Vulnerabilities

Threat intelligence · GHSA-2hw4-c9m4-f2fj

knative.dev/pkg admission webhooks allow unbounded request-body memory exhaustion

Published Oct 28, 2026 · catalog updated Oct 28, 2026 · knative.dev/pkg admission webhooks allow

Last checked Oct 30, 2026, 9 days ago.

Open the advisory

Exploitation likelihood is not on this record for GHSA-2hw4-c9m4-f2fj. A missing EPSS score is not a low risk. Compare knative.dev/pkg admission webhooks allow with the linked advisory before you change production.

Severity

Medium6.5

Known exploitation

Not in CISA/VulnCheck KEV

EPSS

Awaiting synchronization with FIRST

Remediation

Version ranges on file

Affected products

go knative.dev/pkg

## ⚠️ Update (verification, 2026-05-19): C-2 confirmed end-to-end — **4.5× amplification observed** (worse than the advisory's conservative 2× claim) Built a Go PoC ([ ](https://github.com/knative/pkg/blob/1ad5a1ce0ee7043081cb50fd1acc6d36a9126844/cmd/poc-c2/main.go)) that runs the exact 8-line vulnerable snippet from [ ](https://github.com/knative/pkg/blob/1ad5a1ce0ee7043081cb50fd1acc6d36a9126844/webhook/admission.go#L109-L117) verbatim under the same shape (the full production at requires + + an informer-synced channel + controller-runtime / client-go scaffolding — the bug is the missing , nothing else in the handler). Bound to . PoC client streamed 2 GiB of bytes over a raw TCP socket. sampler: The advisory predicted ~2× from doubling the body once in . The observed 4.5× ratio comes from the interaction of two amplifiers: (used by -fed ) doubles its backing slice on every grow, so peak occupancy during a realloc is the live data — and holds its own working buffer in parallel. The advisory's 2× floor is comfortably exceeded. **Downstream impact reminder**: this library backs every Knative serving / eventing controller, every Tekton Triggers / Pipelines webhook, KEDA, Sigstore policy-controller, Kpack, and many more. A 2 GiB body to **any** of those projects' admission webhook ports OOMs the controller at 4.5× the body — i.e., even a 128 MiB request OOMs a 512 MiB-limited Pod. --- # Summary 's admission webhook handler reads the incoming HTTP request body via with **no wrapping the body**. Because every consumer of inherits this handler (Knative serving / eventing, Tekton Pipelines, KEDA, Sigstore policy-controller, and others), a single oversized POST to **any of those projects' admission webhook ports** drives the controller pod into memory exhaustion and OOM-kill. Worse, the design keeps a verbatim copy of the body in while the decoder also materializes the parsed struct — so the peak memory usage is roughly **twice the body size**. This is the same shape as **CV