CVE-2026-50012

Отчет

To exploit this issue, an attacker must control a trusted cache peer server. Also, cache digests are not enabled in the default configuration. Squid deployments that do not use cache peering are not affected. Furthermore, even those that do are only vulnerable when the attacker controls a configured peer server within the same administrative domain. A compromised peer can reliably crash the Squid process via a heap-based buffer overflow during digest exchange, but code execution faces considerable practical security barriers. Default Red Hat Enterprise Linux security features, including SELinux enforcement, Address Space Layout Randomization (ASLR) and NX (No-Execute) stack protection, significantly increase the difficulty of achieving arbitrary code execution, limiting the impact of this vulnerability. Due to these reasons, this vulnerability has been rated with a moderate severity.

Меры по смягчению последствий

To mitigate this vulnerability, restrict or disable cache digest processing for cache peers by adding the appropriate settings to the squid.conf configuration file: Disable cache digests for any cache_peer not under direct control by adding the no-digest option. This completely removes the vulnerability for those peers at the cost of increased bandwidth. Use of HTCP or ICP as alternative protocols can reduce that cost.

Audit the Squid configuration and ensure that all configured cache_peer entries are under direct control and are trusted. Restricting cache peering exclusively to trusted domains greatly reduces the risk of exploitation.