Chapters0/54 ▾
- 1.Architecture & Log Sources0/5
- 2.Indicators of Malicious Activity0/5
- 3.Investigation Tools & Threat Hunting0/5
- 4.Process Improvement, Automation & AI0/5
- 5.Vulnerability Scanning Methods0/5
- 6.Analyzing Scan Output0/5
- 7.Prioritization, Mitigation & Controls0/5
- 8.Attack Frameworks & the Response Process0/5
- 9.Incident Response Techniques0/5
- 10.Vulnerability Reporting & Communication0/5
- 11.Incident Reporting & Metrics0/4
Chapter 6 · Lesson 1 of 5
Validating findings and spotting false positives
1 min read
A false positive is a reported condition that is not actually present. The most common source is version inference: the scanner reads a service banner or an HTTP Server header and assumes every flaw listed for that version.
When an uncredentialed scan reports many critical items and a credentialed scan of the same unpatched host reports far fewer, the extra banner-only detections are unconfirmed claims, not fixed issues.
On a reverse proxy the header may describe the proxy, a backend or nothing at all, so the analyst confirms on the host which package is listening and at what patch level, and records that evidence before anyone is asked to act.
Some false positives come from the tool's own configuration. If forty internal interfaces are flagged as untrusted because their certificates come from an internal certificate authority missing from the scanner's trust store, the fix is to import that root into the scanner and rescan. Blanket suppression would also hide a real expired certificate or weak key later.
Web application findings need the same care. A database error returned after the scanner appended a quote proves only that malformed input changed the response. Confirmation means reproducing the request in a test environment with a harmless payload that changes the query's results, or reading the code path or query log for unparameterized input.
Check yourself
1. A scan report lists 'web server 2.4.29 - multiple critical vulnerabilities' on port 443 of a host, and the evidence field contains only an HTTP Server response header. The host is a reverse proxy that terminates TLS and forwards requests to application servers. Before a remediation ticket is opened, what is the correct action?
2. An uncredentialed scan of a Linux web server reports eleven critical findings. A credentialed scan of the same host two hours later reports three, and the eight that disappeared were all detected from service banners. Nothing was patched in between. What does this comparison most reliably establish?
Facts checked 2026-11-15 · Sources: CompTIA: CySA+ (CS0-004) certification and exam objectives, NIST SP 800-115: Technical Guide to Information Security Testing and Assessment
CompTIA CySA+ cheat sheet: every key fact on one page →