Every website we manage was protected before this week’s critical PHP vulnerabilities could be exploited, and every site / system is now running a fixed version of PHP.
It has been quite a week for the internet. On Tuesday morning, a power failure in one of Google Cloud’s London data centre zones disrupted services across the UK for several hours (The Register). Revolut cards and app payments failed for many customers, Lloyds, Halifax, Bank of Scotland and HSBC customers struggled to reach online banking, and Royal Mail parcel tracking was affected (MoneyCheck). Some UK court services were caught up in it too (Tech Insider).
Then, on Tuesday evening, the PHP security team announced an emergency fix for two critical flaws in the software that powers most of the web, including WordPress.
We want to explain what happened, what we did about it and why this is a good example of how website security really works.
What happened
Two vulnerabilities in PHP (CVE-2026-109618 and CVE-2026-109619) were each rated 9.2 out of 10 for severity. PHP is the language that runs WordPress on the web server, so the flaws affected a huge share of websites worldwide.
In the worst case, the flaws could let an attacker run code on a server without logging in. To make things more urgent, some details became public before the fix was ready on Wednesday 21st October at 3pm (UK time).
The important point is where the problem sat. It was in PHP itself, not in WordPress, its plugins or any website code. WordPress confirmed that no WordPress update was needed. The fix was a new version of PHP, installed at server level by the hosting platform.
While others took their websites offline, we didn’t have to
The response elsewhere shows how serious this was. Cyber security agencies across Europe, including the UK’s NCSC, advised organisations that could not patch quickly to take their PHP websites offline until they could (Safeguard). Many high-profile organisations did exactly that:
- Dutch government services, including DigiD and MijnOverheid, went offline as a precaution, along with public bodies in Finland, Denmark and France (Newsweek).
- In Germany, the Charité hospital took its entire website offline and parts of Berlin’s city portal were disabled (heise).
- Some Canadian federal government websites were also taken down temporarily (CBC).
Even on Thursday, a day after the fix was released, some had still not returned. One public tracker found the French and Dutch data protection authorities’ websites still unavailable, along with several French prefecture sites (PHP2Shell tracker). Early attack attempts against unpatched servers were also being reported, although no successful exploits had been confirmed at the time of writing (Malwarebytes, heise).
Not one website we manage had to be taken offline.
How our infrastructure responded
The websites we manage were shielded before the fix even existed, and our clients did not need to do anything. Here is how the two days unfolded.
- Tuesday evening: as soon as the PHP announcement landed, our hosting infrastructure added server-level protection rules and switched off the affected PHP features. Our team started monitoring.
- Early Wednesday: network-level protection rules were active across our platforms, blocking attack attempts before they reached a server.
- Wednesday morning: we reviewed the security advisory, checked the PHP version of every website we manage and applied extra protective settings where needed. We also confirmed directly with our hosting partners that their protections covered every PHP version in use.
- Wednesday 3pm: the PHP project released the fixed versions. Our hosting infrastructure began rolling them out straight away, with no downtime.
- Overnight into Thursday: the rollout completed across all of our platforms.
- Thursday morning: we checked every server ourselves and confirmed that every website we manage is running a fixed version of PHP.
We are now completing checks on each website for any sign of compromise, covering user accounts, WordPress core files, website files and server logs. So far we have found no issues, and we will keep monitoring closely over the coming days.
WordPress is secure: hosting and maintenance make the difference
WordPress often gets the blame when security stories hit the headlines, but this week shows why that is usually misplaced. WordPress itself needed no changes at all. The risk sat in the server software underneath it, and the protection came from how those servers are hosted and looked after.
When we talk to clients about website security, we focus on three things:
- Managed, enterprise-grade hosting. Platforms that can add protection rules across every server within hours, and install fixes at scale, are what kept websites safe before a patch existed.
- Layered protection. Defences at the network level and the server level mean an attack has to get past more than one barrier.
- Active maintenance. Someone has to read the advisory, check every site, confirm the fix is really in place and look for signs of trouble afterwards. That is what our team did, site by site.
A website left on unmanaged hosting with nobody watching would have faced this week alone. That is the real difference between a secure WordPress site and a vulnerable one.
Still making headlines
This story is far from over. Amazon’s threat intelligence team says the flaws are now being broadly exploited, with the first attempts spotted on 17th October, four days before the fix was even released (BleepingComputer).
The impact is also spreading beyond websites. Palo Alto Networks has confirmed that its firewall management software is affected and has not yet released a fix, while Fortinet and Cisco are still checking dozens of their products (Help Net Security).
The longer-term risk sits with websites nobody is looking after. PHP 8.1 and older no longer receive security updates from the PHP project, so sites still running them cannot simply be patched (BleepingComputer). That is exactly why keeping server software current matters long before an emergency arrives.
What this means for you
If we manage your website, there is nothing you need to do. Your site is running a fixed version of PHP, and our checks so far have found no issues. If anything does need your attention, we will contact you directly.
We take incidents like this seriously because our clients’ websites are central to how they work, communicate and serve their members. Weeks like this one are exactly why we invest in robust infrastructure and proactive maintenance.
If your WordPress site is not managed by us and you are unsure whether it is protected, or you would simply like to talk about how your website is hosted and maintained, get in touch and we will be happy to help.
