Monitor Google Indexing Status #58
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| name: Monitor Google Indexing Status | |
| on: | |
| schedule: | |
| - cron: '0 8 * * 1,4' # Fixed: Run twice weekly (Monday and Thursday at 8:00 UTC) | |
| workflow_dispatch: # Allow manual triggering | |
| inputs: | |
| email: | |
| description: 'Send email notification?' | |
| required: false | |
| default: 'true' | |
| type: choice | |
| options: | |
| - 'true' | |
| - 'false' | |
| jobs: | |
| check-indexing: | |
| runs-on: ubuntu-latest | |
| steps: | |
| - name: Checkout code | |
| uses: actions/checkout@v4 | |
| - name: Set up Python | |
| uses: actions/setup-python@v4 | |
| with: | |
| python-version: '3.11' | |
| - name: Install dependencies | |
| run: | | |
| python -m pip install --upgrade pip | |
| pip install google-api-python-client google-auth-oauthlib google-auth-httplib2 jq | |
| - name: Write credentials to file | |
| run: | | |
| echo '${{ secrets.GOOGLE_SERVICE_ACCOUNT }}' > service-account.json | |
| - name: Validate service account JSON | |
| run: | | |
| # Check if file exists | |
| if [ ! -f "service-account.json" ]; then | |
| echo "::error::service-account.json file not found" | |
| exit 1 | |
| fi | |
| # Validate JSON structure | |
| if ! jq empty service-account.json 2>/dev/null; then | |
| echo "::error::Invalid JSON format in service-account.json" | |
| exit 1 | |
| fi | |
| # Check required fields for Google service account | |
| if [ -z "$(jq -r '.client_email' service-account.json 2>/dev/null)" ] || \ | |
| [ -z "$(jq -r '.private_key' service-account.json 2>/dev/null)" ]; then | |
| echo "::error::Missing required fields in service-account.json" | |
| exit 1 | |
| fi | |
| echo "Service account JSON validated successfully" | |
| - name: Check indexing status and send email | |
| run: | | |
| # Determine if we should send email | |
| SHOULD_SEND_EMAIL="false" | |
| # Send email for scheduled runs (always) | |
| if [ "${{ github.event_name }}" == "schedule" ]; then | |
| SHOULD_SEND_EMAIL="true" | |
| echo "📅 Scheduled run - email will be sent" | |
| fi | |
| # Send email for manual runs if requested | |
| if [ "${{ github.event_name }}" == "workflow_dispatch" ]; then | |
| if [ "${{ github.event.inputs.email }}" == "true" ]; then | |
| SHOULD_SEND_EMAIL="true" | |
| echo "📧 Manual run with email requested" | |
| else | |
| echo "🔇 Manual run without email" | |
| fi | |
| fi | |
| # Run the monitoring script with appropriate flags | |
| if [ "$SHOULD_SEND_EMAIL" == "true" ]; then | |
| echo "Running indexing check with email notification..." | |
| python .github/scripts/monitor_indexing_status.py \ | |
| --site https://sednabcn.github.io/ \ | |
| --output indexing_status.json \ | |
| --email ${{ secrets.NOTIFICATION_EMAIL }} | |
| else | |
| echo "Running indexing check without email..." | |
| python .github/scripts/monitor_indexing_status.py \ | |
| --site https://sednabcn.github.io/ \ | |
| --output indexing_status.json | |
| fi | |
| env: | |
| EMAIL_FROM: ${{ secrets.EMAIL_FROM }} | |
| EMAIL_PASSWORD: ${{ secrets.EMAIL_PASSWORD }} | |
| SMTP_SERVER: smtp.gmail.com | |
| SMTP_PORT: 587 | |
| GOOGLE_APPLICATION_CREDENTIALS: service-account.json | |
| - name: Display report summary | |
| if: always() | |
| run: | | |
| if [ -f indexing_status.json ]; then | |
| echo "=== Indexing Status Report Summary ===" | |
| echo "Report generated at: $(date)" | |
| # Extract key metrics if available | |
| if command -v jq >/dev/null 2>&1; then | |
| TOTAL_PAGES=$(jq -r '.total_pages // "N/A"' indexing_status.json 2>/dev/null) | |
| INDEXED_PAGES=$(jq -r '.indexed_pages // "N/A"' indexing_status.json 2>/dev/null) | |
| CRAWL_ERRORS=$(jq -r '.crawl_errors // "N/A"' indexing_status.json 2>/dev/null) | |
| echo "Total pages: $TOTAL_PAGES" | |
| echo "Indexed pages: $INDEXED_PAGES" | |
| echo "Crawl errors: $CRAWL_ERRORS" | |
| else | |
| echo "Report content:" | |
| cat indexing_status.json | |
| fi | |
| else | |
| echo "❌ No indexing status report generated" | |
| fi | |
| - name: Upload indexing status report | |
| uses: actions/upload-artifact@v4 | |
| if: always() | |
| with: | |
| name: indexing-status-report-${{ github.run_number }} | |
| path: indexing_status.json | |
| retention-days: 90 | |
| - name: Create issue on critical errors | |
| if: always() | |
| run: | | |
| if [ -f indexing_status.json ]; then | |
| if ! ERRORS=$(jq '.crawl_errors // 0' indexing_status.json 2>/dev/null); then | |
| echo "::warning::Failed to parse indexing_status.json with jq" | |
| exit 0 | |
| fi | |
| # Check if ERRORS is a number and greater than threshold | |
| if [[ "$ERRORS" =~ ^[0-9]+$ ]] && [ "$ERRORS" -gt 5 ]; then | |
| echo "Found $ERRORS indexing errors (threshold: 5), creating GitHub issue" | |
| # Create issue with detailed information | |
| ISSUE_BODY="🚨 **Google Indexing Errors Detected** | |
| **Summary:** | |
| - Crawl errors detected: $ERRORS | |
| - Threshold exceeded: 5 | |
| - Workflow run: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }} | |
| **Next Steps:** | |
| 1. Review the indexing status report in the workflow artifacts | |
| 2. Check Google Search Console for detailed error information | |
| 3. Fix any identified issues with your site | |
| 4. Monitor the next scheduled run | |
| **Auto-generated by:** Google Indexing Monitor | |
| **Timestamp:** $(date -u)" | |
| gh issue create \ | |
| --title "🚨 Google Indexing Errors: $ERRORS errors detected" \ | |
| --body "$ISSUE_BODY" \ | |
| --label "indexing-error,automated" | |
| elif [[ "$ERRORS" =~ ^[0-9]+$ ]] && [ "$ERRORS" -gt 0 ]; then | |
| echo "Found $ERRORS indexing errors (below threshold of 5)" | |
| else | |
| echo "No significant indexing errors found" | |
| fi | |
| else | |
| echo "::warning::indexing_status.json file not found - cannot check for errors" | |
| fi | |
| env: | |
| GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }} |