Censys Explains Why PHP2Shell Exposure Counts Differ

· Original article ↗

Summary

Censys explains why public PHP2Shell exposure counts vary, citing different scan sources, counting units, version banners, query methods and refresh rates, and describes its own scanning and historical tracking.

Key points

  • PHP announced an emergency release on October 20 for two pre-authentication RCE flaws, CVE-2026-109618 and CVE-2026-109619, rated CVSS 9.2 and collectively called PHP2Shell.
  • Exposure counts can differ because scanners use different indexes, count different units, handle hidden version banners differently, and may overcount similar version strings or duplicate hosts.
  • Banner-based counts can include systems whose Linux distributions backported fixes without changing the reported version; the article says these counts are upper bounds.
  • Censys says its scanners cover all 65,535 ports daily and revisit services about every 8 to 14 hours; its measurements found newly exposed hosts within 2 to 11 hours.
  • Censys says it treats websites as separate properties and retains four years of history to track service changes and patch adoption.
  • Reports of in-the-wild exploitation were changing at publication; at least two security vendors had retracted claims linking incidents to these CVEs, while the flaws were listed in CISA’s Known Exploited Vulnerabilities catalog.

Article Details

Topic
Censys’ Internet Map methodology for measuring PHP2Shell exposure and explaining differences in public PHP vulnerability counts.

CVE

Vendors

Products

Related Articles