proxmox/docs/archive/completion/EXPLORER_RESTORATION_COMPLETE.md

Explorer Restoration - Complete Status and Next Steps

Date: January 27, 2025
Status: 🔴 EXPLORER REQUIRES MANUAL INTERVENTION

---

📊 Current Status Summary

✅ What's Working

• Container VMID 5000: Running on node pve2
• Nginx: Running and serving frontend (HTTP 200 on direct IP)
• Ports 80 & 443: Open and accessible
• Frontend HTML: Being served correctly

❌ What's Not Working

• Blockscout Service: Not running (port 4000 not accessible)
• Nginx Proxy: Returns 502 Bad Gateway (can't connect to Blockscout)
• Public URL: Returns 404 (Cloudflare routing issue)
• API Endpoints: Not responding (depends on Blockscout)

---

🔍 Diagnostic Results

1. Container Status

• VMID: 5000
• Node: pve2
• Status: ✅ Running
• IP: 192.168.11.140

2. Service Status

• Nginx: ✅ Running (serving frontend)
• Blockscout: ❌ Not running (service inactive)
• PostgreSQL: ⚠️ Status unknown (needs verification)

3. Network Status

• Direct IP (192.168.11.140): ✅ HTTP 200 (frontend served)
• Port 4000: ❌ Not accessible (Blockscout not running)
• Public URL (explorer.d-bis.org): ❌ HTTP 404 (Cloudflare routing)

---

🛠️ Required Actions

Step 1: Access Container and Check Blockscout

On Proxmox Host:

ssh root@192.168.11.10

# Check container status
pct list | grep 5000
pct status 5000

# Enter container
pct exec 5000 -- bash

Inside Container:

# Check Blockscout service
systemctl status blockscout
journalctl -u blockscout -n 50

# Check Docker containers
docker ps -a
docker-compose -f /opt/blockscout/docker-compose.yml ps

# Check if Blockscout directory exists
ls -la /opt/blockscout/

Step 2: Start Blockscout Service

Option A: Using systemd service

pct exec 5000 -- systemctl start blockscout
pct exec 5000 -- systemctl enable blockscout
pct exec 5000 -- systemctl status blockscout

Option B: Using docker-compose

pct exec 5000 -- cd /opt/blockscout && docker-compose up -d
# OR
pct exec 5000 -- cd /opt/blockscout && docker compose up -d

Option C: Manual Docker start

pct exec 5000 -- docker ps -a | grep blockscout
# If containers exist but stopped:
pct exec 5000 -- docker start <container-name>

Step 3: Verify Blockscout is Running

Check port 4000:

# From inside container
pct exec 5000 -- ss -tlnp | grep :4000

# Test API
pct exec 5000 -- curl http://127.0.0.1:4000/api/v2/status

# From external
curl http://192.168.11.140:4000/api/v2/status

Expected Response:

{
  "success": true,
  "chain_id": 138,
  "block_number": "..."
}

Step 4: Fix Nginx Configuration (if needed)

Check Nginx config:

pct exec 5000 -- nginx -t
pct exec 5000 -- cat /etc/nginx/sites-available/blockscout

If Nginx config has errors, fix it:

# The config should proxy to http://127.0.0.1:4000
pct exec 5000 -- cat > /etc/nginx/sites-available/blockscout <<'EOF'
server {
    listen 80;
    listen [::]:80;
    server_name explorer.d-bis.org 192.168.11.140;

    location / {
        proxy_pass http://127.0.0.1:4000;
        proxy_http_version 1.1;
        proxy_set_header Host $host;
        proxy_set_header X-Real-IP $remote_addr;
        proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
        proxy_set_header X-Forwarded-Proto $scheme;
        proxy_set_header Upgrade $http_upgrade;
        proxy_set_header Connection "upgrade";
        proxy_read_timeout 300s;
        proxy_connect_timeout 75s;
    }

    location /api {
        proxy_pass http://127.0.0.1:4000/api;
        proxy_http_version 1.1;
        proxy_set_header Host $host;
        proxy_set_header X-Real-IP $remote_addr;
        proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
        proxy_set_header X-Forwarded-Proto $scheme;
    }
}
EOF

# Enable site
pct exec 5000 -- ln -sf /etc/nginx/sites-available/blockscout /etc/nginx/sites-enabled/blockscout
pct exec 5000 -- rm -f /etc/nginx/sites-enabled/default

# Test and reload
pct exec 5000 -- nginx -t
pct exec 5000 -- systemctl reload nginx

Step 5: Verify Nginx Proxy

Test from external:

curl http://192.168.11.140/api/v2/stats
curl http://192.168.11.140/api/v2/status

Should return Blockscout API responses, not 502 Bad Gateway

Step 6: Fix Cloudflare Configuration

Check Cloudflare tunnel:

# Inside container
pct exec 5000 -- systemctl status cloudflared
pct exec 5000 -- cat /etc/cloudflared/config.yml

Verify DNS record:

• Go to Cloudflare Dashboard
• Check DNS record for explorer.d-bis.org
• Should be CNAME pointing to tunnel (🟠 Proxied)

Verify tunnel route:

• Go to Cloudflare Zero Trust → Networks → Tunnels
• Check route: explorer.d-bis.org → http://192.168.11.140:80

---

📋 Verification Checklist

After completing the steps above, verify:

• Container VMID 5000 is running
• Blockscout service is active
• Port 4000 is listening
• Blockscout API responds: curl http://192.168.11.140:4000/api/v2/status
• Nginx configuration is valid: nginx -t
• Nginx proxy works: curl http://192.168.11.140/api/v2/stats (not 502)
• Cloudflare DNS record exists
• Cloudflare tunnel route configured
• Public URL works: curl https://explorer.d-bis.org/api/v2/stats

---

🔧 Troubleshooting Common Issues

Issue 1: Blockscout Service Won't Start

Check logs:

pct exec 5000 -- journalctl -u blockscout -n 100
pct exec 5000 -- docker-compose -f /opt/blockscout/docker-compose.yml logs

Common causes:

• PostgreSQL not running
• Database connection issues
• Missing environment variables
• Docker issues

Solution:

# Check PostgreSQL
pct exec 5000 -- docker ps | grep postgres
pct exec 5000 -- docker-compose -f /opt/blockscout/docker-compose.yml up -d postgres

# Check environment
pct exec 5000 -- cat /opt/blockscout/.env

# Restart all services
pct exec 5000 -- cd /opt/blockscout && docker-compose restart

Issue 2: Nginx Returns 502 Bad Gateway

Cause: Nginx can't connect to Blockscout on port 4000

Solution:

1. Ensure Blockscout is running (see Step 2)
2. Verify port 4000 is listening: ss -tlnp | grep :4000
3. Test direct connection: curl http://127.0.0.1:4000/api/v2/status
4. Check Nginx error logs: tail -f /var/log/nginx/blockscout-error.log

Issue 3: Public URL Returns 404

Cause: Cloudflare routing issue

Solution:

1. Verify DNS record in Cloudflare dashboard
2. Check tunnel configuration
3. Verify tunnel is running: systemctl status cloudflared
4. Check tunnel logs: journalctl -u cloudflared -n 50

---

📝 Scripts Created

The following diagnostic and fix scripts have been created:

1. scripts/diagnose-explorer-status.sh - Comprehensive status check
2. scripts/fix-explorer-service.sh - Automated fix attempts
3. scripts/restore-explorer-complete.sh - Complete restoration script
4. scripts/fix-nginx-blockscout-config.sh - Nginx configuration fix
5. scripts/check-blockscout-logs.sh - Blockscout logs and status check

Usage:

cd /home/intlc/projects/proxmox
./scripts/diagnose-explorer-status.sh
./scripts/check-blockscout-logs.sh

---

🎯 Priority Actions

Immediate (Required)

1. ✅ Access container VMID 5000
2. ✅ Check Blockscout service status
3. ✅ Start Blockscout service
4. ✅ Verify port 4000 is accessible

High Priority

5. ✅ Fix Nginx configuration if needed
6. ✅ Verify Nginx proxy works
7. ✅ Check Cloudflare tunnel configuration

Medium Priority

8. ⏳ Verify public URL accessibility
9. ⏳ Test all API endpoints
10. ⏳ Monitor service stability

---

📚 Related Documentation

• docs/EXPLORER_STATUS_REVIEW.md - Complete status review
• docs/BLOCKSCOUT_EXPLORER_FIX.md - Fix scripts documentation
• docs/BLOCKSCOUT_COMPREHENSIVE_ANALYSIS.md - Technical analysis
• scripts/fix-blockscout-explorer.sh - Existing fix script

---

✅ Summary

Current State: Explorer container is running, Nginx is serving frontend, but Blockscout backend service is not running.

Root Cause: Blockscout service (port 4000) is not active, causing Nginx to return 502 Bad Gateway.

Solution: Start Blockscout service using one of the methods in Step 2 above.

Next Steps: Follow the step-by-step actions above to restore full functionality.

---

Last Updated: January 27, 2025
Status: 🔴 AWAITING MANUAL INTERVENTION