Staying Secure & Up to Date – Part 1: Why CVE Alerts Should Concern You
“Another CVE in my feed – should I rebuild everything now?”
If you’re into tech, you’ve seen headlines like “Critical flaw in XYZ – patch immediately.” The jargon word behind it is CVE (Common Vulnerabilities and Exposures). Think of it as a uniform, internationally registered label for security flaws: a unique address that means “this program has a flaw at this spot that can be exploited.”
Such an address looks like CVE-2026-109618 – a kind of postal code for the vulnerability.
Every one gets its own, so vendors, scanners, and security teams talk about the same thing: “Are you
affected? Have you patched?”
Why should I care? I don’t host anything valuable
That’s the classic response – and it’s wrong for three reasons:
1. You host something valuable: yourself. Your mail, photos, password vault (e.g. Vaultwarden/Bitwarden), backups, your website. Criminals often target exactly such private systems because they are an easy target: one flaw and your whole server is under someone else’s control – often turned into a zombie for attacks against others.
2. Attacks are not personal – they are automated. Bots scan the internet around the clock for known flaws. They don’t mean you, they mean anyone. If your service has the known flaw, you’re just the next checkbox on the list. And modern AI makes this automation even tougher: AI helps attackers find flaws faster, craft deceptively real phishing messages, and adapt attacks on the fly – including against small, private targets.
3. Criminal networks love “unremarkable” services. A small blog looks harmless – that’s exactly why it’s useful: it can serve as a springboard for spam, phishing pages, or malware distribution. And the trouble stays with you: your server, your IP, your risk.
What does “actively exploited” mean?
There is a crucial distinction for prioritization. Most CVEs are theoretical flaws that might be exploited some day. Some are actually under attack right now – the pros call this “exploited in the wild.” That’s exactly what CISA’s KEV list (Known Exploited Vulnerabilities) tracks: “These flaws are being actively exploited right now.”
A real-world example from November 2026 made this painfully obvious: PHP2Shell
(CVE-2026-109618 / CVE-2026-109619). Two flaws in the core scripting language
behind countless websites that allowed unauthenticated attackers to run their own code on the
site – i.e. take control of it. Both landed on the KEV list within days. Those
who patched stayed safe; those who didn’t were on the scanners’ hit list.
Real-world example: When the supermarket fridges defrosted (12 Oct 2026)
“I don’t host anything valuable” – really? On the night of 11–12 October 2026, unknown attackers switched over cooling systems in hundreds of supermarkets across several European countries (including Intermarché, Coopérative U, Carrefour, Monoprix, E.Leclerc). The targets were the remote-management systems of refrigeration units made by Danish manufacturer Danfoss. The attackers triggered the built-in defrost cycles and let the freezers heat up to 50 °C for hours. Result: fresh and frozen goods worth tens of millions of euros had to be thrown away – in a few places there were even small fires. (Sources: BBC, Reuters, CERT-FR, 12 Oct 2026)
Broken down, this case is a perfect textbook example of everything this series covers:
1. Any internet-connected device is a target – not just “real computers.” Refrigeration controllers, routers, webcams: if the software is outdated and the device reachable, attackers care.
2. “End-of-life” means: no more updates. The manufacturer confirmed that most affected devices were old models that had not received security updates for years; others ran outdated firmware. Months earlier, a firm called Claroty had already examined exactly these devices and reported flaws – patches were rarely applied. Our lesson: “up to date” is a state that must be maintained – “it worked fine for years” does not help.
3. The authorities’ advice came in three parts: don’t expose control systems directly to the internet, change default passwords, and update firmware regularly. All points you’ll be watching on your own systems from this series onward.
“Then I’ll just update everything, right?” – almost
Keeping things up to date is right and important. But “updating” is not the same as “being secure,” for practical reasons:
• There is no “the one update.” Your server is made of an operating system, a kernel, a Docker service, container images, applications, PHP libraries … Each layer has its own update channel.
• Some updates arrive months late. On free systems the patch depends on the distribution; some packages (“Universe”) only get security updates via a free add-on subscription (Ubuntu Pro/ESM). That’s part 6.
• Some updates never arrive. When a project is abandoned or a vendor rarely rebuilds its images (see our Watchtower example in part 7), you sit on holes – until you notice it yourself.
That’s why I wrote this series: I found this out the hard way myself – again and again I faced the task of keeping my own systems reliably secure and up to date while staying calm. I turned my path into this small series: a fixed routine that asks you regularly “What do I run? Is it current? Are there known flaws?” – and, step by step, automates the answers.
Pro Corner
• CVE vs. exploit vs. 0-day:
• CVE = publicly registered vulnerability (ID, description, affected
versions)
• Exploit = code that leverages the flaw (proof-of-concept)
• 0-day = flaw with no patch yet
• KEV instead of “all CVEs”: CISA’s KEV feed
(known_exploited_vulnerabilities.json) is a curated watchlist of actually attacked flaws –
ideal as a mandatory watchlist. EPSS scores add an estimate of exploit likelihood within the next 30 days.
• In practice: many attack toolkits (Cobalt Strike, phishing kits) leverage KEV CVEs; ransomware campaigns disproportionately use KEV IDs. Automating a KEV match against your scan results (part 5) encodes the industry’s #1 prioritization rule.
Why you need this: because “up to date” is a state that must be maintained – and because you can maintain it yourself with a few repeatable steps.
Next part: Part 2 – Know what you run (inventory). (Note: further parts of this series will appear in this overview over the coming weeks.)