
ABUSE.MOM — BEHAVE OR GET EXPOSED
| Signature | Description | Points | Severity |
|---|---|---|---|
| 404 ratio >= 60% | Majority of requests returned 404 — enumeration | +25 | |
| Danger strong hits: 1 | High-risk paths: shells, RCE vectors, exploits | +25 | |
| Foreign referer | Referer from unrelated external domain | +10 |
Reconstructed HTTP requests from server access logs. Target domains redacted for security.
* Typical request patterns for detected signatures. Actual target domains are redacted.
Block scanning from 105.172.114.11: rate-limit 404 responses per IP, deploy a honeypot 404 page, ensure no backup files are web-accessible.
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.
105.172.114.11 has been assigned a threat score of 60/100 (High). This classifies it as a high-severity threat. Proactive blocking is recommended for sensitive infrastructure.
The following attack categories were identified:
Network traffic from 105.172.114.11, located in Luanda, AO, operating on the network of UNITEL SA, has been classified as malicious by our automated threat scoring engine. The address has been active for 6 days in our monitoring system, producing 630 flagged requests at a rate of ~105/day. This is a mobile network IP. While mobile addresses are typically shared via CGNAT, persistent malicious activity from this specific address suggests automated abuse. The IP exhibits directory enumeration behavior, systematically requesting non-existent paths to discover hidden files and misconfigured resources. Our records show 25 malicious IPs originating from AO, positioning it as a notable contributor to global threat activity. At 60/100, this IP presents a meaningful threat. Implement rate limiting with escalation to blocking.
Path traversal attacks attempt to access files outside the intended directory by manipulating file path references. Attackers use sequences like ../ to reach sensitive system files such as /etc/passwd or application configuration files.
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.