Code execution via unsafe crash-dump deserialization
Published Mar 12, 2026 · Updated Mar 12, 2026
Unsafe deserialization in SGLang 0.5.5 through 0.5.9 allows attackers to execute code via a crafted request-dump file. The read_records function in replay_request_dump.py passes each .pkl input directly to pickle.load, which invokes attacker-selected reduction callables. Execution requires an operator to replay a planted or socially supplied file and runs with the script process's privileges.
Summary
What happened
Unsafe deserialization in SGLang 0.5.5 through 0.5.9 allows attackers to execute code via a crafted request-dump file. The read_records function in replay_request_dump.py passes each .pkl input directly to pickle.load, which invokes attacker-selected reduction callables. Execution requires an operator to replay a planted or socially supplied file and runs with the script process's privileges.
The record
- CVE
- CVE-2026-3989
- Published
- Mar 12, 2026
- Updated
- Mar 12, 2026
- Vendor
- SGLang
- Product
- SGLang
- Classifications
- CWE-502, T1204.002
- Attack vector
- local
- Privileges
- unauthenticated
Timeline
How it unfolded
- Mar 12, 2026CVE publishedPublication date reported by the CVE source.
- Mar 12, 2026Record updatedLatest update available in the CVE record.
Exploitability
Present is not the same as exploitable
Compare your product and version with the public record. A matching version still requires validation against your environment.
Is a vulnerable build present?
Compare these published version ranges with your installed build and any vendor patches.
- Affected versionversion=0.5.5 <=0.5.9
What conditions does exploitation require?
What is affected?
Published CVSS scores
CVSS describes severity. EPSS estimates exploitation probability.
Attacks
What attackers are doing with it
Daily unique IPs observed by Shadowserver honeypots for known exploited vulnerabilities (KEVs). Missing observations do not establish an absence of attacks.
Weakness, pattern, technique
Public exploit references
- CVE-2026-3989 malicious pickle proof of conceptproof of concept · demonstrated
Labels summarize the accepted research assessment. They do not indicate a test against your environment.
Technologies
Your stack
See the directory against your own environment.
Your stack
Check the software in your environment
Book a demo to see how Hinoki identifies affected software and validates exploitability in your environment.
Book a demo