Jessica Entwistle
November 19 2026
Today's briefing reflects a week in which widely used web infrastructure has come under significant scrutiny. Critical vulnerabilities in NGINX and PHP affect millions of servers globally, including many operated by UK businesses, while a supply chain compromise via Optimizely demonstrates how trusted third-party tools can become vectors for malware delivery at scale. Alongside these technical developments, the NCSC has published timely guidance on proactive threat hunting — a reminder that reactive monitoring alone is no longer sufficient for organisations that want to stay ahead of attackers.
Five critical vulnerabilities have been discovered in NGINX, the open-source web server software that underpins an estimated 225 million websites worldwide. Reporting from Infosecurity Magazine and The Hacker News confirms that the most significant flaw, tracked as CVE-2026-123007 and dubbed "nginx2root", has been present in NGINX's codebase since version 1.9.5, released in September 2015 — meaning it has existed for over a decade. The vulnerabilities affect NGINX's HTTP/2 support and could allow an unauthenticated attacker to remotely execute code on affected servers. F5, the company behind NGINX, released fixed versions 1.31.9 and 1.30.7 on 19 November 2026.
NGINX is one of the most widely deployed web server and reverse proxy technologies in use across UK businesses, cloud environments, and managed hosting platforms. Because the flaw requires no authentication to exploit, the barrier to attack is low — any internet-facing server running a vulnerable version of NGINX with HTTP/2 enabled could be at risk. For organisations running their own infrastructure or using managed services that rely on NGINX, this represents a meaningful exposure that warrants prompt attention. The fact that the vulnerability has existed for over ten years also raises important questions about how thoroughly legacy code is reviewed, even in widely trusted open-source projects.
For UK businesses running web applications, APIs, or reverse proxies on NGINX, this is a prompt to verify which version is deployed across your estate and confirm that patching to 1.31.9 or 1.30.7 is underway. Where NGINX is managed by a hosting provider or managed service partner, it is worth confirming their patching timeline directly rather than assuming it has been handled.
Source: Infosecurity Magazine
The maintainers of the PHP programming language issued an advisory this week announcing a scheduled security release to address critical vulnerabilities affecting all supported PHP versions. As reported by The Hacker News, the release was scheduled for 19 November 2026 and covers flaws including CVE-2026-115560, which relates to session upload progress handling and could allow remote code execution. SecurityWeek notes that this disclosure came just two days after a separate round of PHP patches — referred to as PHP2Shell — had already been released, indicating an unusually active period of security activity around the language.
PHP remains one of the most widely used server-side programming languages on the web, powering content management systems including WordPress, Drupal, and many bespoke web applications used by UK businesses. Remote code execution vulnerabilities in PHP are particularly serious because they can allow an attacker to run arbitrary commands on the underlying server, potentially leading to full system compromise, data theft, or ransomware deployment. Organisations that manage their own hosting environments or use shared hosting platforms should treat PHP patching as a priority, particularly given the rapid succession of releases this week.
For UK organisations running PHP-based applications, this is a prompt to confirm which PHP versions are deployed across your web estate and to verify that patching processes are in place to apply the 19 November releases promptly. Where PHP is managed through a hosting provider or CMS platform, check with your provider that updates are being applied and ask for confirmation of their patching timeline.
Source: The Hacker News
Krebs on Security reports that at least 178 high-profile websites — including the BBC, CNN, Equifax, Ledger, and Zoom — unwittingly served malware to visitors over a 24-hour period following a security compromise involving Optimizely, a widely used platform that helps websites personalise user experiences and run A/B tests. Attackers exploited a vulnerability in Optimizely to inject data-stealing malicious code into the platform's scripts, which were then loaded by any website using the service. The incident highlights how third-party tools embedded into websites can become a significant attack surface, particularly when those tools are trusted to load and execute JavaScript directly in visitors' browsers.
This type of supply chain attack — sometimes referred to as a web skimming or Magecart-style attack — is particularly difficult to detect because the malicious code originates from a trusted third-party source rather than the website itself. For UK businesses that use Optimizely or similar experience optimisation, analytics, or A/B testing platforms, this incident is a direct reminder that the security of your website is only as strong as the security of every third-party script it loads. Many organisations have limited visibility into what third-party JavaScript is doing on their pages, and few enforce strict Content Security Policies that would limit the damage from this type of compromise.
For UK businesses running customer-facing websites, this is a prompt to review which third-party scripts are loaded on your pages and to assess whether a Content Security Policy is in place to restrict unauthorised script execution. Organisations using Optimizely should check vendor communications for guidance on the incident and review whether their site was affected during the window of compromise.
Source: Krebs on Security
The National Cyber Security Centre published a blog post on 17 November 2026 making the case for proactive threat hunting as a board-level concern, not just a technical one. The NCSC argues that monitoring and alerting alone cannot detect every threat — particularly sophisticated attackers who move slowly, blend in with normal activity, or exploit gaps that standard detection tools are not configured to catch. Threat hunting involves security teams actively searching for signs of compromise or attacker presence within an environment, rather than waiting for automated alerts to fire. The NCSC highlights that threat hunting can also help organisations learn from the experiences of others facing similar threats, improving collective resilience across sectors.
For many UK organisations, security monitoring is handled reactively — alerts are investigated when they arise, but there is limited capacity or appetite for proactive investigation. The NCSC's guidance is a useful prompt for boards and senior leaders to ask whether their current security operations model includes any proactive detection capability, and whether their teams have the tools, time, and skills to look for threats that have not yet triggered an alert. This is particularly relevant for organisations in sectors that are frequently targeted, including healthcare, legal, financial services, and critical infrastructure, where a slow-moving attacker could cause significant harm before being detected through conventional means.
For UK businesses, this is a prompt to review whether your security operations model includes any proactive threat detection activity, or whether your team is entirely dependent on automated alerting. Boards and IT leaders should ask their security teams — or their managed security service providers — what visibility they have into attacker activity that falls below the threshold of automated detection, and whether threat hunting is part of the service they receive.
Source: NCSC UK
This week's developments are a useful reminder that good security hygiene across web infrastructure — keeping software patched, understanding what third-party code is running on your platforms, and maintaining visibility into your environment — remains the foundation of effective defence. The NGINX and PHP vulnerabilities are serious, but they are also fixable through prompt, well-organised patching processes. The Optimizely incident illustrates that supply chain risk extends beyond software vendors and procurement decisions to include the scripts and services embedded in everyday digital operations. Organisations that have already invested in understanding their attack surface, maintaining clear ownership of patching, and building proactive detection capability are far better placed to respond to weeks like this one — not because they anticipated every specific vulnerability, but because their security habits were already in good shape.