Blog
EnglishPolski

A vulnerability report. First check what is missing from it

A transport company with several branches and its own freight forwarding. The compliance manager receives the monthly vulnerability report: two critical items, both being fixed, deadlines met 98 percent of the time. She forwards it to the board with a note that things look good.

A month later it turns out the report covered 14 servers. The company has 31. Some run Windows, which the scanner does not support. Two were set up at the Poznań branch and nobody installed the agent on them. The fleet tracking system runs at a supplier, in the cloud. One server stopped sending results three weeks ago because its address changed after an update.

The report did not lie. It described exactly the servers it knew about. The problem was that nobody asked which ones it did not know about.

What the law says

The Polish National Cybersecurity System Act (KSC), as amended in 2026, lists asset management in Article 8(1)(2)(m) among the measures the information security management system must provide. The NIS2 Directive lists it in Article 21(2)(i) together with access control policies and human resources security.

The law does not say how to keep an asset register or with what tool. It does, however, imply something that applies to every vulnerability report: you cannot manage vulnerabilities in a system you do not know exists. A report without an asset list as a reference point does not say whether the company is secure. It only says what the visible part of it looks like.

Where the gap is

Servers missing from a report do not look like a problem. They look like good news. A server that is not in the report has no vulnerabilities in the report either. That is why scope gaps usually come out by accident, during an incident or an audit.

There are four typical causes.

The tool supports only some systems. A Linux server scanner will not see Windows, network devices or controllers at a plant. That is not a flaw, it is the scope, but someone has to know how those other systems are handled.

New systems do not get added to scanning. A server is set up in a day, and installing the agent is not part of the commissioning procedure. Six months later nobody remembers it was added.

Systems stop reporting. A changed address, an expired key, a stopped service. The last result stays in the system and looks as if everything is fine, just with an ever older date.

Some services run at suppliers. A system in a supplier's cloud will not appear in a scan of your servers. Its security has to be handled through the contract and supplier assessment, but you still need to know the system exists.

How to check it

The steps below can be done by the person responsible for KSC together with IT, in an hour. All it takes is two lists.

1. Take the list of servers from the report. Not the count, but the list with names and the date of each server's last scan.

2. Take a list from another source. The asset register, if there is one. If not, anything produced independently of the scanner: invoices for servers and cloud, the list of machines in the virtualisation system, monitoring, the fixed asset register, the list of services the company pays for.

3. Compare them and describe each difference in one word. For every system missing from the report: "unsupported" (a different type of system, needs a different way of checking), "missed" (should be there, needs adding), "retired" (no longer exists, remove from the register) or "supplier" (covered by a risk assessment, contractual requirements and checks of safeguards proportionate to the risk). Every difference gets an owner.

4. Check the last scan dates. A server that last reported three weeks ago is in the report in name only. Agree with IT after how many days without a result a server counts as stale, and who checks this.

5. Ask for the report to state what it covers. One sentence at the top: "the report covers 14 of 31 servers, 2 without a current scan, the rest handled in other ways, list attached". That sentence changes how the whole report is read and misleads nobody.

6. Add scanning to the server commissioning procedure. Installing the agent or registering the system for scanning as a checklist item, without which the server is not considered ready.

7. Repeat the comparison every quarter. The asset register and the scanning scope drift apart on their own, without bad intent. A quarterly comparison catches it before someone from outside does.

Where Secvalis helps, and where it does not

The Secvalis main dashboard shows the number of projects and machines, counts of critical and high vulnerabilities and a timeliness summary. The machine list shows each machine's status and last scan time. A machine with no scan or with a last scan older than 7 days is flagged in the machine list and counted in the management report as a stale or missing scan. Its open items are left out of timeliness measurement until the next scan, which is why the management report will not show a "Good" verdict in that case.

A machine-level check compares the latest full scan with the deadline ledger and flags machines with no entries or with a ledger older than the scan. A machine's criticality and whether it is exposed affect the remediation order without changing the severity of the vulnerability itself.

Secvalis sees only machines on which the agent has been installed. It does not discover unregistered devices on its own and is not an asset register. It covers Linux servers, their system and language packages and the images of running containers, not Windows workstations and servers, network devices, industrial controllers or services running in suppliers' clouds. The comparison with the company's full asset list, described above, has to be done outside the product.

You can see the machine list with last scan dates in the public demo, with no registration.

In the next part: the server runs, vendor support has ended, and the migration keeps waiting. Why that is a board decision, not the IT person's problem.

Sources

Weekly CVE digest

One email a week with newly published vulnerabilities worth knowing about. No account needed.

This digest covers new vulnerabilities in general, not your servers. If you want to know which of them actually run in your infrastructure, that is what Secvalis does: it scans your machines and reports only what concerns them.

I agree to receive the weekly CVE digest at the address above. I can withdraw this consent at any time using the link in the footer of every message.