OpenSSL CVE-2026-35191

Requires a server deployed using OpenSSL's QUIC implementation (versions 3.5.0-3.5.8, 3.6.0-3.6.4, or 4.0.0-4.0.2) reachable over the network, plus the ability to spoof source IP addresses so the over-accounted amplification credit directs reflected traffic at a chosen victim. The weakness lives in QUIC's address-validation/anti-amplification path before the client address is validated, so no authentication (PR:N) or user interaction (UI:N) is needed, but AC:H reflects the specific timing/traffic conditions required to realize meaningful amplification. Exploitation is limited to deployments that actually use the OpenSSL QUIC stack (a relatively new feature of the 3.5+/4.0 series), not TLS-only OpenSSL users, and impact is bounded to amplified traffic volume rather than code execution or data exposure.

This is a low-real-risk issue despite being part of a security release whose most severe CVE is rated High (that High rating belongs to a different CVE in the OpenSSL 4.0.3 batch, not this one). The CVSS base score is only 3.7 with CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:N/I:N/A:L - network vector but high attack complexity, no privileges, and only a low availability impact with zero confidentiality or integrity impact. All three decision signals align downward: SSVC reports Exploitation=none, Automatable=no, Technical Impact=partial, and there is no EPSS score, no CISA KEV listing, and no public PoC identified. The flaw is an under-enforcement of QUIC's anti-amplification credit (the server may over-account the 3x unvalidated-data limit), meaning the OpenSSL QUIC server could be coaxed into reflecting/amplifying traffic toward a spoofed victim rather than being directly compromised itself - this is why impact is availability-only and modest. Note the vendor scored scope S:U even though an amplification/reflection weakness primarily harms a third-party victim, which could arguably justify S:C or a CVSS 4.0 subsequent-system availability impact; this is a defensible modeling nuance, not an error. Overall this is a correctly-rated low-severity hardening fix, not a priority patch on its own - though operators should still take the 4.0.3 upgrade for the batch's higher-severity QUIC/DTLS fixes.

An attacker spoofs a victim's source IP address and initiates QUIC handshakes against an internet-facing server built on a vulnerable OpenSSL QUIC stack. Because the server over-accounts its anti-amplification credit, it sends more response data to the spoofed address than the QUIC 3x limit should allow, turning the server into a traffic amplifier that floods the victim. No authentication or user interaction is required, but exploitation is bandwidth-limited and high-complexity; no public exploit identified at time of analysis.