/

CVE-2026-27490

Severity:

HIGH RISK

(7.5)

Weak Randomness (CWE-330) in iTop Image Access Links

CVE-2026-27490

Deferred

Vulnerability type

Category:

Insufficiently Random Values

CWE:

CWE-330, CWE-331

Impact:

Unauthorized Disclosure of Inline Images

Severity:

HIGH RISK

(7.5)

CVSS version:

3.1

Catch this CVE while code is being written

Detect vulnerable code directly in pull requests, so CVEs are fixed before merge.

Summary

Combodo iTop uses a weak 24-bit pseudo-random secret to protect inline images that can be accessed without authentication. An attacker who can discover or guess the secret may gain unauthorized access to those images, particularly when the deployment exposes the relevant image links.

Affected context

The issue occurs in iTop deployments where inline images are available through unauthenticated access paths and their links or access tokens rely on a short pseudo-random secret. Risk depends on whether image URLs are exposed, whether the secret can be guessed within a practical timeframe, and what sensitive information the images contain.

Exploit preconditions

  • Attack surface: Unauthenticated inline-image access endpoints
  • Required access: No authenticated account may be required; access to image URLs or the ability to discover them is relevant
  • Trigger point: Request processing for an inline image protected by the pseudo-random secret
  • Why exploitation is possible: The protection uses a weak 24-bit value, reducing the search space and making guessing or enumeration more feasible than with a cryptographically secure secret

Why this happens

The underlying bug is the use of insufficiently unpredictable and overly short values for access protection. A 24-bit pseudo-random secret does not provide adequate entropy for protecting potentially sensitive resources, especially when the value is exposed through URLs or can be tested repeatedly. This behavior maps to CWE-330: Use of Insufficiently Random Values and is also related to CWE-331: Insufficient Entropy.

Potential impact

  • An attacker may access inline images without authorization
  • Confidential information embedded in images could be disclosed
  • Repeated guessing attempts could enable enumeration of protected image resources
  • Exposure of sensitive service-management or user-related content could occur

General mitigation guidance

Mitigation typically focuses on replacing weak resource-protection tokens with cryptographically secure, sufficiently long values.

Common defensive patterns include:

  • Generating access tokens with a cryptographically secure random number generator
  • Using sufficient entropy and avoiding short, predictable token formats
  • Binding resource access to authenticated sessions or explicit authorization checks where appropriate
  • Applying expiration, revocation, and rate limiting to temporary image-access links
  • Reducing sensitive information included in unauthenticated images
  • Running the application with least privilege and protecting image storage from direct exposure

How teams detect issues like this

Organizations can detect related weaknesses through:

  • Static analysis and code review searches for non-cryptographic random generators used in security tokens
  • Auditing token length, entropy, predictability, and reuse characteristics
  • Reviewing unauthenticated resource endpoints and URL-based authorization mechanisms
  • Dynamic testing for enumeration, repeated token guessing, missing expiration, and ineffective rate limiting
  • Policy enforcement requiring approved cryptographic randomness APIs for access-control secrets

Automated security review tools may help identify weak randomness and insecure direct resource access patterns.

Sources

SourceLink
github.comgithub.com/Combodo/iTop/commit/9c39efd9af53a1deeb578133eff7333a7b8816b0
github.comgithub.com/Combodo/iTop/security/advisories/GHSA-3jr5-rqmx-97gc

Frequently asked

questions

Potentially. If inline-image endpoints are reachable over the network without authentication, an attacker may attempt to discover or guess the weak secret remotely, subject to endpoint exposure and rate limiting.

Not necessarily. The affected access path is described as unauthenticated, although network controls, URL availability, and deployment-specific restrictions may limit exposure.

Use cryptographically secure, high-entropy access tokens and enforce authorization checks rather than relying on a short pseudo-random secret alone.

Establish minimum entropy requirements for security tokens, review unauthenticated endpoints during code assessment, apply expiration and rate limiting, and test resource-access controls for enumeration resistance.

Find out what's already
exploitable in your codebase.