CVE-2026-34204: MinIO is Vulnerable to SSE Metadata Injection via Replication Headers
(updated )
What kind of vulnerability is it? Who is impacted?
A flaw in extractMetadataFromMime() allows any authenticated user with s3:PutObject permission to inject internal server-side encryption metadata into objects by sending crafted X-Minio-Replication-* headers on a normal PutObject request. The server unconditionally maps these headers to X-Minio-Internal-* encryption metadata without verifying that the request is a legitimate replication request. Objects written this way carry bogus encryption keys and become permanently unreadable through the S3 API.
Any authenticated user or service with s3:PutObject permission on any bucket can make objects permanently unreadable by injecting fake SSE encryption metadata. The attacker sends a standard PutObject request with X-Minio-Replication-Server-Side-Encryption-* headers but without the X-Minio-Source-Replication-Request header that marks legitimate replication traffic. The server maps these headers to internal encryption metadata (X-Minio-Internal-Server-Side-Encryption-Sealed-Key, etc.), causing all subsequent GetObject and HeadObject calls to treat the object as encrypted with keys that do not exist.
This is a targeted denial-of-service vulnerability. An attacker can selectively corrupt individual objects or entire buckets. The ReplicateObjectAction IAM permission is never checked because the request is a normal PutObject, not a replication request.
Affected component: cmd/handler-utils.go, function extractMetadataFromMime().
References
Code Behaviors & Features
Detect and mitigate CVE-2026-34204 with GitLab Dependency Scanning
Secure your software supply chain by verifying that all open source dependencies used in your projects contain no disclosed vulnerabilities. Learn more about Dependency Scanning →