              

Flag	Coverage Δ
api	85.95% <ø> (ø)
api-client	85.95% <ø> (ø)
core	85.95% <ø> (ø)
crypto	85.95% <ø> (ø)
desktop	16.45% <ø> (-14.99%)	⬇️
rust	68.64% <ø> (?)
sdk	85.95% <ø> (ø)
sdk-core	85.95% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.
see 85 files with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.