CVE-2026-17557
A flaw was found in the `lm-evaluation-harness` component of Red Hat OpenShift AI. An authenticated user, able to submit an evaluation job,
CVSS
6.5
Média
EPSS
0.3%
p26
KEV
—
Exploit Today
8
0-100
Publicado: 21 de out. de 2026 · Última mod.: 22 de out. de 2026 · CWE-918
0.3%EPSS · 30 dias0.3%
2026-10-212026-11-07
A flaw was found in the `lm-evaluation-harness` component of Red Hat OpenShift AI. An authenticated user, able to submit an evaluation job, could exploit this by providing a malicious model URL in the `JobSpec`. This vulnerability allows for the disclosure of the tenant's model authentication secret to an attacker-controlled external listener. Additionally, it enables Server-Side Request Forgery (SSRF) attacks against internal network endpoints reachable from the evaluation job pod.
CVECVSSEPSSKEVRExploitTítuloMod.
CVE-2026-1235526.5 MÉD—
———OpenAM before 16.1.4 contains a server-side request forgery vulnerability in certificate CRL checking that fetches CRLs from the distribution point named in a presented certificate without restricting the destination or verifying the downloaded CRL's signature. Attackers holding a trusted-CA-chained certificate can make OpenAM request internal hosts, capture configured CRL update credentials, overwrite stored revocation data to block certificate authentication for that CA, and tie up authentication threads.12hCVE-2026-1235375.0 MÉD—
———OpenAM before 16.1.4 contains a server-side request forgery vulnerability that allows parties registering UMA resource sets to trigger internal requests through unvalidated scope description URLs. Attackers can register resource sets with scope URLs pointing to internal hosts or cloud metadata endpoints, which OpenAM fetches when sending pending-request notification emails, enabling internal network probing.12hCVE-2026-1235364.1 MÉD—
———OpenAM before 16.1.4 contains a server-side request forgery vulnerability that allows administrators, including delegated realm administrators, to make the server fetch arbitrary URLs through federation configuration wizard metadata URLs. Attackers can enter metadata or OpenID Connect discovery URLs pointing to internal hosts or cloud metadata endpoints to probe the internal network and read partial responses.12hCVE-2026-755907.2 ALT29.0%
——9IBM API Connect 12.1.0.0 through 12.1.1.2 could allow a remote attacker to access internal network resources and obtain sensitive information due to server-side request forgery. A protection that blocks requests to internal addresses can be bypassed, so the server could be made to contact internal services and return part of their responses.1dCVE-2026-191287.5 ALT33.8%
——10IBM API Connect 12.1.0.0 through 12.1.1.2 could allow an authenticated attacker to send unauthorized requests from the system due to server-side request forgery. This could be used to enumerate network resources or to support further attacks.12hCVE-2026-191223.8 BAI17.2%
——5IBM API Connect 12.1.0.0 through 12.1.1.2 could allow an authenticated privileged attacker to send unauthorized requests from the system due to server-side request forgery. This could be used to enumerate network resources or to support further attacks.1d