Merge branch 'master' into fix/cache-hit-rate

Two cache bugs fixed:

1. Hit rate formula excluded stale hits — reported rate was artificially low
   because stale-while-revalidate responses (which ARE cache hits from the
   caller's perspective) were not counted. Changed formula from
   hits/(hits+misses) to (hits+staleHits)/(hits+staleHits+misses).

2. Bulk-health cache invalidated on every advert packet — in a mesh with
   dozens of nodes advertising every few seconds, this caused the expensive
   bulk-health query to be recomputed on nearly every request, defeating
   the cache entirely. Switched to 30s debounced invalidation via
   debouncedInvalidateBulkHealth().

Added regression test for hit rate formula in test-server-routes.js.

.github/workflows/deploy.yml

@@ -41,10 +41,10 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - name: Checkout code
-        uses: actions/checkout@v5
+        uses: actions/checkout@v4
 
       - name: Set up Go 1.22
-        uses: actions/setup-go@v6
+        uses: actions/setup-go@v5
         with:
           go-version: '1.22'
           cache-dependency-path: |
@@ -122,16 +122,9 @@ jobs:
           echo "| Server | ${SERVER_COV}% |" >> $GITHUB_STEP_SUMMARY
           echo "| Ingestor | ${INGESTOR_COV}% |" >> $GITHUB_STEP_SUMMARY
 
-      - name: Cancel workflow on failure
-        if: failure()
-        run: |
-          curl -s -X POST \
-            -H "Authorization: Bearer ${{ github.token }}" \
-            "https://api.github.com/repos/${{ github.repository }}/actions/runs/${{ github.run_id }}/cancel"
-
       - name: Upload Go coverage badges
         if: always()
-        uses: actions/upload-artifact@v5
+        uses: actions/upload-artifact@v4
         with:
           name: go-badges
           path: .badges/go-*.json
@@ -143,15 +136,15 @@ jobs:
   # ───────────────────────────────────────────────────────────────
   node-test:
     name: "🧪 Node.js Tests"
-    runs-on: [self-hosted, Linux]
+    runs-on: self-hosted
     steps:
       - name: Checkout code
-        uses: actions/checkout@v5
+        uses: actions/checkout@v4
         with:
           fetch-depth: 2
 
       - name: Set up Node.js 22
-        uses: actions/setup-node@v5
+        uses: actions/setup-node@v4
         with:
           node-version: '22'
 
@@ -197,11 +190,7 @@ jobs:
 
       - name: Install Playwright browser
         if: steps.changes.outputs.frontend == 'true'
-        run: |
-          # Install chromium (skips download if already cached on self-hosted runner)
-          npx playwright install chromium 2>/dev/null || true
-          # Install system deps only if missing (apt-get is slow)
-          npx playwright install-deps chromium 2>/dev/null || true
+        run: npx playwright install chromium --with-deps 2>/dev/null || true
 
       - name: Instrument frontend JS for coverage
         if: steps.changes.outputs.frontend == 'true'
@@ -231,32 +220,19 @@ jobs:
             sleep 1
           done
 
-      - name: Run Playwright E2E + coverage collection concurrently
+      - name: Run Playwright E2E tests
         if: steps.changes.outputs.frontend == 'true'
-        run: |
-          # Run E2E tests and coverage collection in parallel — both use the same server
-          BASE_URL=http://localhost:13581 node test-e2e-playwright.js 2>&1 | tee e2e-output.txt &
-          E2E_PID=$!
-          BASE_URL=http://localhost:13581 node scripts/collect-frontend-coverage.js 2>&1 | tee fe-coverage-output.txt &
-          COV_PID=$!
+        run: BASE_URL=http://localhost:13581 node test-e2e-playwright.js 2>&1 | tee e2e-output.txt
 
-          # Wait for both — E2E must pass, coverage is best-effort
-          E2E_EXIT=0
-          wait $E2E_PID || E2E_EXIT=$?
-          wait $COV_PID || true
-
-          # Fail if E2E failed
-          [ $E2E_EXIT -ne 0 ] && exit $E2E_EXIT
-          true
-
-      - name: Generate frontend coverage badges
+      - name: Collect frontend coverage report
         if: always() && steps.changes.outputs.frontend == 'true'
         run: |
+          BASE_URL=http://localhost:13581 node scripts/collect-frontend-coverage.js 2>&1 | tee fe-coverage-output.txt
+          
           E2E_PASS=$(grep -oP '[0-9]+(?=/)' e2e-output.txt | tail -1)
           
           mkdir -p .badges
-          # Merge E2E + coverage collector data if both exist
-          if [ -f .nyc_output/frontend-coverage.json ] || [ -f .nyc_output/e2e-coverage.json ]; then
+          if [ -f .nyc_output/frontend-coverage.json ]; then
             npx nyc report --reporter=text-summary --reporter=text 2>&1 | tee fe-report.txt
             FE_COVERAGE=$(grep 'Statements' fe-report.txt | head -1 | grep -oP '[\d.]+(?=%)' || echo "0")
             FE_COVERAGE=${FE_COVERAGE:-0}
@@ -283,24 +259,13 @@ jobs:
           fuser -k 13581/tcp 2>/dev/null || true
           PORT=13581 node server.js &
           SERVER_PID=$!
-          # Wait for server to be ready (up to 15s)
-          for i in $(seq 1 15); do
-            curl -sf http://localhost:13581/api/stats > /dev/null 2>&1 && break
-            sleep 1
-          done
+          sleep 5
           BASE_URL=http://localhost:13581 node test-e2e-playwright.js || true
           kill $SERVER_PID 2>/dev/null || true
 
-      - name: Cancel workflow on failure
-        if: failure()
-        run: |
-          curl -s -X POST \
-            -H "Authorization: Bearer ${{ github.token }}" \
-            "https://api.github.com/repos/${{ github.repository }}/actions/runs/${{ github.run_id }}/cancel"
-
       - name: Upload Node.js test badges
         if: always()
-        uses: actions/upload-artifact@v5
+        uses: actions/upload-artifact@v4
         with:
           name: node-badges
           path: .badges/
@@ -313,14 +278,14 @@ jobs:
   build:
     name: "🏗️ Build Docker Image"
     if: github.event_name == 'push'
-    needs: [go-test, node-test]
-    runs-on: [self-hosted, Linux]
+    needs: [go-test]
+    runs-on: self-hosted
     steps:
       - name: Checkout code
-        uses: actions/checkout@v5
+        uses: actions/checkout@v4
 
       - name: Set up Node.js 22
-        uses: actions/setup-node@v5
+        uses: actions/setup-node@v4
         with:
           node-version: '22'
 
@@ -339,10 +304,10 @@ jobs:
     name: "🚀 Deploy Staging"
     if: github.event_name == 'push'
     needs: [build]
-    runs-on: [self-hosted, Linux]
+    runs-on: self-hosted
     steps:
       - name: Checkout code
-        uses: actions/checkout@v5
+        uses: actions/checkout@v4
 
       - name: Start staging on port 82
         run: |
@@ -384,21 +349,21 @@ jobs:
     name: "📝 Publish Badges & Summary"
     if: github.event_name == 'push'
     needs: [deploy]
-    runs-on: [self-hosted, Linux]
+    runs-on: self-hosted
     steps:
       - name: Checkout code
-        uses: actions/checkout@v5
+        uses: actions/checkout@v4
 
       - name: Download Go coverage badges
         continue-on-error: true
-        uses: actions/download-artifact@v5
+        uses: actions/download-artifact@v4
         with:
           name: go-badges
           path: .badges/
 
       - name: Download Node.js test badges
         continue-on-error: true
-        uses: actions/download-artifact@v5
+        uses: actions/download-artifact@v4
         with:
           name: node-badges
           path: .badges/

scripts/collect-frontend-coverage.js

@@ -64,9 +64,9 @@ async function collectCoverage() {
   // ══════════════════════════════════════════════
   console.log('  [coverage] Home page — chooser...');
   // Clear localStorage to get chooser
-  await page.goto(BASE, { waitUntil: 'domcontentloaded', timeout: 10000 }).catch(() => {});
+  await page.goto(BASE, { waitUntil: 'networkidle', timeout: 15000 }).catch(() => {});
   await page.evaluate(() => localStorage.clear()).catch(() => {});
-  await page.goto(`${BASE}/#/home`, { waitUntil: 'domcontentloaded', timeout: 10000 }).catch(() => {});
+  await page.goto(`${BASE}/#/home`, { waitUntil: 'networkidle', timeout: 15000 }).catch(() => {});
 
   // Click "I'm new"
   await safeClick('#chooseNew');
@@ -105,7 +105,7 @@ async function collectCoverage() {
 
   // Switch to experienced mode
   await page.evaluate(() => localStorage.clear()).catch(() => {});
-  await page.goto(`${BASE}/#/home`, { waitUntil: 'domcontentloaded', timeout: 10000 }).catch(() => {});
+  await page.goto(`${BASE}/#/home`, { waitUntil: 'networkidle', timeout: 15000 }).catch(() => {});
   await safeClick('#chooseExp');
 
   // Interact with experienced home page
@@ -120,7 +120,7 @@ async function collectCoverage() {
   // NODES PAGE
   // ══════════════════════════════════════════════
   console.log('  [coverage] Nodes page...');
-  await page.goto(`${BASE}/#/nodes`, { waitUntil: 'domcontentloaded', timeout: 10000 }).catch(() => {});
+  await page.goto(`${BASE}/#/nodes`, { waitUntil: 'networkidle', timeout: 15000 }).catch(() => {});
 
   // Sort by EVERY column
   for (const col of ['name', 'public_key', 'role', 'last_seen', 'advert_count']) {
@@ -168,7 +168,7 @@ async function collectCoverage() {
   try {
     const firstNodeKey = await page.$eval('#nodesBody tr td:nth-child(2)', el => el.textContent.trim());
     if (firstNodeKey) {
-      await page.goto(`${BASE}/#/nodes/${firstNodeKey}`, { waitUntil: 'domcontentloaded', timeout: 10000 }).catch(() => {});
+      await page.goto(`${BASE}/#/nodes/${firstNodeKey}`, { waitUntil: 'networkidle', timeout: 15000 }).catch(() => {});
 
       // Click tabs on detail page
       await clickAll('.tab-btn, [data-tab]', 10);
@@ -191,7 +191,7 @@ async function collectCoverage() {
   try {
     const firstKey = await page.$eval('#nodesBody tr td:nth-child(2)', el => el.textContent.trim()).catch(() => null);
     if (firstKey) {
-      await page.goto(`${BASE}/#/nodes/${firstKey}?scroll=paths`, { waitUntil: 'domcontentloaded', timeout: 10000 }).catch(() => {});
+      await page.goto(`${BASE}/#/nodes/${firstKey}?scroll=paths`, { waitUntil: 'networkidle', timeout: 15000 }).catch(() => {});
     }
   } catch {}
 
@@ -199,7 +199,7 @@ async function collectCoverage() {
   // PACKETS PAGE
   // ══════════════════════════════════════════════
   console.log('  [coverage] Packets page...');
-  await page.goto(`${BASE}/#/packets`, { waitUntil: 'domcontentloaded', timeout: 10000 }).catch(() => {});
+  await page.goto(`${BASE}/#/packets`, { waitUntil: 'networkidle', timeout: 15000 }).catch(() => {});
 
   // Open filter bar
   await safeClick('#filterToggleBtn');
@@ -285,13 +285,13 @@ async function collectCoverage() {
   } catch {}
 
   // Navigate to specific packet by hash
-  await page.goto(`${BASE}/#/packets/deadbeef`, { waitUntil: 'domcontentloaded', timeout: 10000 }).catch(() => {});
+  await page.goto(`${BASE}/#/packets/deadbeef`, { waitUntil: 'networkidle', timeout: 15000 }).catch(() => {});
 
   // ══════════════════════════════════════════════
   // MAP PAGE
   // ══════════════════════════════════════════════
   console.log('  [coverage] Map page...');
-  await page.goto(`${BASE}/#/map`, { waitUntil: 'domcontentloaded', timeout: 10000 }).catch(() => {});
+  await page.goto(`${BASE}/#/map`, { waitUntil: 'networkidle', timeout: 15000 }).catch(() => {});
 
   // Toggle controls panel
   await safeClick('#mapControlsToggle');
@@ -345,7 +345,7 @@ async function collectCoverage() {
   // ANALYTICS PAGE
   // ══════════════════════════════════════════════
   console.log('  [coverage] Analytics page...');
-  await page.goto(`${BASE}/#/analytics`, { waitUntil: 'domcontentloaded', timeout: 10000 }).catch(() => {});
+  await page.goto(`${BASE}/#/analytics`, { waitUntil: 'networkidle', timeout: 15000 }).catch(() => {});
 
   // Click EVERY analytics tab
   const analyticsTabs = ['overview', 'rf', 'topology', 'channels', 'hashsizes', 'collisions', 'subpaths', 'nodes', 'distance'];
@@ -383,7 +383,7 @@ async function collectCoverage() {
 
   // Deep-link to each analytics tab via URL
   for (const tab of analyticsTabs) {
-    await page.goto(`${BASE}/#/analytics?tab=${tab}`, { waitUntil: 'domcontentloaded', timeout: 10000 }).catch(() => {});
+    await page.goto(`${BASE}/#/analytics?tab=${tab}`, { waitUntil: 'networkidle', timeout: 15000 }).catch(() => {});
   }
 
   // Region filter on analytics
@@ -396,7 +396,7 @@ async function collectCoverage() {
   // CUSTOMIZE
   // ══════════════════════════════════════════════
   console.log('  [coverage] Customizer...');
-  await page.goto(BASE, { waitUntil: 'domcontentloaded', timeout: 10000 }).catch(() => {});
+  await page.goto(BASE, { waitUntil: 'networkidle', timeout: 15000 }).catch(() => {});
   await safeClick('#customizeToggle');
 
   // Click EVERY customizer tab
@@ -503,7 +503,7 @@ async function collectCoverage() {
   // CHANNELS PAGE
   // ══════════════════════════════════════════════
   console.log('  [coverage] Channels page...');
-  await page.goto(`${BASE}/#/channels`, { waitUntil: 'domcontentloaded', timeout: 10000 }).catch(() => {});
+  await page.goto(`${BASE}/#/channels`, { waitUntil: 'networkidle', timeout: 15000 }).catch(() => {});
   // Click channel rows/items
   await clickAll('.channel-item, .channel-row, .channel-card', 3);
   await clickAll('table tbody tr', 3);
@@ -512,7 +512,7 @@ async function collectCoverage() {
   try {
     const channelHash = await page.$eval('table tbody tr td:first-child', el => el.textContent.trim()).catch(() => null);
     if (channelHash) {
-      await page.goto(`${BASE}/#/channels/${channelHash}`, { waitUntil: 'domcontentloaded', timeout: 10000 }).catch(() => {});
+      await page.goto(`${BASE}/#/channels/${channelHash}`, { waitUntil: 'networkidle', timeout: 15000 }).catch(() => {});
     }
   } catch {}
 
@@ -520,7 +520,7 @@ async function collectCoverage() {
   // LIVE PAGE
   // ══════════════════════════════════════════════
   console.log('  [coverage] Live page...');
-  await page.goto(`${BASE}/#/live`, { waitUntil: 'domcontentloaded', timeout: 10000 }).catch(() => {});
+  await page.goto(`${BASE}/#/live`, { waitUntil: 'networkidle', timeout: 15000 }).catch(() => {});
 
   // VCR controls
   await safeClick('#vcrPauseBtn');
@@ -603,14 +603,14 @@ async function collectCoverage() {
   // TRACES PAGE
   // ══════════════════════════════════════════════
   console.log('  [coverage] Traces page...');
-  await page.goto(`${BASE}/#/traces`, { waitUntil: 'domcontentloaded', timeout: 10000 }).catch(() => {});
+  await page.goto(`${BASE}/#/traces`, { waitUntil: 'networkidle', timeout: 15000 }).catch(() => {});
   await clickAll('table tbody tr', 3);
 
   // ══════════════════════════════════════════════
   // OBSERVERS PAGE
   // ══════════════════════════════════════════════
   console.log('  [coverage] Observers page...');
-  await page.goto(`${BASE}/#/observers`, { waitUntil: 'domcontentloaded', timeout: 10000 }).catch(() => {});
+  await page.goto(`${BASE}/#/observers`, { waitUntil: 'networkidle', timeout: 15000 }).catch(() => {});
   // Click observer rows
   const obsRows = await page.$$('table tbody tr, .observer-card, .observer-row');
   for (let i = 0; i < Math.min(obsRows.length, 3); i++) {
@@ -631,7 +631,7 @@ async function collectCoverage() {
   // PERF PAGE
   // ══════════════════════════════════════════════
   console.log('  [coverage] Perf page...');
-  await page.goto(`${BASE}/#/perf`, { waitUntil: 'domcontentloaded', timeout: 10000 }).catch(() => {});
+  await page.goto(`${BASE}/#/perf`, { waitUntil: 'networkidle', timeout: 15000 }).catch(() => {});
   await safeClick('#perfRefresh');
   await safeClick('#perfReset');
 
@@ -641,14 +641,14 @@ async function collectCoverage() {
   console.log('  [coverage] App.js — router + global...');
 
   // Navigate to bad route to trigger error/404
-  await page.goto(`${BASE}/#/nonexistent-route`, { waitUntil: 'domcontentloaded', timeout: 10000 }).catch(() => {});
+  await page.goto(`${BASE}/#/nonexistent-route`, { waitUntil: 'networkidle', timeout: 15000 }).catch(() => {});
 
   // Navigate to every route via hash
   const allRoutes = ['home', 'nodes', 'packets', 'map', 'live', 'channels', 'traces', 'observers', 'analytics', 'perf'];
   for (const route of allRoutes) {
     try {
       await page.evaluate((r) => { location.hash = '#/' + r; }, route);
-      await new Promise(r => setTimeout(r, 200));
+      await page.waitForLoadState('networkidle').catch(() => {});
     } catch {}
   }
 
@@ -788,14 +788,14 @@ async function collectCoverage() {
   console.log('  [coverage] Region filter...');
   try {
     // Open region filter on nodes page
-    await page.goto(`${BASE}/#/nodes`, { waitUntil: 'domcontentloaded', timeout: 10000 }).catch(() => {});
+    await page.goto(`${BASE}/#/nodes`, { waitUntil: 'networkidle', timeout: 15000 }).catch(() => {});
     await safeClick('#nodesRegionFilter');
     await clickAll('#nodesRegionFilter input[type="checkbox"]', 3);
   } catch {}
 
   // Region filter on packets
   try {
-    await page.goto(`${BASE}/#/packets`, { waitUntil: 'domcontentloaded', timeout: 10000 }).catch(() => {});
+    await page.goto(`${BASE}/#/packets`, { waitUntil: 'networkidle', timeout: 15000 }).catch(() => {});
     await safeClick('#packetsRegionFilter');
     await clickAll('#packetsRegionFilter input[type="checkbox"]', 3);
   } catch {}
@@ -807,7 +807,7 @@ async function collectCoverage() {
   for (const route of allRoutes) {
     try {
       await page.evaluate((r) => { location.hash = '#/' + r; }, route);
-      await new Promise(r => setTimeout(r, 200));
+      await page.waitForLoadState('networkidle').catch(() => {});
     } catch {}
   }
 

server.js

@@ -207,6 +207,13 @@ class TTLCache {
       if (key.startsWith(prefix)) this.store.delete(key);
     }
   }
+  debouncedInvalidateBulkHealth() {
+    if (this._bulkHealthTimer) return;
+    this._bulkHealthTimer = setTimeout(() => {
+      this._bulkHealthTimer = null;
+      this.invalidate('bulk-health');
+    }, 30000);
+  }
   debouncedInvalidateAll() {
     if (this._debounceTimer) return;
     this._debounceTimer = setTimeout(() => {
@@ -410,7 +417,7 @@ app.get('/api/perf', (req, res) => {
     avgMs: perfStats.requests ? Math.round(perfStats.totalMs / perfStats.requests * 10) / 10 : 0,
     endpoints: Object.fromEntries(sorted),
     slowQueries: perfStats.slowQueries.slice(-20),
-    cache: { size: cache.size, hits: cache.hits, misses: cache.misses, staleHits: cache.staleHits, recomputes: cache.recomputes, hitRate: cache.hits + cache.misses > 0 ? Math.round(cache.hits / (cache.hits + cache.misses) * 1000) / 10 : 0 },
+    cache: { size: cache.size, hits: cache.hits, misses: cache.misses, staleHits: cache.staleHits, recomputes: cache.recomputes, hitRate: cache.hits + cache.staleHits + cache.misses > 0 ? Math.round((cache.hits + cache.staleHits) / (cache.hits + cache.staleHits + cache.misses) * 1000) / 10 : 0 },
     packetStore: pktStore.getStats(),
     sqlite: (() => {
       try {
@@ -519,7 +526,7 @@ app.get('/api/health', (req, res) => {
       misses: cache.misses,
       staleHits: cache.staleHits,
       recomputes: cache.recomputes,
-      hitRate: cache.hits + cache.misses > 0 ? Math.round(cache.hits / (cache.hits + cache.misses) * 1000) / 10 : 0,
+      hitRate: cache.hits + cache.staleHits + cache.misses > 0 ? Math.round((cache.hits + cache.staleHits) / (cache.hits + cache.staleHits + cache.misses) * 1000) / 10 : 0,
     },
     websocket: {
       clients: wsClients,
@@ -723,7 +730,7 @@ for (const source of mqttSources) {
             // Invalidate this node's caches on advert
             cache.invalidate('node:' + p.pubKey);
             cache.invalidate('health:' + p.pubKey);
-            cache.invalidate('bulk-health');
+            cache.debouncedInvalidateBulkHealth();
 
             // Cross-reference: if this node's pubkey matches an existing observer, backfill observer name
             if (p.name && p.pubKey) {

test-e2e-playwright.js

@@ -909,19 +909,6 @@ async function run() {
     assert(hexDump, 'Hex dump should be visible after selecting a packet');
   });
 
-  // Extract frontend coverage if instrumented server is running
-  try {
-    const coverage = await page.evaluate(() => window.__coverage__);
-    if (coverage) {
-      const fs = require('fs');
-      const path = require('path');
-      const outDir = path.join(__dirname, '.nyc_output');
-      if (!fs.existsSync(outDir)) fs.mkdirSync(outDir, { recursive: true });
-      fs.writeFileSync(path.join(outDir, 'e2e-coverage.json'), JSON.stringify(coverage));
-      console.log(`Frontend coverage from E2E: ${Object.keys(coverage).length} files`);
-    }
-  } catch {}
-
   await browser.close();
 
   // Summary

test-server-routes.js

@@ -1254,6 +1254,24 @@ seedTestData();
     lastPathSeenMap.delete(liveNode);
   });
 
+  // ── Cache hit rate includes stale hits ──
+  await t('Cache hitRate includes staleHits in formula', async () => {
+    cache.clear();
+    cache.hits = 0;
+    cache.misses = 0;
+    cache.staleHits = 0;
+    // Simulate: 3 hits, 2 stale hits, 5 misses => rate = (3+2)/(3+2+5) = 50%
+    cache.hits = 3;
+    cache.staleHits = 2;
+    cache.misses = 5;
+    const r = await request(app).get('/api/health').expect(200);
+    assert(r.body.cache.hitRate === 50, 'hitRate should be (hits+staleHits)/(hits+staleHits+misses) = 50%, got ' + r.body.cache.hitRate);
+    // Reset
+    cache.hits = 0;
+    cache.misses = 0;
+    cache.staleHits = 0;
+  });
+
   // ── Summary ──
   console.log(`\n═══ Server Route Tests: ${passed} passed, ${failed} failed ═══`);
   if (failed > 0) process.exit(1);