Skip to main content
Exostruction
Back to overview
Alexander Kreitmair5 min read

Securing Remote Access: Why VPN Gateways Are an Attractive Target

VPN gateways are an attractive target for attackers. How small businesses secure remote access with updates, a second factor and segmentation.

Guide
Securing Remote Access: Why VPN Gateways Are an Attractive Target

The short answer: a VPN gateway has to be reachable from the internet and is therefore an attractive target. Risk drops mostly through three things: prompt security updates, a second factor before access, and a network where a successful login does not automatically mean access to everything, in other words segmentation after login. The five questions for your remote access help you place your own situation quickly.

Why VPN gateways are an attractive target

Remote access promises attackers a direct path into the company network. In their joint guidance, NSA and CISA describe remote access VPN servers as entry points into protected networks and therefore attractive targets. According to CISA, a successful attack on the device can lead to stolen credentials, code execution on the VPN device or hijacked encrypted sessions (CISA). According to the NSA, the key recommendations include tested VPN products, strong authentication such as multi-factor authentication, prompt updates, and a smaller attack surface by disabling features that are not part of the VPN (NSA).

The guidance dates from 2021, but the recommendations are still a useful benchmark. The risk grows when a gateway runs unchanged for years and nobody is responsible for updates.

A recent example from June 2026

On 8 June 2026, the US agency CISA added the vulnerability CVE-2026-50751 to its catalog of actively exploited vulnerabilities (CISA). It affects Check Point Remote Access VPN and Mobile Access when they use the deprecated IKEv1 key exchange: a remote attacker can establish a VPN connection without knowing a valid password (NVD).

According to the vendor, further steps are needed after the connection is established to reach internal resources, and the observed exploitation was limited to a few dozen targeted organizations worldwide. Check Point names 7 May 2026 as the earliest observed exploitation, and the hotfix was released with the advisory on 8 June (Check Point).

Our assessment: this concerns one product in specific configurations, not VPNs in general. Two things are still worth taking away. First, the vulnerability was in use for about a month before the advisory, and no update could help during that time. Second, further steps were needed after login, and that is exactly where it is decided how far a breach reaches.

Three levers for remote access

  1. Patch promptly. Everything reachable from the internet gets security updates without delay. Our recommendation: a fixed time window of days. The first step is an inventory: which gateways, firewalls and routers are exposed, and who is responsible for them?
  2. A second factor before access. Ideally through a central login, so the check is tied to the identity and not to an add-on feature of a single device.
  3. Shrink the attack surface. Old protocols, legacy clients and features nobody uses anymore get switched off. In the example above, a deprecated protocol was the starting point.

The effort for these points is manageable. They do not answer what an attacker can reach who gets in anyway.

After login: segmentation instead of a flat network

In a flat network, a connected device can, depending on the configuration, reach a great many systems: file servers, administration, backups, other sites. A single hijacked account then quickly becomes major damage. Segmentation reverses the principle: a device reaches only the systems it has an explicit rule for.

CriterionFlat network behind the gatewaySegmented network with groups and rules
Access after logindepending on configuration, large parts of the internal networkonly what a rule explicitly allows
Consequence of a breachattacker can move laterallylimited to the allowed targets
Maintenance effortfew rules, but gateway updatesgroups and rules have to be maintained
Typical mistakerules too broad, forgotten legacy accessexceptions that are never cleaned up

The table describes the principle, not individual products. Segmentation is possible with classic gateways as well as with modern mesh or zero-trust solutions, it just takes the discipline to allow no connection without an explicit rule.

A VPN is not a wall, it is a door. What matters is what can be reached behind it.

Our recommendation: five questions for your remote access

  1. What is reachable from the internet, and who is responsible for its updates?
  2. How quickly are vulnerabilities closed that are already being actively exploited, for example those in the CISA catalog of known exploited vulnerabilities?
  3. Is there a second factor before anyone gets into the network?
  4. Which old protocols and features are still active although nobody needs them anymore?
  5. What can a connected device reach, and does it really need that?

If you cannot answer these questions with confidence, you have a good starting point. When we set up servers, firewall, SSH hardening and automatic security updates are part of the hardening, more on that is on our page about Linux servers. If you want your infrastructure reviewed: our IT consulting offers a security check that looks for vulnerabilities in code and infrastructure.

What segmentation does not solve

  • Compromised but authorized devices. A hijacked admin laptop uses the same allowed connection as a legitimate admin. The content of an allowed connection is not inspected by segmentation.
  • State of the endpoints. Patch level and disk encryption are not checked automatically.
  • The management layer. Whoever may change the rules can lift the protection. These accounts need the same protection as the rules themselves.
  • Everything underneath. Server hardening, patch management and monitoring remain necessary. Segmentation complements them but does not replace them.
Share

Frequently asked questions

  • No. A VPN encrypts the connection but does not decide who may log in or what they can see afterwards. You also need a second factor before access, timely security updates for the device at the edge of the network, and segmentation so that one hijacked account does not open the whole network.

  • The authorities NSA and CISA recommend applying updates promptly. Our assessment: for vulnerabilities that are already being actively exploited, for example those listed in the CISA catalog of known exploited vulnerabilities, immediately and not at the next maintenance window. For everything reachable from the internet we recommend a fixed time window of days.

  • Segmentation means that a connected device reaches only the systems it has an explicit rule for, instead of seeing the whole internal network after login. This limits how far a hijacked account reaches. It works with classic gateways as well as with mesh or zero-trust solutions and replaces neither updates nor a second factor.

Author of this article

AK

Alexander Kreitmair

Co-Founder & Infrastructure

Certified systems integration specialist. Makes sure the systems keep running - servers, networks, deployment pipelines.

This article is for general information only and does not replace individual legal, tax, or business advice. Prices, timelines, and assessments are non-binding estimates provided without guarantee.