Go · github.com/zalando/skipper
Skipper: Unbounded Request Body Read in Admission Webhook Causes Memory Exhaustion DoS
The Kubernetes admission webhook handler reads the entire request body using io.ReadAll(r.Body) without any size limit. Any client that can reach the webhook port within the cluster can send a multi-GB payload, causing the skipper process to exhaust memory and be OOM-killed. This disrupts all Kubernetes admission control, potentially blocking all pod creation and updates.
// dataclients/kubernetes/admission/admission.go:76
body, err := io.ReadAll(r.Body) // <-- NO SIZE LIMIT
if err != nil {
log.Errorf("Failed to read request: %v", err)
w.WriteHeader(http.StatusInternalServerError)
invalidRequests.WithLabelValues(admitterName).Inc()
return
}
var review admissionReview
err = json.Unmarshal(body, &review)
For comparison, the OPA filter has a body size limit:
// filters/openpolicyagent/openpolicyagent.go:68-70
const DefaultMaxRequestBodySize = 1 << 20 // 1MB
// OPA uses a bufferedBodyReader with size limits
:9443/admission or configured path)POST /admission HTTP/1.1, Content-Type: application/json with a multi-GB request bodyio.ReadAll(r.Body) allocates unbounded memory for the entire bodyDefaultMaxRequestBodySize (1MB) and semaphore-based memory limit; admission handler has neither| File | Lines | Description |
|---|---|---|
dataclients/kubernetes/admission/admission.go |
76 | io.ReadAll(r.Body) without size limit |
filters/openpolicyagent/openpolicyagent.go |
68-70 | OPA filter has DefaultMaxRequestBodySize = 1MB |
filters/openpolicyagent/openpolicyagent.go |
1333-1336 | OPA uses bufferedBodyReader with size limits |
dataclients/kubernetes/admission/admission_test.go exists but does not test body size limitsThe admission webhook handler reads the entire request body using io.ReadAll(r.Body) without a size limit. An attacker with in-cluster network access and a valid Kubernetes client certificate can send a multi-GB payload to the webhook endpoint, causing the skipper process to exhaust memory and be OOM-killed. This disrupts admission control for Ingress and RouteGroup resources until the process is automatically restarted by the kubelet.
Scope of impact: Ingress and RouteGroup admission only — not pod creation or other admission controllers.
Recovery: Kubernetes automatically restarts the OOM-killed process, limiting downtime.
Prerequisites: (1) In-cluster network access to the webhook port, (2) valid Kubernetes client certificate.
http.MaxBytesReader or equivalent body size limit before io.ReadAllDefaultMaxRequestBodySize and use a buffered reader with size limits--admission-max-body-size flagIs your project exposed to this? Stateward checks every dependency on every pull request and flags it only if your code actually reaches it.
Check my repoSources: CISA KEV (public domain), OSV.dev & GitHub Advisory Database (CC-BY-4.0), FIRST EPSS, NVD/CWE (public domain). Served live from the Stateward advisory database.