Origin IP Discovery
Discover the real server IP behind CDN/WAF protections (Cloudflare, Akamai, Fastly). When the origin IP is found, the raw server is exposed without firewall rules, rate limiting, or application-layer filtering. Techniques include favicon hash fingerprinting across Shodan, historical DNS records from passive sources, SSL certificate SAN field matching, and Google Analytics ID cross-referencing.
When to Use
- Target is behind Cloudflare/Akamai and returns 403 or CAPTCHA challenges on all requests.
- You need direct access to the origin to bypass WAF rules.
- Subdomain enumeration reveals internal/staging hosts on non-CDN IPs.
- The target uses a single favicon across all infrastructure.
- SSL certificates share the same organization name across IPs.
Prerequisites
terminalwith curl, python3, and shodan CLI.- Shodan API key:
shodan init <KEY>. - Target favicon file or URL.
Quick Detection
# Check Cloudflare presence
curl --max-time 30 --connect-timeout 10 -sI "https://target.com" | grep -i "cf-ray\|server: cloudflare"
# If Cloudflare detected, check for common origin leaks
curl --max-time 30 --connect-timeout 10 -sk "https://target.com/cdn-cgi/trace" | grep -E "ip=|colo="
Procedure
Phase 1 — Favicon Hash Fingerprinting
# Get favicon hash
FAVICON_URL="https://target.com/favicon.ico"
curl --max-time 30 --connect-timeout 10 -sk "$FAVICON_URL" -o favicon_target.ico
# Calculate hash
python3 -c "
import hashlib, base64
with open('favicon_target.ico', 'rb') as f:
hash_bytes = base64.b64encode(hashlib.md5(f.read()).digest())
print(f'favicon hash: {hash_bytes.decode()}')
"
# Search Shodan for IPs serving this favicon
HASH=$(python3 -c "
import hashlib, base64
with open('favicon_target.ico','rb') as f:
print(base64.b64encode(hashlib.md5(f.read()).digest()).decode())
")
shodan search "http.favicon.hash:$HASH" --fields ip_str,port,org,hostnames
# Automatedtools
python3 favUp.py -ff favicon_target.ico --shodan-cli
python3 favUp.py --web target.com -sc
Phase 2 — Historical DNS Records
# SecurityTrails — DNS history
curl --max-time 30 --connect-timeout 10 -s "https://securitytrails.com/domain/target.com/history/a" \
-H "APIKEY: $SECURITYTRAILS_KEY" \
| jq '.records[].values[].ip' | sort -u
# AlienVault OTX — passive DNS
curl --max-time 30 --connect-timeout 10 -s "https://otx.alienvault.com/api/v1/indicators/domain/target.com/passive_dns" \
| jq '.passive_dns[].address' | sort -u
# Automatedtools
echo "target.com" | originiphunter
cat domains.txt | originiphunter
Phase 3 — SSL Certificate Search
# crt.sh — find all certs for the domain, extract unique IPs from SAN fields
curl --max-time 30 --connect-timeout 10 -s "https://crt.sh/?q=%25.target.com&output=json" \
| jq -r '.[].name_value' | tr ',' '\n' | sort -u > cert_domains.txt
# Censys — search by parsed names
# Web: https://search.censys.io — query: parsed.names: target.com
# Shodan — search by SSL subject CN
shodan search "ssl.cert.subject.cn:target.com" --fields ip_str,port,org
# Netlas — deep infrastructure search
# Web: https://netlas.io — query: domain:*.target.com
Phase 4 — Google Analytics ID Cross-Referencing
# Extract Analytics ID from target pages
curl --max-time 30 --connect-timeout 10 -s "https://target.com" | grep -Eo '(?:UA-|G-|GTM-)[A-Z0-9]+'
# Find all domains sharing the same GA ID
curl --max-time 30 --connect-timeout 10 -s "https://api.hackertarget.com/analyticslookup/?q=UA-XXXXXXXX-X"
curl --max-time 30 --connect-timeout 10 -s "https://builtwith.com/relationships/target.com"
# Automated
cat subdomains.txt | analyticsrelationships
Phase 5 — Common Origin Leak Vectors
# Direct IP in page source
curl --max-time 30 --connect-timeout 10 -sk "https://target.com" | grep -Eo '\b(?:\d{1,3}\.){3}\d{1,3}\b' | sort -u
# DNS CNAME chain — may expose origin
dig target.com ANY +noall +answer
dig www.target.com CNAME +short
# MX records — mail server often shares infrastructure
dig target.com MX +short
# SPF records — may contain non-CDN IPs
dig target.com TXT +short | grep -Eo '\b(?:\d{1,3}\.){3}\d{1,3}\b'
# Subdomain with different IP pattern
subfinder -d target.com -silent | dnsx -silent -a -resp-only | sort -u > all_ips.txt
# Filter out Cloudflare IPs (104.x, 172.64-71.x)
grep -vE '^104\.(1[6-9]|2[0-9]|3[0-1])\.|^172\.(6[4-9]|7[0-1])\.' all_ips.txt > non_cf_ips.txt
# Probe each for the target's content
for ip in $(cat non_cf_ips.txt); do
curl --max-time 30 -sk --connect-timeout 5 "https://$ip" -H "Host: target.com" -o /dev/null -w "%{http_code} $ip\n"
done | grep -v "^403\|^000"
Phase 6 — Validate Origin
# Check if IP serves the target's content
curl --max-time 30 --connect-timeout 10 -sk "https://ORIGIN_IP" -H "Host: target.com" | grep -o '<title>[^<]*</title>'
# Verify favicon matches
curl --max-time 30 --connect-timeout 10 -sk "https://ORIGIN_IP/favicon.ico" -H "Host: target.com" -o favicon_origin.ico
md5sum favicon_target.ico favicon_origin.ico # should match
# Probe ports directly on origin
naabu -host ORIGIN_IP -p - -rate 2000
Pitfalls
- Cloudflare may block your IP on repeated scans. Rotate user-agents and proxies.
- Shodan queries require an API key. Free tier is rate-limited.
- The origin may also be behind firewall rules. Even if you find the IP, it may only accept traffic from Cloudflare's IP ranges.
- Google Analytics IDs are shared across unrelated sites. Verify by favicon or content match before claiming a finding.
- Non-CDN IPs from subdomains may be load balancers, not the actual web server origin. Probe with Host header to confirm.
Verification
- Favicon hash matches between CDN-cached site and direct origin IP.
- HTTPS request to origin IP with
Host: target.comreturns the same page content. - Open ports on origin differ from CDN-proxied ports (origin exposes SSH, MySQL, etc.).
- SSL certificate on the origin IP lists the target domain in SAN.
Related Skills
subdomain-enumeration— Generate the subdomain list needed for IP deduplication.vhost-enumeration— Once origin IP is found, enumerate virtual hosts.port-mass-scan— Scan origin IP for exposed services.