Bug 2514853

Summary: CVE-2026-43632 ollama: llama.cpp: Potential code execution via a race condition in tokenization endpoints [fedora-all]
Product: [Fedora] Fedora Reporter: Joel Chamberlain <jochambe>
Component: ollamaAssignee: Tom.Rix
Status: NEW --- QA Contact:
Severity: high Docs Contact:
Priority: high    
Version: rawhideCC: go-sig, lx, rocm-packagers-sig, Tom.Rix
Target Milestone: ---Keywords: Security, SecurityTracking
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard: {"flaws": ["36225ca8-b123-4047-b2c5-9fc6eddd012f"]}
Fixed In Version: Doc Type: ---
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Bug Depends On:    
Bug Blocks: 2512244    

Description Joel Chamberlain 2026-08-12 18:36:20 UTC 
Disclaimer: Community trackers are created by Red Hat Product Security team on a best effort basis. Package maintainers are required to ascertain if the flaw indeed affects their package, before starting the update process.

llama.cpp builds b7492 through the latest b9060 contains a use-after-free vulnerability in llama-server affecting six tokenization endpoints (/tokenize, /detokenize, /infill, /apply-template, /rerank, and /anthropic/count_tokens) that bypass the task queue and access ctx_server.vocab directly on HTTP worker threads. Attackers can exploit a time-of-check-time-of-use race condition where the main thread destroys and frees vocab after the synchronization lock is released but before the handler finishes using it, causing a crash or potential code execution when --sleep-idle-seconds is configured.