Short answer: the Apache flaw tracked as CVE-2026-119618 affects Apache HTTP Server from 2.4.55 through 2.4.70, and Red Hat says only servers with HTTP/2 enabled are affected. The separately distributed mod_h2 module from GitHub is also affected, in releases from 2.0.8 through 2.0.42. The fix is to upgrade: Apache HTTP Server 2.4.71 was released on 2026-11-11 and carries the fix.

Handle this without reading any server code. Below is a plain-English way to check whether your site is exposed, a short message you can send your hosting company today, and what we'd do if the answer comes back "we're working on it."

What happened, in plain words

Think of your web server as the program on the other end that hands over your pages when someone visits. Here is the version a busy owner needs.

The same Apache advisory lists a second, separate issue. CVE-2026-106023 is a mod_md NULL pointer dereference that leads to a child process crash, and it affects Apache HTTP Server 2.4.30 through 2.4.70. When you talk to your host, ask about both CVEs by number so nothing slips through.

Who found it is worth a sentence, because it shows the flaw was spotted from more than one direction. Anthropic's OSS Scanner is credited as a finder, under identifier ANT-2026-MAPKT6MZ, and Anthony Weems and Simon Scannell of Google Cloud Vulnerability Research are credited too.

How widespread is it?

Cloud security company Wiz published two figures from the environments it can see.

64%
of cloud environments have resources vulnerable to CVE-2026-119618 (Wiz data)
51%
of cloud environments have publicly exposed instances (Wiz data)

According to Wiz data, 64% of cloud environments have resources vulnerable to CVE-2026-119618, and 51% of cloud environments have publicly exposed instances. Keep in mind that exposure figures from scanners can't tell you how any given server is set up. They are industry figures, not a count of small-business sites. Our reading is simple: this isn't a rare corner case, so assume you need to check rather than assuming you're fine.

Why a small business site is a real target

It's tempting to think "nobody is going to hack a dental office's website." In our experience, bugs like this aren't exploited by someone choosing your business by hand. Automated scanning doesn't care whether a server belongs to a plumber or a bank; if it answers like a vulnerable server, it gets treated like one.

What an attacker might want from a small site usually falls into one of these buckets:

The cost side is real even at the industry level. IBM reports the average global breach cost dropped to USD 4.44 million from USD 4.88 million, a 9% fall. Treat an industry average as a sign of direction, not a quote for your business. The point stands either way: cleanup, downtime, notifying customers and rebuilding trust cost far more than an afternoon of patching.

The San Diego angle

For San Diego businesses, the website often does real work. If your site collects anything from customers through a form, booking link or intake page, the server underneath it deserves the same attention as the front desk. In our experience, many local sites are built once, handed to a budget host and rarely looked at again. That "set it and forget it" setup is exactly how a web server ends up unpatched.

If your business sits in a regulated field, such as healthcare, legal or financial services, a breach of a form that collects client details isn't just an IT headache. Our advice is to loop in whoever handles your compliance as soon as you know you were exposed, rather than after.

Step 1: Find out what you're running

There are a few ways to get an answer, from easiest to most technical. A clear answer from any of them is enough.

Option A: Ask the people who run your server

If you're on shared hosting or a managed WordPress plan, you probably can't see or change the server setup yourself. Your host can. Skip to the email template below; we'd start there.

Option B: Check whether HTTP/2 is on

Because only servers with HTTP/2 enabled are affected, checking HTTP/2 tells you a lot. Red Hat's command points at the server itself; a web person can try the same idea against your public address:

curl -sI --http2 https://www.yourbusiness.com

Reading the result is straightforward. When HTTP/2 is enabled, the first line of the output returns HTTP/2 200. When HTTP/2 is disabled, the first line returns HTTP/1.1 200 OK.

No command line? Open Chrome's developer tools and look at the list of requests for your page. In Chrome DevTools, you right-click the header of the Requests table to hide or show columns.

Heads up

**Treat the browser or curl result as a clue, not a verdict.** If anything sits in front of your site, such as a CDN or a security service, confirm the settings on the actual origin server with your host before you relax or panic.

Option C: Check the version number

Some hosting control panels show a server version on an information page. If you see an Apache version from 2.4.55 through 2.4.70, you're in the affected range for CVE-2026-119618.

One caution from our side: don't rely on the version number alone. Ask it in exactly those words.

Step 2: What to send your host this week

Here's a short message you can copy, paste and send to your hosting support today. Keep it this specific; vague tickets get vague answers.

Hi, we're checking our exposure to the recent Apache HTTP Server vulnerabilities. Please confirm, for the server hosting [yourbusiness.com]:

1. Which server software and release are running, and does the installed package include both fixes?

  1. Is HTTP/2 enabled on the server, via the Protocols directive, since only servers with HTTP/2 enabled are affected?
  2. Have you seen any signs of unusual access on our account since the advisory came out?

    Thank you.

What a good answer looks like: a version number, a yes or no on the fix, and a date. A weak answer is a vague promise that the servers are secure and regularly updated. If you get the weak version, reply once asking for the specific version and patch status. If you still don't get a straight answer, that tells you something about your host.

Step 3: The one setting that buys time

If your host, or your own server, can't upgrade right away, the workaround is about which protocols your web server offers.

Wiz's mitigation guidance is to edit the configuration file to restrict active protocols to HTTP/1.1 by removing h2 and h2c references, and it lists Protocols http/1.1 as the directive to use. In case h2c looks unfamiliar, h2c is HTTP/2 over cleartext TCP (without TLS).

Placement matters. Apache's guide says that, depending on where you put this directive, it affects all connections or only those to a specific virtual host. Its own example allows only HTTP/1 on connections, except SSL connections to test.example.org, which offer HTTP/2. So a single tidy line in the main file doesn't prove that every site on the server is covered.

In plain steps, for whoever manages the server:

  1. Back up the Apache configuration files before touching anything.
  2. Search the main configuration and every virtual host file for Protocols lines.
  3. Test the configuration, then restart Apache.
  4. Re-run the curl check from Option B and confirm the reply.
Tip

**Treat this as a seatbelt, not a repair.** Red Hat says servers configured to serve only HTTP/1.1 are not vulnerable, but we'd still treat the switch as temporary and get the patched version installed. Put a date on the upgrade, and undo the workaround once it's done.

If you're on shared hosting, you usually can't make this change yourself, because the main server configuration belongs to the host. That's why the stopgap question in the email above matters.

Worked example

Hypothetical: a three-chair dental practice in Mission Valley has a WordPress site on a low-cost shared plan. The office manager follows the steps above.

StepWhat they didResult
Browser checkDeveloper tools, Protocol columnShows "h2"
Version checkControl panel server info pageShows a version inside the affected range
Email to hostSent the template above on MondayHost replies Tuesday: patch scheduled Thursday night
StopgapAsked host to switch off HTTP/2 until thenHost declines on shared servers; offers no interim fix
DecisionWeigh waiting three days vs. movingPractice waits, but turns on login alerts and checks the site daily
ConfirmRe-check on FridayHost confirms the patched package is installed

The interesting row is the "host declines" one. On shared hosting you're one tenant among many, and a host may not change a server-wide setting for you. That's not automatically a failure, as long as they patch quickly and tell you when. If a host can't give you a date at all, that's when we'd start planning a move.

Pitfalls we see

Your Apache checklist

Print this or paste it into your task list:

What we'd do

If we were looking after your site, we'd treat this as a same-week job, in this order: confirm the exact server and package your site runs on, check HTTP/2 from outside and inside, apply the HTTP/1.1-only workaround where we control the configuration, push the host for a dated patch where we don't, and then clean up afterward with password rotation and a check for anything unfamiliar.

The bigger lesson is about ownership. In our experience, small-business sites often lack a named person responsible for "is the server patched?" That's the gap bugs like this fall through. Our web development work for small businesses includes setting up hosting you can actually see into, documenting who owns updates, and putting simple monitoring in place so the next advisory is a quick check instead of a scramble.

If you'd rather not watch security advisories yourself, this is also a good fit for automation. We'd suggest a small AI automation workflow that watches vendor advisories for the software your site runs, checks your server's version and settings, and sends you a plain-English alert only when something actually applies to you. For businesses with more than one site or location, that saves a lot of guesswork.

We work with businesses across San Diego, from single-location offices to multi-site groups, and we'll tell you plainly if your current host is fine and you don't need us.

Next step

Send the email to your host today; it costs nothing and you'll know where you stand quickly. If the answer is unclear, or you don't know who manages your server in the first place, book a Free Automation Audit. We'll check what your site runs on, whether this Apache bug touches you, and what it would take to keep your website patched without you having to think about it.

Sources