PHP2Shell vulnerabilities force global scramble to patch critical web infrastructure
Two critical PHP bugs, CVE-2026-109618 and CVE-2026-109619, allowed unauthenticated remote code execution, prompting widespread emergency patching across the internet.
On Wednesday, October 21, 2026, the internet faced a near-catastrophic disruption due to two critical vulnerabilities in PHP, the programming language that powers approximately 70% of the web. The PHP Foundation released emergency patches for CVE-2026-109618 and CVE-2026-109619, known collectively as PHP2Shell, after researchers at Calif demonstrated they could be exploited for unauthenticated remote code execution. Major industries, including banking, healthcare, and retail, temporarily took services offline to apply fixes before malicious actors could achieve mass exploitation.
What happened
Security researchers at Calif identified two bugs in PHP that had been present since 2024 but were previously thought to be non-exploitable. Upon discovering a viable attack vector, Calif alerted the PHP Foundation, which initiated a coordinated disclosure process. This involved sharing patches under strict embargo with major internet companies and web hosts to allow them time to prepare defenses before the details became public. The foundation publicly released the fixes on Wednesday, aiming to secure systems before widespread abuse could occur.
Despite these precautions, evidence suggests that exploitation began before the official patch release. Amazon reported detecting active exploitation attempts as early as October 17, a week prior to the public fix. It remains unclear how attackers gained early access to the vulnerability details, though theories include independent discovery, espionage, or a leak during the embargo period. The PHP Foundation noted that one company may have pushed patches live hours early, potentially exposing the fix to reverse-engineering by malicious actors.
The response was swift and severe. Governments urged immediate action, and many organizations chose to shut down their web-facing servers entirely rather than risk operating with vulnerable software. While mass exploitation was largely avoided thanks to this coordinated effort, security firms like Palo Alto Networks and Trail of Bits have confirmed post-exploit activity on specific targets. Researchers warn that the threat persists for unpatched systems, particularly legacy infrastructure that may not receive updates.
Key details
- The vulnerabilities are designated CVE-2026-109618 and CVE-2026-109619, collectively dubbed PHP2Shell.
- PHP powers roughly 70% of the web, including major platforms like WordPress, Drupal, and Joomla.
- The bugs allow unauthenticated remote code execution, granting attackers full control without passwords.
- Calif researchers discovered the exploits in 2024 but only recently found a way to weaponize them.
- Amazon detected exploitation attempts starting October 17, before the October 21 public patch release.
- AI tools assisted in the discovery process, though human expertise remained central to the find.
Background
PHP is an open-source scripting language widely used for server-side web development. Its longevity and ease of use have made it ubiquitous, but also mean that much of the internet relies on aging codebases. These older systems often suffer from "memory unsafe" flaws, where improper handling of data in memory can lead to severe security breaches. According to the National Security Agency, about 75% of actively exploited vulnerabilities stem from such memory-related issues.
The concept of coordinated disclosure is standard practice in cybersecurity. When a critical flaw is found, researchers share it privately with the vendor and major stakeholders to allow for patch development before public announcement. This window aims to minimize the time attackers have to exploit the bug once it is known. However, as seen with PHP2Shell, maintaining secrecy is difficult, and even a short leak can provide enough information for skilled attackers to develop exploits.
Why it matters
For teams running self-hosted software, this incident highlights the fragility of relying on widely used, open-source components. If you manage WordPress sites, Laravel applications, or any PHP-based internal tools, your infrastructure is directly exposed to these types of threats. The fact that exploitation began before the public patch means that simply waiting for an official announcement is not always sufficient. Proactive monitoring and rapid response capabilities are essential for maintaining security in such environments.
Furthermore, the role of AI in discovering these bugs signals a shift in the threat landscape. As AI tools become more capable of analyzing code for vulnerabilities, the window between discovery and exploitation may shrink. This places greater pressure on IT and DevOps teams to maintain rigorous update schedules and robust incident response plans. Legacy systems, which are often harder to patch, remain particularly vulnerable and may require isolation or replacement to mitigate long-term risks.
What you can do
- Immediately update all PHP installations to the latest patched version provided by the PHP Foundation.
- Audit your web servers for any signs of unauthorized access or unusual activity since October 17.
- Review and restrict network access to internal PHP applications to limit exposure to external threats.
- Implement automated monitoring for certificate and domain expiry to ensure service continuity during maintenance.
- Consider migrating critical services to memory-safe languages like Rust or Python where feasible.
- Establish a clear incident response plan that includes procedures for emergency patching and service restoration.



