ci: remove temporary npm-hang diagnostic instrumentation

The diagnostic blocks (pre-npm-ci probes, --loglevel=http on
npm ci, post-failure debug-log artifact upload) were added to
identify the root cause of the npm-ci silent 8-minute hang on
protected runners. That root cause is identified and fixed: the
lockfile contained 415 `resolved:` URLs pointing to
npm-proxy.dev.databricks.com (the internal dev-only proxy), which
the protected runners cannot reach. The fix is the `Rewrite
lockfile to JFrog registry` step in `.github/actions/setup-jfrog`,
which rewrites both `registry.npmjs.org` and the dev-proxy URLs
to JFrog before `npm ci`.

With the root cause fixed and CI now green across the full
[16, 18, 20, 22, 24] matrix on both unit-test and e2e-test, the
diagnostics are no longer needed. Strip them so the workflow
stays focused on its actual responsibilities.

Co-authored-by: Isaac
Signed-off-by: Vikrant Puppala <vikrant.puppala@databricks.com>

Author: vikrantpuppala

.github/workflows/main.yml

@@ -34,69 +34,11 @@ jobs:
             ${{ runner.os }}-build-${{ env.cache-name }}-
             ${{ runner.os }}-build-
             ${{ runner.os }}-
-      # DIAGNOSTIC (temporary — remove once npm-hang root cause identified).
-      # Captures effective npm config, lockfile registry distribution, and
-      # registry reachability BEFORE npm ci runs.
-      - name: Diag — pre-npm-ci
-        run: |
-          set +e
-          echo "=== effective npm config ==="
-          npm config list -l 2>&1 | grep -E '^(registry|fetch-|cache|loglevel|prefer-|@databricks)' || true
-          echo "=== ~/.npmrc (auth masked) ==="
-          sed 's/_authToken=.*/_authToken=***/' ~/.npmrc 2>/dev/null || echo "(no ~/.npmrc)"
-          echo "=== npm cache dir ==="
-          npm config get cache
-          ls -la "$(npm config get cache)" 2>/dev/null | head -10 || echo "(empty)"
-          echo "=== package-lock.json resolved URLs by registry ==="
-          grep -oE '"resolved": "https://[^/]+' package-lock.json | sort | uniq -c
-          echo "=== reachability probes ==="
-          for url in https://registry.npmjs.org/ https://databricks.jfrog.io/artifactory/api/npm/db-npm/; do
-            echo "--- $url ---"
-            curl -sS -o /dev/null -w "HTTP=%{http_code}  connect=%{time_connect}s  total=%{time_total}s\n" \
-              --max-time 10 --connect-timeout 5 \
-              -H "Authorization: Bearer $JFROG_ACCESS_TOKEN" \
-              "$url" || echo "FAIL (curl exit $?)"
-          done
-          echo "=== sample package metadata fetch (basic-ftp, new in this PR) ==="
-          curl -sS -o /tmp/probe.json -w "basic-ftp: HTTP=%{http_code} size=%{size_download}b time=%{time_total}s\n" \
-            --max-time 10 -H "Authorization: Bearer $JFROG_ACCESS_TOKEN" \
-            "https://databricks.jfrog.io/artifactory/api/npm/db-npm/basic-ftp" || echo "FAIL"
-          head -c 200 /tmp/probe.json 2>/dev/null; echo
-          echo "=== /probe ==="
-          true
       - name: Check code style
         run: |
-          npm ci --loglevel=http --no-progress --foreground-scripts
+          npm ci
           npm run prettier
           npm run lint
-      # DIAGNOSTIC (temporary — remove once npm-hang root cause identified).
-      # Captures npm debug log, cache state, and node_modules state AFTER
-      # the failure so we can see exactly what npm did during the silent
-      # 8-minute hang.
-      - name: Diag — post-npm-ci on failure
-        if: failure()
-        run: |
-          set +e
-          DIAG=/tmp/npm-diag
-          mkdir -p "$DIAG"
-          cp -r ~/.npm/_logs "$DIAG/npm_logs" 2>/dev/null || echo "no _logs dir"
-          du -sh ~/.npm/_cacache 2>/dev/null > "$DIAG/cacache_size.txt"
-          ls -la node_modules/.bin/ 2>/dev/null > "$DIAG/node_modules_bin.txt" || echo "(no .bin)" > "$DIAG/node_modules_bin.txt"
-          ls node_modules/ 2>/dev/null | wc -l > "$DIAG/node_modules_pkg_count.txt"
-          ps auxf > "$DIAG/ps_snapshot.txt" 2>&1 || true
-          dmesg 2>&1 | tail -50 > "$DIAG/dmesg_tail.txt" || true
-          cp package-lock.json "$DIAG/package-lock.json.in-ci"
-          echo "=== diag bundle contents ==="
-          ls -la "$DIAG"
-          true
-      - name: Diag — upload bundle
-        if: failure()
-        uses: actions/upload-artifact@ea165f8d65b6e75b540449e92b4886f43607fa02 # v4
-        with:
-          name: npm-diag-lint
-          path: /tmp/npm-diag
-          if-no-files-found: warn
-          retention-days: 7
 
   unit-test:
     runs-on:
@@ -129,61 +71,10 @@ jobs:
             ${{ runner.os }}-${{ matrix.node-version }}-build-${{ env.cache-name }}-
             ${{ runner.os }}-${{ matrix.node-version }}-build-
             ${{ runner.os }}-${{ matrix.node-version }}-
-      - name: Diag — pre-npm-ci
-        run: |
-          set +e
-          echo "=== effective npm config ==="
-          npm config list -l 2>&1 | grep -E '^(registry|fetch-|cache|loglevel|prefer-|@databricks)' || true
-          echo "=== ~/.npmrc (auth masked) ==="
-          sed 's/_authToken=.*/_authToken=***/' ~/.npmrc 2>/dev/null || echo "(no ~/.npmrc)"
-          echo "=== npm cache dir ==="
-          npm config get cache
-          ls -la "$(npm config get cache)" 2>/dev/null | head -10 || echo "(empty)"
-          echo "=== package-lock.json resolved URLs by registry ==="
-          grep -oE '"resolved": "https://[^/]+' package-lock.json | sort | uniq -c
-          echo "=== reachability probes ==="
-          for url in https://registry.npmjs.org/ https://databricks.jfrog.io/artifactory/api/npm/db-npm/; do
-            echo "--- $url ---"
-            curl -sS -o /dev/null -w "HTTP=%{http_code}  connect=%{time_connect}s  total=%{time_total}s\n" \
-              --max-time 10 --connect-timeout 5 \
-              -H "Authorization: Bearer $JFROG_ACCESS_TOKEN" \
-              "$url" || echo "FAIL (curl exit $?)"
-          done
-          echo "=== sample package metadata fetch (basic-ftp, new in this PR) ==="
-          curl -sS -o /tmp/probe.json -w "basic-ftp: HTTP=%{http_code} size=%{size_download}b time=%{time_total}s\n" \
-            --max-time 10 -H "Authorization: Bearer $JFROG_ACCESS_TOKEN" \
-            "https://databricks.jfrog.io/artifactory/api/npm/db-npm/basic-ftp" || echo "FAIL"
-          head -c 200 /tmp/probe.json 2>/dev/null; echo
-          echo "=== /probe ==="
-          true
       - name: Run unit tests
         run: |
-          npm ci --loglevel=http --no-progress --foreground-scripts
+          npm ci
           npm run test
-      - name: Diag — post-npm-ci on failure
-        if: failure()
-        run: |
-          set +e
-          DIAG=/tmp/npm-diag
-          mkdir -p "$DIAG"
-          cp -r ~/.npm/_logs "$DIAG/npm_logs" 2>/dev/null || echo "no _logs dir"
-          du -sh ~/.npm/_cacache 2>/dev/null > "$DIAG/cacache_size.txt"
-          ls -la node_modules/.bin/ 2>/dev/null > "$DIAG/node_modules_bin.txt" || echo "(no .bin)" > "$DIAG/node_modules_bin.txt"
-          ls node_modules/ 2>/dev/null | wc -l > "$DIAG/node_modules_pkg_count.txt"
-          ps auxf > "$DIAG/ps_snapshot.txt" 2>&1 || true
-          dmesg 2>&1 | tail -50 > "$DIAG/dmesg_tail.txt" || true
-          cp package-lock.json "$DIAG/package-lock.json.in-ci"
-          echo "=== diag bundle contents ==="
-          ls -la "$DIAG"
-          true
-      - name: Diag — upload bundle
-        if: failure()
-        uses: actions/upload-artifact@ea165f8d65b6e75b540449e92b4886f43607fa02 # v4
-        with:
-          name: npm-diag-unit-test-node${{ matrix.node-version }}
-          path: /tmp/npm-diag
-          if-no-files-found: warn
-          retention-days: 7
       - run: tar -cvf ${{ env.NYC_REPORT_DIR }}.tar ${{ env.NYC_REPORT_DIR }}
       - name: Store coverage report
         uses: actions/upload-artifact@ea165f8d65b6e75b540449e92b4886f43607fa02 # v4
@@ -236,61 +127,10 @@ jobs:
             ${{ runner.os }}-${{ matrix.node-version }}-build-${{ env.cache-name }}-
             ${{ runner.os }}-${{ matrix.node-version }}-build-
             ${{ runner.os }}-${{ matrix.node-version }}-
-      - name: Diag — pre-npm-ci
-        run: |
-          set +e
-          echo "=== effective npm config ==="
-          npm config list -l 2>&1 | grep -E '^(registry|fetch-|cache|loglevel|prefer-|@databricks)' || true
-          echo "=== ~/.npmrc (auth masked) ==="
-          sed 's/_authToken=.*/_authToken=***/' ~/.npmrc 2>/dev/null || echo "(no ~/.npmrc)"
-          echo "=== npm cache dir ==="
-          npm config get cache
-          ls -la "$(npm config get cache)" 2>/dev/null | head -10 || echo "(empty)"
-          echo "=== package-lock.json resolved URLs by registry ==="
-          grep -oE '"resolved": "https://[^/]+' package-lock.json | sort | uniq -c
-          echo "=== reachability probes ==="
-          for url in https://registry.npmjs.org/ https://databricks.jfrog.io/artifactory/api/npm/db-npm/; do
-            echo "--- $url ---"
-            curl -sS -o /dev/null -w "HTTP=%{http_code}  connect=%{time_connect}s  total=%{time_total}s\n" \
-              --max-time 10 --connect-timeout 5 \
-              -H "Authorization: Bearer $JFROG_ACCESS_TOKEN" \
-              "$url" || echo "FAIL (curl exit $?)"
-          done
-          echo "=== sample package metadata fetch (basic-ftp, new in this PR) ==="
-          curl -sS -o /tmp/probe.json -w "basic-ftp: HTTP=%{http_code} size=%{size_download}b time=%{time_total}s\n" \
-            --max-time 10 -H "Authorization: Bearer $JFROG_ACCESS_TOKEN" \
-            "https://databricks.jfrog.io/artifactory/api/npm/db-npm/basic-ftp" || echo "FAIL"
-          head -c 200 /tmp/probe.json 2>/dev/null; echo
-          echo "=== /probe ==="
-          true
       - name: Run e2e tests
         run: |
-          npm ci --loglevel=http --no-progress --foreground-scripts
+          npm ci
           NODE_OPTIONS="--max-old-space-size=4096" npm run e2e
-      - name: Diag — post-npm-ci on failure
-        if: failure()
-        run: |
-          set +e
-          DIAG=/tmp/npm-diag
-          mkdir -p "$DIAG"
-          cp -r ~/.npm/_logs "$DIAG/npm_logs" 2>/dev/null || echo "no _logs dir"
-          du -sh ~/.npm/_cacache 2>/dev/null > "$DIAG/cacache_size.txt"
-          ls -la node_modules/.bin/ 2>/dev/null > "$DIAG/node_modules_bin.txt" || echo "(no .bin)" > "$DIAG/node_modules_bin.txt"
-          ls node_modules/ 2>/dev/null | wc -l > "$DIAG/node_modules_pkg_count.txt"
-          ps auxf > "$DIAG/ps_snapshot.txt" 2>&1 || true
-          dmesg 2>&1 | tail -50 > "$DIAG/dmesg_tail.txt" || true
-          cp package-lock.json "$DIAG/package-lock.json.in-ci"
-          echo "=== diag bundle contents ==="
-          ls -la "$DIAG"
-          true
-      - name: Diag — upload bundle
-        if: failure()
-        uses: actions/upload-artifact@ea165f8d65b6e75b540449e92b4886f43607fa02 # v4
-        with:
-          name: npm-diag-e2e-test-node${{ matrix.node-version }}
-          path: /tmp/npm-diag
-          if-no-files-found: warn
-          retention-days: 7
       - run: tar -cvf ${{ env.NYC_REPORT_DIR }}.tar ${{ env.NYC_REPORT_DIR }}
       - name: Store coverage report
         uses: actions/upload-artifact@ea165f8d65b6e75b540449e92b4886f43607fa02 # v4