feat: Add real-time Service Status tiles for active viewers and packe… #85
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: Deploy to Fly.io | |
| on: | |
| push: | |
| branches: [ main ] | |
| paths: | |
| - 'fly/**' | |
| - '.github/workflows/deploy-fly.yml' | |
| workflow_dispatch: | |
| inputs: | |
| service: | |
| description: 'Service to deploy (all, web, grafana, influxdb, eagle-monitor)' | |
| required: false | |
| default: 'all' | |
| type: choice | |
| options: | |
| - all | |
| - web | |
| - grafana | |
| - influxdb | |
| - eagle-monitor | |
| permissions: | |
| contents: read | |
| env: | |
| FLY_API_TOKEN: ${{ secrets.FLY_API_TOKEN }} | |
| # Concurrency control to prevent multiple deployments | |
| concurrency: | |
| group: deploy-${{ github.ref }}-${{ github.event.inputs.service || 'all' }} | |
| cancel-in-progress: false # Don't cancel deployments in progress | |
| jobs: | |
| # Monitor queue time before main job | |
| queue-monitor: | |
| name: Queue Time Monitor | |
| uses: ./.github/workflows/queue-monitor-reusable.yml | |
| deploy: | |
| runs-on: ubuntu-latest | |
| needs: [queue-monitor] # Run after queue monitor | |
| timeout-minutes: 30 # Total timeout including queue time | |
| steps: | |
| - name: Checkout code | |
| uses: actions/checkout@v4 | |
| - name: Detect changed services | |
| uses: dorny/paths-filter@v3 | |
| id: changes | |
| with: | |
| filters: | | |
| web: | |
| - 'fly/web/**' | |
| influxdb: | |
| - 'fly/influxdb/**' | |
| grafana: | |
| - 'fly/grafana/**' | |
| eagle: | |
| - 'fly/eagle-monitor/**' | |
| workflow: | |
| - '.github/workflows/deploy-fly.yml' | |
| - name: Setup Fly CLI | |
| uses: superfly/flyctl-actions/setup-flyctl@master | |
| - name: Verify Fly.io Authentication | |
| run: | | |
| echo "Checking Fly.io authentication..." | |
| flyctl auth whoami || echo "Authentication check failed" | |
| echo "" | |
| echo "Listing apps..." | |
| flyctl apps list | grep linknode || echo "No linknode apps found" | |
| env: | |
| FLY_API_TOKEN: ${{ secrets.FLY_API_TOKEN }} | |
| - name: Capture Current Versions | |
| id: versions | |
| run: | | |
| echo "Capturing current deployment versions for potential rollback..." | |
| # Function to safely get version with error handling | |
| get_version() { | |
| local app=$1 | |
| local version | |
| if ! version=$(flyctl image show -a "$app" -j 2>&1); then | |
| echo "::warning::Failed to get version for $app" >&2 | |
| echo "none" | |
| return 0 # Don't fail the workflow | |
| fi | |
| version=$(echo "$version" | jq -r .Tag 2>/dev/null) | |
| if [ -z "$version" ] || [ "$version" = "null" ]; then | |
| echo "::warning::No version tag found for $app" >&2 | |
| echo "none" | |
| return 0 # Don't fail the workflow | |
| fi | |
| echo "$version" | |
| return 0 | |
| } | |
| # Capture versions with validation | |
| INFLUXDB_VERSION=$(get_version linknode-influxdb) | |
| EAGLE_VERSION=$(get_version linknode-eagle-monitor) | |
| GRAFANA_VERSION=$(get_version linknode-grafana) | |
| WEB_VERSION=$(get_version linknode-web) | |
| # Store versions in outputs | |
| echo "influxdb_version=$INFLUXDB_VERSION" >> $GITHUB_OUTPUT | |
| echo "eagle_version=$EAGLE_VERSION" >> $GITHUB_OUTPUT | |
| echo "grafana_version=$GRAFANA_VERSION" >> $GITHUB_OUTPUT | |
| echo "web_version=$WEB_VERSION" >> $GITHUB_OUTPUT | |
| echo "Current versions:" | |
| echo "- InfluxDB: $INFLUXDB_VERSION" | |
| echo "- Eagle Monitor: $EAGLE_VERSION" | |
| echo "- Grafana: $GRAFANA_VERSION" | |
| echo "- Web: $WEB_VERSION" | |
| # Warn if no versions were captured successfully | |
| if [ "$INFLUXDB_VERSION" = "none" ] && [ "$EAGLE_VERSION" = "none" ] && | |
| [ "$GRAFANA_VERSION" = "none" ] && [ "$WEB_VERSION" = "none" ]; then | |
| echo "::warning::No deployment versions could be captured - rollback may not be possible" | |
| fi | |
| env: | |
| FLY_API_TOKEN: ${{ secrets.FLY_API_TOKEN }} | |
| - name: Deploy InfluxDB | |
| id: deploy-influxdb | |
| if: | | |
| (github.event_name == 'push' && steps.changes.outputs.influxdb == 'true') || | |
| (github.event_name == 'workflow_dispatch' && (github.event.inputs.service == 'all' || github.event.inputs.service == 'influxdb')) | |
| run: | | |
| cd fly/influxdb | |
| # Deploy with retry logic | |
| RETRY_COUNT=0 | |
| MAX_RETRIES=2 | |
| while [ $RETRY_COUNT -le $MAX_RETRIES ]; do | |
| echo "Deployment attempt $((RETRY_COUNT + 1)) of $((MAX_RETRIES + 1))..." | |
| if flyctl deploy --remote-only --wait-timeout 300; then | |
| echo "✅ InfluxDB deployed successfully" | |
| # Verify deployment health | |
| sleep 15 # Give InfluxDB more time to start | |
| # Check InfluxDB health - Fly.io may handle port routing differently | |
| echo "Checking InfluxDB health..." | |
| # Primary check: Use flyctl status which is more reliable | |
| if flyctl status -a linknode-influxdb --json 2>/dev/null | jq -r '.Allocations[0].Status' 2>/dev/null | grep -q "running"; then | |
| echo "✅ InfluxDB is running (verified via flyctl)" | |
| # Give it a bit more time to be fully ready | |
| sleep 5 | |
| break | |
| else | |
| echo "⚠️ InfluxDB not yet healthy" | |
| # Try HTTP endpoint without port (Fly.io handles routing) | |
| if curl -sf https://linknode-influxdb.fly.dev/ping -m 10 2>/dev/null; then | |
| echo "✅ InfluxDB health check passed (HTTPS)" | |
| break | |
| elif curl -sf http://linknode-influxdb.fly.dev/ping -m 10 2>/dev/null; then | |
| echo "✅ InfluxDB health check passed (HTTP)" | |
| break | |
| fi | |
| if [ $RETRY_COUNT -eq $MAX_RETRIES ]; then | |
| echo "❌ Maximum retries reached" | |
| echo "Debug: Checking app status..." | |
| flyctl status -a linknode-influxdb || true | |
| exit 1 | |
| fi | |
| fi | |
| else | |
| echo "❌ Deployment failed" | |
| if [ $RETRY_COUNT -eq $MAX_RETRIES ]; then | |
| echo "🔄 Attempting rollback to version ${{ steps.versions.outputs.influxdb_version }}" | |
| if [ "${{ steps.versions.outputs.influxdb_version }}" != "none" ]; then | |
| flyctl deploy --image "${{ steps.versions.outputs.influxdb_version }}" -a linknode-influxdb --yes || true | |
| fi | |
| exit 1 | |
| fi | |
| fi | |
| RETRY_COUNT=$((RETRY_COUNT + 1)) | |
| [ $RETRY_COUNT -le $MAX_RETRIES ] && echo "Waiting 30 seconds before retry..." && sleep 30 | |
| done | |
| env: | |
| FLY_API_TOKEN: ${{ secrets.FLY_API_TOKEN }} | |
| continue-on-error: false | |
| - name: Deploy Eagle Monitor | |
| id: deploy-eagle | |
| if: | | |
| (github.event_name == 'push' && steps.changes.outputs.eagle == 'true') || | |
| (github.event_name == 'workflow_dispatch' && (github.event.inputs.service == 'all' || github.event.inputs.service == 'eagle-monitor')) | |
| run: | | |
| cd fly/eagle-monitor | |
| # Deploy with retry logic | |
| RETRY_COUNT=0 | |
| MAX_RETRIES=2 | |
| while [ $RETRY_COUNT -le $MAX_RETRIES ]; do | |
| echo "Deployment attempt $((RETRY_COUNT + 1)) of $((MAX_RETRIES + 1))..." | |
| if flyctl deploy --remote-only --wait-timeout 300; then | |
| echo "✅ Eagle Monitor deployed successfully" | |
| # Verify deployment health | |
| sleep 10 | |
| if curl -sf https://linknode-eagle-monitor.fly.dev/health -m 10; then | |
| echo "✅ Eagle Monitor health check passed" | |
| break | |
| else | |
| echo "⚠️ Eagle Monitor health check failed" | |
| if [ $RETRY_COUNT -eq $MAX_RETRIES ]; then | |
| echo "❌ Maximum retries reached" | |
| exit 1 | |
| fi | |
| fi | |
| else | |
| echo "❌ Deployment failed" | |
| if [ $RETRY_COUNT -eq $MAX_RETRIES ]; then | |
| echo "🔄 Attempting rollback to version ${{ steps.versions.outputs.eagle_version }}" | |
| if [ "${{ steps.versions.outputs.eagle_version }}" != "none" ]; then | |
| flyctl deploy --image "${{ steps.versions.outputs.eagle_version }}" -a linknode-eagle-monitor --yes || true | |
| fi | |
| exit 1 | |
| fi | |
| fi | |
| RETRY_COUNT=$((RETRY_COUNT + 1)) | |
| [ $RETRY_COUNT -le $MAX_RETRIES ] && echo "Waiting 30 seconds before retry..." && sleep 30 | |
| done | |
| env: | |
| FLY_API_TOKEN: ${{ secrets.FLY_API_TOKEN }} | |
| continue-on-error: false | |
| - name: Deploy Grafana | |
| id: deploy-grafana | |
| if: | | |
| (github.event_name == 'push' && steps.changes.outputs.grafana == 'true') || | |
| (github.event_name == 'workflow_dispatch' && (github.event.inputs.service == 'all' || github.event.inputs.service == 'grafana')) | |
| run: | | |
| cd fly/grafana | |
| # Deploy with retry logic | |
| RETRY_COUNT=0 | |
| MAX_RETRIES=2 | |
| while [ $RETRY_COUNT -le $MAX_RETRIES ]; do | |
| echo "Deployment attempt $((RETRY_COUNT + 1)) of $((MAX_RETRIES + 1))..." | |
| if flyctl deploy --remote-only --wait-timeout 300; then | |
| echo "✅ Grafana deployed successfully" | |
| # Verify deployment health | |
| sleep 10 | |
| if curl -sf https://linknode-grafana.fly.dev/api/health -m 10; then | |
| echo "✅ Grafana health check passed" | |
| break | |
| else | |
| echo "⚠️ Grafana health check failed" | |
| if [ $RETRY_COUNT -eq $MAX_RETRIES ]; then | |
| echo "❌ Maximum retries reached" | |
| exit 1 | |
| fi | |
| fi | |
| else | |
| echo "❌ Deployment failed" | |
| if [ $RETRY_COUNT -eq $MAX_RETRIES ]; then | |
| echo "🔄 Attempting rollback to version ${{ steps.versions.outputs.grafana_version }}" | |
| if [ "${{ steps.versions.outputs.grafana_version }}" != "none" ]; then | |
| flyctl deploy --image "${{ steps.versions.outputs.grafana_version }}" -a linknode-grafana --yes || true | |
| fi | |
| exit 1 | |
| fi | |
| fi | |
| RETRY_COUNT=$((RETRY_COUNT + 1)) | |
| [ $RETRY_COUNT -le $MAX_RETRIES ] && echo "Waiting 30 seconds before retry..." && sleep 30 | |
| done | |
| env: | |
| FLY_API_TOKEN: ${{ secrets.FLY_API_TOKEN }} | |
| continue-on-error: false | |
| - name: Deploy Web Interface | |
| id: deploy-web | |
| if: | | |
| (github.event_name == 'push' && steps.changes.outputs.web == 'true') || | |
| (github.event_name == 'workflow_dispatch' && (github.event.inputs.service == 'all' || github.event.inputs.service == 'web')) | |
| run: | | |
| cd fly/web | |
| # Generate build info | |
| echo "Generating build-info.json..." | |
| cat > build-info.json << EOF | |
| { | |
| "version": "v1.0.${{ github.run_number }}", | |
| "buildDate": "$(date -u +%Y-%m-%d)", | |
| "commit": "$(echo ${{ github.sha }} | cut -c1-7)", | |
| "environment": "production", | |
| "branch": "${{ github.ref_name }}", | |
| "buildNumber": "${{ github.run_number }}", | |
| "actor": "${{ github.actor }}" | |
| } | |
| EOF | |
| echo "Build info:" | |
| cat build-info.json | |
| # Deploy with retry logic | |
| RETRY_COUNT=0 | |
| MAX_RETRIES=2 | |
| while [ $RETRY_COUNT -le $MAX_RETRIES ]; do | |
| echo "Deployment attempt $((RETRY_COUNT + 1)) of $((MAX_RETRIES + 1))..." | |
| if flyctl deploy --remote-only --wait-timeout 300; then | |
| echo "✅ Web interface deployed successfully" | |
| # Verify deployment health | |
| sleep 10 | |
| if curl -sf https://linknode-web.fly.dev -m 10 | grep -q "Linknode"; then | |
| echo "✅ Web interface health check passed" | |
| break | |
| else | |
| echo "⚠️ Web interface health check failed" | |
| if [ $RETRY_COUNT -eq $MAX_RETRIES ]; then | |
| echo "❌ Maximum retries reached" | |
| exit 1 | |
| fi | |
| fi | |
| else | |
| echo "❌ Deployment failed" | |
| if [ $RETRY_COUNT -eq $MAX_RETRIES ]; then | |
| echo "🔄 Attempting rollback to version ${{ steps.versions.outputs.web_version }}" | |
| if [ "${{ steps.versions.outputs.web_version }}" != "none" ]; then | |
| flyctl deploy --image "${{ steps.versions.outputs.web_version }}" -a linknode-web --yes || true | |
| fi | |
| exit 1 | |
| fi | |
| fi | |
| RETRY_COUNT=$((RETRY_COUNT + 1)) | |
| [ $RETRY_COUNT -le $MAX_RETRIES ] && echo "Waiting 30 seconds before retry..." && sleep 30 | |
| done | |
| env: | |
| FLY_API_TOKEN: ${{ secrets.FLY_API_TOKEN }} | |
| continue-on-error: false | |
| - name: Verify Deployments | |
| if: always() | |
| run: | | |
| echo "=== Deployment Status ===" | |
| flyctl status -a linknode-influxdb || echo "InfluxDB: Failed to get status" | |
| echo "" | |
| flyctl status -a linknode-eagle-monitor || echo "Eagle Monitor: Failed to get status" | |
| echo "" | |
| flyctl status -a linknode-grafana || echo "Grafana: Failed to get status" | |
| echo "" | |
| flyctl status -a linknode-web || echo "Web: Failed to get status" | |
| env: | |
| FLY_API_TOKEN: ${{ secrets.FLY_API_TOKEN }} | |
| - name: Create deployment summary | |
| if: always() | |
| run: | | |
| echo "## 🚀 Fly.io Deployment Status" >> $GITHUB_STEP_SUMMARY | |
| echo "" >> $GITHUB_STEP_SUMMARY | |
| # Show which services were deployed | |
| echo "### Deployed Services" >> $GITHUB_STEP_SUMMARY | |
| if [ "${{ github.event_name }}" = "workflow_dispatch" ]; then | |
| echo "- **Mode**: Manual deployment (${{ github.event.inputs.service }})" >> $GITHUB_STEP_SUMMARY | |
| else | |
| echo "- **Mode**: Automatic deployment based on changed files" >> $GITHUB_STEP_SUMMARY | |
| echo "" >> $GITHUB_STEP_SUMMARY | |
| echo "**Services deployed:**" >> $GITHUB_STEP_SUMMARY | |
| [ "${{ steps.changes.outputs.web }}" = "true" ] && echo "- ✅ Web Interface" >> $GITHUB_STEP_SUMMARY | |
| [ "${{ steps.changes.outputs.grafana }}" = "true" ] && echo "- ✅ Grafana" >> $GITHUB_STEP_SUMMARY | |
| [ "${{ steps.changes.outputs.eagle }}" = "true" ] && echo "- ✅ Eagle Monitor" >> $GITHUB_STEP_SUMMARY | |
| [ "${{ steps.changes.outputs.influxdb }}" = "true" ] && echo "- ✅ InfluxDB" >> $GITHUB_STEP_SUMMARY | |
| fi | |
| echo "" >> $GITHUB_STEP_SUMMARY | |
| echo "### Public Access URLs" >> $GITHUB_STEP_SUMMARY | |
| echo "- **Main Site**: https://linknode.com" >> $GITHUB_STEP_SUMMARY | |
| echo "- **Direct Fly URL**: https://linknode-web.fly.dev" >> $GITHUB_STEP_SUMMARY | |
| echo "- **Grafana Dashboard**: https://linknode-grafana.fly.dev" >> $GITHUB_STEP_SUMMARY | |
| echo "" >> $GITHUB_STEP_SUMMARY | |
| echo "### Service Endpoints" >> $GITHUB_STEP_SUMMARY | |
| echo "- **Eagle Monitor API**: https://linknode-eagle-monitor.fly.dev/api/stats" >> $GITHUB_STEP_SUMMARY | |
| echo "- **Health Check**: https://linknode-eagle-monitor.fly.dev/health" >> $GITHUB_STEP_SUMMARY | |
| echo "" >> $GITHUB_STEP_SUMMARY | |
| echo "### Deployment Details" >> $GITHUB_STEP_SUMMARY | |
| echo "- **Commit**: ${{ github.sha }}" >> $GITHUB_STEP_SUMMARY | |
| echo "- **Branch**: ${{ github.ref_name }}" >> $GITHUB_STEP_SUMMARY | |
| echo "- **Time**: $(date -u +%Y-%m-%dT%H:%M:%SZ)" >> $GITHUB_STEP_SUMMARY | |
| echo "- **Triggered by**: ${{ github.actor }}" >> $GITHUB_STEP_SUMMARY | |
| - name: Final Health Check | |
| if: success() | |
| run: | | |
| echo "Performing final health checks for deployed services..." | |
| FAILED_SERVICES="" | |
| # Function to check if service was deployed | |
| was_deployed() { | |
| local service=$1 | |
| if [ "${{ github.event_name }}" = "workflow_dispatch" ]; then | |
| [ "${{ github.event.inputs.service }}" = "all" ] || [ "${{ github.event.inputs.service }}" = "$service" ] | |
| else | |
| # For push events, check if service files changed | |
| case $service in | |
| web) [ "${{ steps.changes.outputs.web }}" = "true" ] ;; | |
| grafana) [ "${{ steps.changes.outputs.grafana }}" = "true" ] ;; | |
| eagle-monitor) [ "${{ steps.changes.outputs.eagle }}" = "true" ] ;; | |
| influxdb) [ "${{ steps.changes.outputs.influxdb }}" = "true" ] ;; | |
| esac | |
| fi | |
| } | |
| # Check only deployed services | |
| if was_deployed "web"; then | |
| echo "🔍 Checking Web Interface..." | |
| if curl -sf https://linknode-web.fly.dev -m 10 | grep -q "Linknode"; then | |
| echo "✅ Web interface is healthy" | |
| else | |
| echo "❌ Web interface is not responding correctly" | |
| FAILED_SERVICES="${FAILED_SERVICES}web " | |
| fi | |
| fi | |
| if was_deployed "grafana"; then | |
| echo "🔍 Checking Grafana..." | |
| if curl -sf https://linknode-grafana.fly.dev/api/health -m 10; then | |
| echo "✅ Grafana is healthy" | |
| else | |
| echo "❌ Grafana is not responding" | |
| FAILED_SERVICES="${FAILED_SERVICES}grafana " | |
| fi | |
| fi | |
| if was_deployed "eagle-monitor"; then | |
| echo "🔍 Checking Eagle Monitor..." | |
| if curl -sf https://linknode-eagle-monitor.fly.dev/health -m 10; then | |
| echo "✅ Eagle Monitor is healthy" | |
| else | |
| echo "❌ Eagle Monitor is not responding" | |
| FAILED_SERVICES="${FAILED_SERVICES}eagle-monitor " | |
| fi | |
| fi | |
| if was_deployed "influxdb"; then | |
| echo "🔍 Checking InfluxDB..." | |
| # InfluxDB health check - try multiple approaches | |
| if flyctl status -a linknode-influxdb --json 2>/dev/null | jq -r '.Allocations[0].Status' 2>/dev/null | grep -q "running"; then | |
| echo "✅ InfluxDB is healthy (verified via flyctl)" | |
| elif curl -sf https://linknode-influxdb.fly.dev/ping -m 10 2>/dev/null; then | |
| echo "✅ InfluxDB is healthy (HTTPS)" | |
| elif curl -sf http://linknode-influxdb.fly.dev/ping -m 10 2>/dev/null; then | |
| echo "✅ InfluxDB is healthy (HTTP)" | |
| else | |
| echo "❌ InfluxDB is not responding" | |
| FAILED_SERVICES="${FAILED_SERVICES}influxdb " | |
| fi | |
| fi | |
| # Exit with error if any services failed | |
| if [ -n "$FAILED_SERVICES" ]; then | |
| echo "⚠️ Failed services: $FAILED_SERVICES" | |
| echo "Consider manual intervention or rollback" | |
| exit 1 | |
| else | |
| echo "✅ All services are healthy!" | |
| fi | |
| # Slack Notifications via GitHub Native Integration | |
| # This workflow is configured to send notifications through GitHub's native Slack integration. | |
| # | |
| # Setup completed in #linknode-com channel: | |
| # - GitHub Slack app installed: https://slack.com/apps/A01BP7R4KNY-github | |
| # - Subscribed to: /github subscribe murr2k/linknode-com workflows | |
| # | |
| # Notifications are sent automatically for: | |
| # - Workflow failures | |
| # - Deployment completions | |
| # - Status changes | |
| # | |
| # No secrets or webhook configuration needed! |