Apache Struts2 lets attackers run code on the server
Struts is a tool many websites use to build their pages. A flaw can let an attacker send a specially crafted request that makes the server run commands they choose.
| Severity | CriticalVulnerability |
|---|---|
| Being exploited? | UnclearNo reports either way yet |
| Fix available? | UnclearNo fixed version published in the sources yet |
| How sure is this? | 1 source, unconfirmedOnly one publisher so far, details may change |
How the attack works simplified
- 1Attacker sends crafted requestRequest contains a malicious OGNL expression
- 2Sandbox check passesAllowed method accepts nested arguments
- 3Reflection runs uncheckedReflection call executes outside the sandbox
- 4Code runs on serverAttacker-chosen code executes server-side
- ResultAttacker controls the server
What happened
Apache Struts2 Core 7.4.0 and below has an OGNL sandbox escape: an allowed method that accepts nested arguments and invokes reflection internally runs outside the sandbox's restrictions, enabling arbitrary code execution.
Are you affected?
You are affected if you run Apache Struts2 Core version 7.4.0 or below in a web application that processes OGNL expressions.
Apache Struts2 Core 7.4.0 and below
What to do
- Check whether your applications use Apache Struts2 Core 7.4.0 or below
- Watch Apache Struts security advisories for a fixed version and update as soon as one is released
- Limit network access to affected Struts2 applications until a fix is confirmed
Timeline
- 2 Nov 2026, 06:10 UTCPublicly disclosed
- 2 Nov 2026, 06:10 UTCReported by Full Disclosure · Full Disclosure
- 2 Nov 2026, 06:31 UTCPicked up by Zero Radar
Sources 1 report, oldest first
Security researchers research teams and mailing lists
- Apache Struts2 Core Arbitrary Code Execution via OGNL Sandbox EscapeFull Disclosure · 2 Nov 2026, 06:10 UTC
Technical details
- ID
- ZR-6d08cb75
- Disclosed
- 2 Nov 2026, 06:10 UTC
- On Zero Radar
- 2 Nov 2026, 06:31 UTC (20 min after disclosure)
- Last change
- 2 Nov 2026, 06:31 UTC
The plain-language parts and the attack diagram were written by an AI model (claude-haiku-5-5) from the sources above. Facts like versions, exploitation and dates come from the sources themselves. How Zero Radar works