Regex Name Finder Vulnerability in Apache OpenNLP
CVE-2026-82617

10CRITICAL

Key Information:

Vendor

Apache

Vendor
CVE Published:
11 September 2026

What is CVE-2026-82617?

CVE-2026-82617 is a vulnerability found in Apache OpenNLP, a popular open-source library used for natural language processing tasks, including named entity recognition. This specific vulnerability pertains to the Regex Name Finder component of OpenNLP, which utilizes built-in patterns to identify email addresses and URLs from text. The flaw originates from the existence of ambiguous nested quantifiers in the regex patterns for email and URL detection.

When an application employs these patterns to process untrusted input, it may encounter severe performance issues due to super-linear backtracking or unbounded recursion. For instance, the EMAIL pattern can lead to extensive CPU consumption, potentially taking several seconds for 32 KB of input with no resulting match. Similarly, the URL pattern may cause a StackOverflowError due to excessive recursion when processing input as small as 1 KB in environments with limited thread stack sizes. This can lead to denial-of-service (DoS) conditions as the application can become unresponsive or crash.

This vulnerability affects multiple versions of Apache OpenNLP, specifically from versions 2.0.0 through 2.5.11 and 3.0.0-M1 through 3.0.0-M5. Users are advised to upgrade to the latest versions, which contain fixes for this issue.

Potential Impact of CVE-2026-82617

  1. Denial of Service (DoS): An attacker could craft specific inputs to exploit the vulnerability, causing the application to slow down significantly or even crash. This disrupts services and can lead to substantial downtime for organizations relying on NLP functionalities in their applications.

  2. Resource Exhaustion: The exploitation may result in excessive consumption of CPU resources, which can impact not only the affected application but also other processes running on the same server. This may lead to an overall degradation of system performance and responsiveness.

  3. Ease of Exploitation: The vulnerability does not require any specialized knowledge or access controls to exploit, making it accessible to a broader range of potential attackers. Any adversary capable of submitting text for analysis can trigger the vulnerability, heightening the risk for organizations utilizing OpenNLP in production environments.

Affected Version(s)

Apache OpenNLP 3.0.0-M1 < 3.0.0-M6

Apache OpenNLP 2.0.0 < 2.5.12

References

CVSS V4

Score:
10
Severity:
CRITICAL
Confidentiality:
High
Integrity:
High
Availability:
High
Attack Vector:
Network
Attack Complexity:
Low
Attack Required:
None
Privileges Required:
Undefined
User Interaction:
None

Timeline

  • Vulnerability published

  • Vulnerability Reserved

Credit

n0mi1k
.