CVE-2026-24117: Rekor SSRF Vulnerability

CVE-2026-24117 Overview

CVE-2026-24117 is a Server-Side Request Forgery (SSRF) vulnerability affecting Rekor, the software supply chain transparency log developed by Sigstore. The vulnerability exists in the /api/v1/index/retrieve endpoint, which supports retrieving a public key via a user-provided URL. This functionality can be abused by attackers to trigger arbitrary GET requests to internal services, potentially enabling reconnaissance and probing of internal network infrastructure.

Critical Impact

Attackers can leverage this Blind SSRF vulnerability to probe internal networks and services, potentially mapping internal infrastructure and identifying other attack surfaces within the target environment.

Affected Products

• Rekor versions 1.4.3 and below
• Sigstore Rekor transparency log server instances with the retrieve API enabled

Discovery Timeline

• January 22, 2026 - CVE-2026-24117 published to NVD
• January 22, 2026 - Last updated in NVD database

Technical Details for CVE-2026-24117

Vulnerability Analysis

This Server-Side Request Forgery vulnerability allows remote attackers to abuse the Rekor server as a proxy for making HTTP GET requests to arbitrary internal services. The vulnerability is classified under CWE-918 (Server-Side Request Forgery). The flaw is accessible over the network without requiring authentication or user interaction, though the exploitability is limited since only GET requests can be triggered and responses are not returned to the attacker.

The vulnerability represents a significant security concern for organizations running Rekor as part of their software supply chain security infrastructure, as it could allow attackers to enumerate internal services, identify open ports, and map network topology from an externally-accessible endpoint.

Root Cause

The root cause of CVE-2026-24117 lies in insufficient validation of user-supplied URLs in the /api/v1/index/retrieve endpoint. When a request is made to retrieve a public key, the endpoint accepts a URL parameter from the user and performs an HTTP GET request to that URL without proper validation or restrictions on the target destination.

The endpoint fails to implement adequate safeguards such as URL allowlisting, blocking of private IP ranges (RFC 1918), or restricting requests to known-safe external domains. This allows attackers to specify internal IP addresses or hostnames as the URL target.

Attack Vector

The attack can be executed by any unauthenticated network attacker who can reach the Rekor server's API endpoint. The attacker crafts a request to /api/v1/index/retrieve with a URL parameter pointing to an internal service or IP address. While the SSRF is "blind" (the response is not returned to the attacker), timing-based analysis and error responses can reveal information about the internal network:

The attacker could probe internal services by observing response times and status codes. For example, requests to active internal services may return faster than requests to non-existent hosts. This technique allows systematic enumeration of internal network resources without direct visibility into the response content.

Detection Methods for CVE-2026-24117

Indicators of Compromise

• Unusual HTTP requests to the /api/v1/index/retrieve endpoint with URLs pointing to internal IP ranges (10.x.x.x, 172.16.x.x-172.31.x.x, 192.168.x.x)
• High volume of requests to the retrieve endpoint from a single source IP
• Requests containing localhost, 127.0.0.1, or internal hostnames in the URL parameter
• Sequential scanning patterns indicating automated network enumeration attempts

Detection Strategies

• Monitor Rekor server access logs for suspicious URL patterns in retrieve API requests
• Implement network monitoring to detect outbound connections from the Rekor server to internal services
• Configure intrusion detection rules to alert on SSRF payload patterns in HTTP requests
• Deploy web application firewall rules to inspect and block requests with internal IP addresses in URL parameters

Monitoring Recommendations

• Enable verbose logging on the Rekor server to capture full request details including URL parameters
• Set up alerting for anomalous request patterns to the /api/v1/index/retrieve endpoint
• Monitor for DNS queries from the Rekor server to internal hostnames that should not be accessed
• Track connection attempts from the Rekor server to internal network segments

How to Mitigate CVE-2026-24117

Immediate Actions Required

• Upgrade Rekor to version 1.5.0 or later, which contains the fix for this vulnerability
• If immediate upgrade is not possible, disable the vulnerable retrieve API endpoint using --enable_retrieve_api=false
• Review network segmentation to limit the Rekor server's ability to reach sensitive internal services
• Audit access logs for any signs of prior exploitation attempts

Patch Information

The vulnerability has been addressed in Rekor version 1.5.0. Organizations should upgrade to this version as soon as possible. The fix is available in GitHub commit 60ef2bceba192c5bf9327d003bceea8bf1f8275f. Additional details are available in the GitHub Security Advisory GHSA-4c4x-jm2x-pf9j and the v1.5.0 release notes.

Workarounds

• Disable the retrieve API endpoint by starting Rekor with --enable_retrieve_api=false
• Implement network-level controls to restrict outbound connections from the Rekor server
• Deploy a reverse proxy or WAF in front of Rekor to filter requests containing internal IP addresses
• Apply egress filtering to prevent the Rekor server from connecting to internal network ranges

# Disable the vulnerable retrieve API endpoint
rekor-server serve --enable_retrieve_api=false