Data story · October 2026

Live· updated

Of websites showing their PHP version, 9.1% in New Zealand and 4.2% in Australia so far were on a patched release when we checked

In the week of 20 October 2026 the PHP project disclosed two critical flaws, nicknamed “PHP2Shell”, that can let someone take over a web server without a password, and released fixes on 21 October: PHP 8.5.12, 8.4.27, 8.3.36 and 8.2.35. We gave .co.nz and .com.au websites the same check: one ordinary homepage request each, reading what the server says about itself.

Head to head · on a patched release for PHP2Shell

Australia Still scanning

4.2%

on a patched release, interim

of 60,425 sites showing a PHP version, from 441,573 .com.au names checked so far

30.2% of an estimated 1.46 million live .com.au names checked

Expected to finish between 1:30 am and 8:00 am on 23 Oct, NZ time. In Sydney that’s between 11:30 pm on 22 Oct and 6:00 am on 23 Oct.

Checked 22 Oct, 5:10 pm to 22 Oct, 7:53 pm NZ time (updates hourly)

New Zealand Census complete

9.1%

on a patched release

of 19,122 sites showing a PHP version, from 277,875 .co.nz domains checked

Checked 22 Oct, 8:53 am to 22 Oct, 12:45 pm NZ time

We have re-checked these New Zealand sites every hour since. That tracker is further down and isn’t compared with Australia.

  • No official fix (PHP 8.1 or older)
  • Fix available, not yet applied
  • On a patched release
  • Australia, not checked yet

A fair fight, with caveats. Australia’s check started about 4½ hours after New Zealand’s finished, so its hosts had more time to install the fix. While the scan runs, names are checked in roughly alphabetical order, so interim figures aren’t a random sample of Australian sites and can shift. Both shares count only sites that show an exact PHP version, and a version number isn’t proof either way: some hosts patch without changing it. This isn’t a measure of which country is more secure.

Dots are illustrative: their positions are made up, their colours follow each country’s real shares. Agents sweep the map to show what automated checking looks like. AI tools helped researchers at Calif, OpenAI and Wiz find PHP2Shell (Wiz).

The scoreboard

Same check, two countries

Every figure is a share of a stated group of sites, so the two countries can be read side by side even though Australia has more websites. Australia’s column is interim while its scan runs: names are checked in roughly alphabetical order, so interim figures aren’t a random sample of Australian sites and can shift.

New Zealand and Australia, census figures side by side
MeasureNew ZealandNZAustraliaAU (interim)
Domains checked277,875441,573 so far
Answered our request % of domains checked91.1%88.0%
Showed an exact PHP version % of those that answered7.6%15.6%
On a patched release % of those showing a version · head to head9.1%4.2%
No official fix, PHP 8.1 or older % of those showing a version44.7%46.5%
End-of-life, PHP 7.4 or older % of those showing a version33.6%37.0%
WordPress % of sites with an identified platform47.8%57.6%
WordPress sites on a patched release % of WordPress sites showing a version10.3%4.5%
Behind Cloudflare % of those that answered32.0%27.3%

New Zealand census: 19,122 sites showing a version, checked 22 Oct, 8:53 am to 22 Oct, 12:45 pm NZ time. Australia: 60,425 sites showing a version, checked 22 Oct, 5:10 pm to 22 Oct, 7:53 pm NZ time, figures as of 22 Oct, 7:53 pm and updated hourly while the scan runs.

01 · Who shows a version

7.6% of New Zealand sites and 15.6% of Australian sites announce their exact PHP version

Most servers keep their PHP version to themselves, and that’s good practice. Everything below about versions and patching is about the sites that do show it, so it’s a floor, not a count of every exposed site.

  • New Zealand
  • Australia (interim)
Answered: % of domains checked (NZ 277,875, AU 441,573). The other two: % of sites that answered (NZ 253,113, AU 388,447). Source: TheColab.ai scans.
View as table
MeasureNew ZealandAustralia

02 · Patched vs not patched

90.9% in New Zealand and 95.8% in Australia weren’t on a patched release yet

Two different problems hide in that number. Sites on PHP 8.2 or newer can fix it with a routine update. Sites on PHP 8.1 or older get no official fix at all: 44.7% in New Zealand and 46.5% in Australia need a bigger upgrade.

  • End-of-life (7.4 or older)
  • 8.0 or 8.1, no official fix
  • 8.2 or newer, fix not applied
  • Patched
Each bar is 100% of the sites in that country that showed an exact PHP version (NZ 19,122, AU 60,425 so far). Patched means 8.5.12, 8.4.27, 8.3.36, 8.2.35 or later on the same branch, at the moment we checked. Australia’s check started about 4½ hours after New Zealand’s finished.
View as table
GroupNew ZealandAustralia

03 · Top versions

The same old releases lead in both countries

PHP 7.4.33, the last 7.4 release from 2022, is near the top on both sides of the Tasman. The supported releases just before the fix, 8.2.34 and 8.3.35, are common too: those sites are one routine update away.

  • End-of-life
  • 8.0/8.1, no official fix
  • Fix available, not applied
  • Patched

New Zealand

Australia (interim)

The eight most common exact versions in each country, as a % of that country’s sites showing a version. Both columns use the same scale.
View as tables
New Zealand
VersionStatusSites% showing a version
Australia
VersionStatusSites% showing a version

04 · Platforms

WordPress leads in both: 47.8% of identifiable New Zealand sites, 57.6% in Australia

Self-hosted software such as WordPress runs on PHP that the owner or their host has to update. Hosted builders such as Squarespace, Wix and Shopify run their own servers and patch them centrally.

Of WordPress sites showing a PHP version, 10.3% in New Zealand (1,201 of 11,636) and 4.5% in Australia (1,861 of 41,601) were on a patched release.

  • New Zealand
  • Australia (interim)
% of sites with an identified platform (NZ 113,184, AU 193,482). WooCommerce is counted separately from WordPress.
View as table
PlatformNew ZealandAustralia

05 · CDN and hosting

32.0% of New Zealand sites and 27.3% of Australian sites answer through Cloudflare

What answered our request: a CDN in front, or the web server software itself. A CDN or firewall can hide the PHP version and block some attacks, but it isn’t a fix. The server behind it still runs PHP and still needs the update.

  • New Zealand
  • Australia (interim)
% of sites that answered (NZ 253,113, AU 388,447), by the CDN or server software named in the response. “Other servers” groups every less common server name.
View as table
CDN or serverNew ZealandAustralia

New Zealand only · Live patch tracker

Are New Zealand sites patching? We check every hour

Every hour we revisit the 18,252 websites that showed an exact PHP version in our first scan, with one ordinary homepage request each, and count how many we have seen move to a release with the fix. That total only rises: each site counts once, the first time we see it upgraded. What each check sees right now is a separate measure, further down, and it can go down as well as up.

This tracker follows New Zealand sites only, after the census. Its figures are a later, separate measurement and aren’t compared with Australia’s census figures above.

–

Loading the hourly series…

View as table
Check (NZ time)Window startsWindowNew upgrades seenCheck statusTotal observed upgrading
The upper chart is the number of distinct tracked sites we have seen move from an older version to a fixed release since our first scan. It only rises because each site is counted once, at the first check that sees it upgraded. A site that later hides its version, stops answering or shows an older release again stays counted as an observed upgrade, so this total is not the number of sites on a fixed release right now; that is the separate chart below. Each bar in the lower chart is one re-check, placed at the time it ran. It counts tracked sites seen on a fixed release (8.5.12, 8.4.27, 8.3.36, 8.2.35 or later) for the first time, after showing an older version at our first scan. The upgrade happened some time in the window since the previous check, not necessarily at that hour, and a site that didn’t answer the previous check may have upgraded earlier than that. Our first scan took hours and the first window starts when it finished, so a site checked early in that scan may have upgraded before the first window and is then counted in the first bar. Re-checks ran every one to two hours until the afternoon of 22 October and every hour since, so early windows are longer; the table gives each window’s exact length. A check whose site results didn’t arrive is shown as a gap and is never counted as zero; a zero appears only when every result of a check arrived. Each site counts once. Sites that were already fixed, hide their version or aren’t on the list are never counted. These are upgrades we observed, not proof a site is secure, and not something this page caused. Servers patched by their Linux distribution may still report an older number.

Detail: what each check sees right now

This is a different measure from the total above: the share of tracked sites that answered, showed a PHP version, and were on a fixed release at that check. It can go down as well as up, because sites that were on a fixed release drop out of it when they don’t answer, stop showing a version or show an older release. Between the 3:21 pm and 4:21 pm checks on 22 October it dipped from 13.7% to 13.4%: 49 sites newly showed a fixed release, while 90 that had shown one didn’t answer (most on LiteSpeed or Cloudflare-fronted hosting), 16 stopped showing a version and 2 showed an older release. The hover details give the number of sites showing a version at each check.

14.4%

on a patched release at the latest check · latest check 22 October 2026, 7:21 pm NZDT

Share of tracked sites on a patched PHP release

Tracking started 22 October 2026, 10:37 am NZDT — first update in ~1 hour.

View as table
Checked (NZ time)PatchedNot yet patchedShowing a PHP version
The first point is our main scan of these sites; each later point is a fresh re-check of the same list. A site counts only if it answered and still showed an exact PHP version that time, so the number checked moves a little between points, and some of the change can come from that mix rather than from patching. Patched means PHP 8.5.12, 8.4.27, 8.3.36 or 8.2.35 or later. Servers patched by their Linux distribution may still report an older number.

Upgrades seen since tracking began

–

Loading the count…

We count a website once, the first time an hourly re-check sees it on a fixed release (8.5.12, 8.4.27, 8.3.36, 8.2.35 or later) after it showed an older version on our tracking list. Sites that hide their version or don’t answer are never counted as upgraded. This is what we observed, not a claim that this page caused it, and a version number can’t prove a site is secure.

Notify me when a change is observed

For one website on our tracking list: an email when an hourly re-check sees its PHP status change, for example from “update needed” to a fixed release.

Not sending email yet. We’re saving requests now. Nothing is sent, including the confirmation email, until delivery is switched on, and alerts start only after you confirm.

Only websites already on our list of 18,252 re-checked sites can get alerts. We don’t start checking new websites. A change is only noticed when a re-check runs, about every hour, and when the site answers and shows its PHP version. If it hides the version or doesn’t answer, we can’t tell you anything. Signing up for incident updates doesn’t sign you up for this. Privacy note

06 · What this means

Two problems: sites that haven’t updated yet, and sites that can’t

The PHP project supports each version for a fixed period. After that it stops publishing security releases. On 21 October 2026 it released fixed versions 8.5.12, 8.4.27, 8.3.36 and 8.2.35.

46.3% NZ · 49.3% AU

Haven’t updated yet. On PHP 8.2 or newer but not yet on a patched release, as a share of sites showing a version. This is a routine update a host can install, and many will catch up over the coming days.

44.7% NZ · 46.5% AU

Can’t update the normal way. Running PHP 8.1 or older, which won’t get the fix through normal updates. That includes 33.6% (NZ) and 37.0% (AU) on end-of-life PHP 7.4 or older. These sites need a major PHP upgrade, which can mean updating the website software too.

  • PHP 7.4, 8.0 and 8.1 get no official fixed release. The PHP security team published minimal source-code patches for 7.4 to 8.1, but a host has to build and install those themselves.
  • PHP 7.3 and older get no fix of any kind from the PHP project.
  • Supported versions are only safe once the host installs the new release and restarts the web server.

If you run a website

Ask your web developer or hosting company which PHP version your site uses and whether this week’s fix is installed. If the site runs PHP 8.1 or older, plan an upgrade to a supported version. Many hosts let you change the PHP version in their control panel, but test the site afterwards, because old themes and plugins can break.

If you want to check a site

Some servers announce their PHP version in a response header called X-Powered-By, for example PHP/7.4.33. In most browsers, open the developer tools (F12), choose the Network tab, reload the page and select the first request to see its headers. No header does not mean a site is safe; it only means the version is not shown. The reliable answer comes from the people who run the site.

Check a site

Is your site affected?

Enter a website address. We make one ordinary request to its homepage, the same as a visitor would, and read what the server says about itself.

For example yourbusiness.co.nz. Any domain works, not just .nz.

Get alerts about your site and incidents like this

We’ll email you when a serious web-server flaw like PHP2Shell affects New Zealand sites, with plain-English steps to take.

We store your email only to send incident updates; unsubscribe anytime. Privacy note

Watch: what happened, in about a minute

A plain-English explainer on this week's critical PHP flaw and what New Zealand site owners should do. It was made before the Australian comparison.

Older figures in this video. It was made before our 22 October correction and shows the original count of 19,644 PHP sites, which included some sites twice. The corrected census count is 19,122 unique sites, and the 91% headline is unchanged after rounding. See the correction

07 · Questions

Frequently asked questions

What is this page?

A public reference on how New Zealand and Australian websites stand on this week’s critical PHP vulnerabilities. It compares a census of 277,875 unique .co.nz domains checked on 22 October 2026 (NZ time) with a census of .com.au domains given the same check, starting about 4½ hours after New Zealand’s finished, which is still running. For New Zealand it also shows an hourly re-check of the sites that showed an exact PHP version. We corrected the census counts on 22 October; see Corrections. It does not name or list any individual site.

How does the New Zealand vs Australia comparison work?

Both countries got the same check: one ordinary homepage request per domain, the same software reading the same headers, and the same rule for counting each domain once. The lists of names differ: the Australian list drew on more public sources (see Method), which may be part of why more Australian sites showed a version. The head-to-head figure is the share of sites showing an exact PHP version that were on a patched release when we checked them: 9.1% of 19,122 in New Zealand and 4.2% of 60,425 in Australia so far. Every figure is a percentage of a group we name next to it, because Australia has many more websites. Australia’s check started about 4½ hours after New Zealand’s finished, which gave its hosts more time to install the fix. While the scan runs, names are checked in roughly alphabetical order, so interim figures aren’t a random sample of Australian sites and can shift. The New Zealand hourly tracker is a later, separate measurement, so we don’t set it against Australia.

Why is Australia still being scanned?

There are many more .com.au names than .co.nz names, and we check them politely, one request per site at a steady rate. The page shows how far through the scan is and a finish estimate, based on how fast recent batches have gone. Until it finishes, the Australian figures are interim and update every hour. Names are checked in roughly alphabetical order, so the interim figures aren’t a random sample and can move as more sites are checked.

Does this show which country is more secure?

No. It compares one thing: how many sites that announce their PHP version were on a patched release at the moment we checked. Most sites don’t announce a version, some hosts patch without changing the version number, and Australia was checked later. A higher share patched is good news for those sites, not a security rating for a country.

What happened with PHP this week?

In the week of 20 October 2026 the PHP project disclosed two critical vulnerabilities (CVE-2026-109618 and CVE-2026-109619, nicknamed “PHP2Shell”). In some set-ups they can let someone take over a web server without a password. PHP is the language behind roughly 70% of websites whose server language is known, including most WordPress sites. Fixed releases exist only for supported versions: PHP 8.5.12, 8.4.27, 8.3.36 and 8.2.35.

What does “end-of-life PHP” mean, and why can’t it be patched?

Each PHP version is supported for a fixed period, usually four years. After that, the PHP project stops publishing security releases for it, and the version is “end-of-life”. PHP 8.1 and older are past that point, so there is no official fixed release. The PHP team published unofficial source-code patches for 7.4 to 8.1, but a host has to build and install those themselves; PHP 7.3 and older get nothing. The dependable fix is to move to a supported version.

Is my website affected? How do I check?

Ask your web developer or hosting company which PHP version your site runs and whether this week’s update is installed. Some sites also show the version in a response header called X-Powered-By, which you can see in your browser’s developer tools (Network tab). If no version is shown, that does not mean the site is safe. The goal is PHP 8.2 or newer, kept up to date. Sites built on hosted services such as Squarespace, Wix or Shopify are maintained by those companies.

I run a site on old PHP. What should I do?

Contact your host and ask them to move your site to a supported PHP version (8.2 or newer) with the latest update installed. Before or alongside that, update your website software, theme and plugins, because older ones may not work on newer PHP. Take a backup first and check the site afterwards. If you don’t know who hosts your site, your web developer or domain provider can usually tell you.

How accurate are these numbers?

They are a floor, not a total. We only count sites that openly disclose their PHP version (19,122 in New Zealand, 60,425 in Australia so far); most servers don’t, so the real number of exposed sites is likely higher. The check is passive: one ordinary request per site for its homepage, reading the response headers and page, with no scanning, probing or attempts to exploit anything. It is a point-in-time snapshot that updates as the scan runs and hosts apply fixes. The “not yet patched” share is the most time-sensitive figure: the scan began on the day the fix was released, and servers patched by their Linux distribution may still report an older version number. See Methodology & caveats for detail.

Who made this?

The data and analysis are by TheColab.ai. You’re welcome to cite the figures with credit.

08 · Methodology & caveats

How we counted, and what the numbers can’t tell you

The domain list

We built a list of registrable .co.nz names from public sources: the Common Crawl web graph (2022 to 2026 releases), the Tranco, Cisco Umbrella and Majestic top-site lists, the Chrome UX Report (New Zealand and global), and the open tb0hdan domains dataset. Names that did not resolve in DNS were dropped. Other New Zealand zones (.nz, .org.nz, .govt.nz and others) are not included.

Australia. We built a list of registrable .com.au names from public sources only, and from more of them than the New Zealand list: Common Crawl (2022 to 2026), the Tranco, Cisco Umbrella and Majestic lists, the Chrome UX Report (Australia and global), the tb0hdan Australia dataset, certificate transparency logs, and names that exist under other .au zones tried under .com.au. Names that did not resolve in DNS were dropped. There is no public list of every .com.au name, so this is a large public-source sample, not every Australian website. Other .au zones are not included. The Australian crawler identifies itself as AUWebsitesCensus/1.0 and started about 4½ hours after the New Zealand census finished; it is still running, in roughly alphabetical order, so Australian figures are interim and not yet a random sample.

Finish estimate. We count the live names still to check, plus names DNS hasn’t answered yet for the late end, and divide by the speed of the last six completed batches of 40,000 names: the early end uses the fastest of those, the late end the slowest plus 15%. It’s recalculated every hour.

What we collected

For each domain we made one ordinary request for the homepage, as a browser would, from a crawler that identifies itself (NZWebsitesCensus/1.0). If a site did not answer we tried at most a few address variations (https, http, www). Sites that failed were re-checked in later retry and rescue passes, so some domains have more than one check. Each domain counts once: we use its latest successful answer, or its latest failed attempt if it never answered. We read only what the server sends back to every visitor: the response headers and the homepage HTML. We did not probe for vulnerabilities, log in, request any other page, or test whether any site can be exploited.

How versions and platforms were identified

A site counts as “disclosing” when its Server or X-Powered-By header, or a technology fingerprint built from those headers, contains a full PHP version number. Platforms were identified with the open-source Wappalyzer fingerprints and the page’s generator tag. Percentages in the version charts use disclosing sites as the denominator; platform percentages use sites with an identifiable platform.

Sample sizes in each country
StageNew ZealandAustralia (so far)
Unique domains checked277,875441,573
Responded to the request253,113388,447
Platform identified113,184193,482
Disclosed a PHP version19,12260,425

Definitions

  • Patched release: the site reports PHP 8.5.12, 8.4.27, 8.3.36 or 8.2.35, or a later release on the same branch. These are the first releases with the fix.
  • Not yet patched: the site reports any other version, that is, a PHP version released before this week’s fix. That includes supported branches that have not been updated yet (for example 8.3.35) and every end-of-life version.
  • No official fix: PHP 8.1 or older. End-of-life: PHP 7.4 or older.

Shares of versions and patching are of the sites in each country that disclosed an exact PHP version, unless stated otherwise. Each figure is a snapshot of the moment the site was checked.

Caveats

  • The figures are a floor, not a total. Most servers do not announce their PHP version, so this counts only the sites that do. Servers configured to hide version details tend to be better maintained, so the share of out-of-date PHP among all sites is unknown and the absolute number of exposed sites is very likely higher than shown here.
  • A version number is not proof of exposure. Some Linux distributions and long-term-support vendors back-port security fixes without changing the version string, so a few sites listed as old may be patched. This matters most for the “not yet patched” figure: a server patched by its distribution can still report, say, 8.2.34. In the other direction, a supported version is only safe if the latest point release is installed.
  • Very old versions. Published analysis traces the affected code back to PHP 5.4. In New Zealand, the 315 sites showing PHP 5.3 or older may not be affected by these two flaws specifically, but have had no security fixes of any kind for over a decade. We count them as end-of-life.
  • A snapshot in time. The New Zealand scan began on 21 October 2026, the day the fixed releases were published, so many sites were checked within hours of the fix appearing. Hosts are updating servers now, and the “not yet patched” share in particular will fall. The “last updated” time above shows when this data was generated.
  • The patch tracker follows a fixed list. The over-time chart re-checks only the 18,252 sites that showed an exact PHP version in the first scan, every hour, one homepage request each with the same settings as the main scan. New sites are not added. Sites that stop answering or start hiding their version drop out of that check’s count, so the denominator moves a little from point to point. It covers New Zealand only and is not compared with Australia.
  • Different times. Australia’s check started about 4½ hours after New Zealand’s finished, so its hosts had more time to install the fix. The comparison is of each census at the time it ran.
  • Different name lists. The Australian list drew on more public sources than the New Zealand one, including certificate transparency logs, so the two groups of sites weren’t assembled identically.
  • One domain, one site. Many domains share a server or hosting account, so site counts are not counts of servers or organisations. We do not name or list individual sites.

Sources on the vulnerabilities

Data & analysis by TheColab.ai. You are welcome to cite and reproduce these figures with credit. The underlying aggregate numbers are published at data.json (as shown on this page) and census.json (the corrected 21 October census, with its selection rule). The two-country figures are at nz-vs-au.json.

09 · Corrections

Corrections

Correction, 22 October 2026. We first published census figures that counted some websites more than once: when a site was re-checked in our retry and rescue passes, each extra check was counted as another website. We originally said we scanned 322,710 .co.nz websites; that was 322,710 checks of 277,875 unique domains. The PHP sample of 19,644 checks becomes 19,122 unique sites when each domain is counted once, at its latest successful check, and the census share not on a patched release moves from 91.0% to 90.9%. We have also withdrawn our claim that about 73% of identifiable New Zealand websites run on PHP. The mistake was ours. The details are below.

22 October 2026: census counted some websites twice

What went wrong. Our census checked each domain once, then re-checked sites that hadn’t answered in a retry pass and a rescue pass. Our report counted every check as a separate website. Sites checked two or three times were counted two or three times, which inflated the totals. This was our error. We have recounted so that each domain appears once, using its latest successful answer by check time, or its latest failed attempt if it never answered.

Census figures as originally published and as corrected
Figure (21 October census)Originally publishedCorrected
.co.nz websites scanned322,710 “websites”277,875 unique domains (322,710 checks)
Sites disclosing an exact PHP version19,64419,122 unique sites
Of those, not on a patched release17,877 (91.0%)17,385 (90.9%)
On a patched release1,7671,737
No official fix (PHP 8.1 or older)8,784 (44.7%)8,540 (44.7%)
Responded to the request269,260253,113
Platform identified114,829113,184
“Run on PHP”about 73% of identifiable sitesWithdrawn (see below)

Both headline shares round to 91%, so the main finding stands, but the counts behind it were too high. Interim totals shown while the scan was still running were counted the same way and are replaced too.

Withdrawn: “about 73% of identifiable New Zealand websites run on PHP”. That figure divided sites showing any sign of PHP by the sites whose platform we could name. Many sites in the first group are not in the second, so it was not a real share and should not have been presented as one. It also came from a .co.nz sample, not every New Zealand website. We have removed it. As a narrower observation: 32.7% of the 253,113 unique domains that answered showed some PHP indicator in their headers or detected technology. That is a lower bound for this sample, not an estimate of how many New Zealand sites use PHP.

The patch tracker was not affected. The hourly tracker re-checks a fixed list of 18,252 sites that showed an exact PHP version in the main census pass, counting each domain once. Its first point (21 October) and the hourly upgrade counts are unchanged. Its figures are a separate, later measurement: the PHP shares at the top of this page come from the latest hourly check of that list, so they differ from the census’s 90.9%.

Update, 22 October 2026. The page now leads with a comparison of the New Zealand and Australian censuses, so the PHP shares at the top no longer come from the hourly tracker. The tracker’s figures are in its own section.

What we changed: the census summary, the downloadable report and chart, the page text and its search and social descriptions. The corrected census is published at census.json with its selection rule. Our explainer video and earlier social posts still show the original counts; we have marked the video on this page.