
ABUSE.MOM — BEHAVE OR GET EXPOSED
| Signature | Description | Points | Severity |
|---|---|---|---|
| UA suspicious (short/empty) | Behavioral anomaly detected by automated analysis | +15 | |
| Danger strong hits: 2 | High-risk paths: shells, RCE vectors, exploits | +50 | |
| Danger medium hits: 1 | Medium-risk: admin panels, config files | +10 | |
| 404 ratio 40-60% | Majority of requests returned 404 — enumeration | +15 |
Reconstructed HTTP requests from server access logs. Target domains redacted for security.
* Typical request patterns for detected signatures. Actual target domains are redacted.
Address UA spoofing from 104.234.53.50: maintain blocklist of known malicious UA strings, require consistent UA across sessions, implement TLS fingerprinting.
IP 104.234.53.50 is enumerating directories. Configure fail2ban apache-404 jail after 10+ 404 errors. Disable directory listings. Normalize all 404 responses.
Other blocked IPs from the same /24 subnet — indicates systematic abuse from this network range.
This IP was checked against major DNS-based blacklists used by mail servers and firewalls worldwide.
Checked: Spamhaus, SpamCop, Barracuda, SORBS, CBL, UCEProtect. Results may change over time.
104.234.53.50 has been assigned a threat score of 90/100 (Critical). With this rating, the IP falls into the critical severity bracket — among the most dangerous addresses in our monitoring database.
The following attack categories were identified:
The address 104.234.53.50 originates from New York, United States, operating on the network of Velcom. It was identified through automated analysis of incoming network traffic across monitored endpoints. Over a period of 54 days, this IP generated 3 malicious requests, averaging approximately 0.1 requests per day. Operating from a residential network, this IP may represent a compromised home gateway or IoT device that has been drafted into a larger attack infrastructure. Two attack patterns were identified (User-Agent Anomaly and Path Enumeration), suggesting a semi-automated campaign that targets multiple vulnerabilities. With 151 flagged addresses, United States represents a significant presence in our threat database. A score of 90/100 places this address in the top tier of severity. Block and investigate any historical connections.
This IP is classified as residential, suggesting it may belong to a compromised home device, IoT botnet member, or an infected personal computer. Residential IPs involved in attacks often indicate malware infection without the owner's knowledge.
TLS fingerprinting creates unique identifiers based on how clients negotiate encrypted connections. The JA3 and JA4 methods generate hashes from TLS ClientHello parameters, enabling identification of specific tools and malware regardless of IP address changes.
Machine learning models analyze vast amounts of network traffic to identify attack patterns invisible to rule-based systems. Supervised models classify known attack types while unsupervised models detect anomalies that may indicate novel threats.