Written by: Mio Kanzaki(Security Analyst (AI))Reviewed by: 千葉圭介Updated
Virtual patching is a stopgap that blocks only the traffic aimed at a hole, before a fixed version exists. It leaves the site's program untouched and spots and stops the traffic that targets the hole in front of the site. OWASP's material describes it as a layer that detects and blocks attempts to exploit a known vulnerability without changing the code.
How does virtual patching work?
A leaking roof makes a good comparison. Suppose the roofer can only come next week. In the meantime, a blue tarp over the spot where the rain gets in limits the damage. Virtual patching is that tarp.
In practice, a rule is added to the layer that inspects traffic: traffic of this shape is aimed at this hole, so stop it. The program with the hole stays as it is, but the traffic that targets it no longer arrives. It is often used on top of a mechanism that inspects traffic, such as a WAF.
OWASP lists situations for using it: the developers have no time, the software is from another company and cannot be fixed by you, or development is outsourced and a fix needs a new order. It also helps when you want to shorten the time before a fix. Unpatched software explains why late updates are a problem.
Why does virtual patching matter?
Because there is a time during which you must wait with the hole open until a fixed version is out. When the maker has not released a fix, there is not yet an update the user can apply. Virtual patching fills that gap, and it matters most for a zero-day, where no fixed version exists.
Its limits are clear, too. The hole itself remains, so it protects only the traffic that gets inspected, and it can be slipped past if the form of the attack changes a little. If the rules are too broad, they can also block legitimate customers.
The basic way to use it is as a temporary measure: put it in, update the moment the fixed version is out, and then remove rules that are no longer needed. OWASP's Top 10 also points to virtual patching as something to consider when a part cannot be fixed. Our Cyber Shield Guard Pro includes virtual patching.
When should you use it?
- A plugin has a published hole, and the maker's fix is not out yet.
- You cannot update right away because the update may break the page, or the agency cannot be reached.
- A part is no longer maintained and you cannot replace it quickly.
In each case, set a date to review it. To decide whether a stopgap suits your site, contact us.
Related terms
- WAF: a wall that stops malicious traffic at the entrance
- Zero-day: a vulnerability with no fixed version yet
- Vulnerability: a security hole in a program or its settings
- CVE: the worldwide number that identifies a hole
Frequently asked questions
If I apply a virtual patch, can I skip the update?
No. A virtual patch does not close the hole. It only blocks traffic heading for it. OWASP says fixing the code and virtual patching go together, and that the virtual patch is reviewed and removed once the real fix is in.
Does virtual patching work on every kind of hole?
Its reach is uneven. OWASP's material says it suits injection-type holes, where malicious input is mixed in, while for other kinds of attack the range it can protect may not be enough. You have to judge how well it works for each type of hole.
Are virtual patching and a WAF the same thing?
No. A WAF is the general name for a mechanism that inspects and blocks traffic. Virtual patching is a way of using such a mechanism by adding rules for a known hole. Even with a WAF, without a rule matched to the hole it is not a virtual patch.
Next step
If a plugin on your site has a hole that cannot be updated, contact us and tell us your situation. The free URL check shows only what is visible from outside. The cost of waiting is described in unpatched software.
Frequently asked questions
If I apply a virtual patch, can I skip the update?
No. A virtual patch does not close the hole. It only blocks traffic heading for it. OWASP says fixing the code and virtual patching go together, and that the virtual patch is reviewed and removed once the real fix is in.
Does virtual patching work on every kind of hole?
Its reach is uneven. OWASP's material says it suits injection-type holes, where malicious input is mixed in, while for other kinds of attack the range it can protect may not be enough. You have to judge how well it works for each type of hole.
Are virtual patching and a WAF the same thing?
No. A WAF is the general name for a mechanism that inspects and blocks traffic. Virtual patching is a way of using such a mechanism by adding rules for a known hole. Even with a WAF, without a rule matched to the hole it is not a virtual patch.