From b695cee9873b04364abb826415709bb06599d48f Mon Sep 17 00:00:00 2001
From: Story Crater Bot <19826264+Riotpiaole@users.noreply.github.com>
Date: Thu, 13 Aug 2026 07:55:21 -0700
Subject: [PATCH] =?UTF-8?q?fix(api):=20label=20Kong=20pods=20llm-client=3D?=
 =?UTF-8?q?true=20so=20llm-serving=20NetworkPolicy=20admits=20them=20?=
 =?UTF-8?q?=E2=80=94=20chat/embeddings/rerank/score=20routes=20silently=20?=
 =?UTF-8?q?hung=20until=20the=20client=20timeout=20because=20Cilium=20drop?=
 =?UTF-8?q?ped=20Kong's=20packets?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

llm-serving-default-deny admits port 8080 only from pods carrying
llm-client=true. Kong lacked it, so every route that actually contacts an
upstream timed out. /v1/models masked the problem: request-termination answers
inside Kong and never touches an upstream, so it returned 200 throughout.

Opting in via podLabels rather than relaxing the policy — it is a compensating
control, not hygiene, since vLLM v0.11.0 is frozen on Volta and will not receive
patches for several remote/unauthenticated advisories.

podLabels land only in the pod template, not spec.selector.matchLabels, so this
is not an immutable-field change.
---
 k8s/apps/api/kong-values.yaml | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/k8s/apps/api/kong-values.yaml b/k8s/apps/api/kong-values.yaml
index 117285e..ac69f34 100644
--- a/k8s/apps/api/kong-values.yaml
+++ b/k8s/apps/api/kong-values.yaml
@@ -29,6 +29,18 @@ fullnameOverride: kong
 # stateless in DB-less mode, so replicas are pure redundancy.
 replicaCount: 2
 
+# Opt in to the `llm-serving-default-deny` NetworkPolicy, which admits port 8080
+# only from pods carrying this label. That policy is a compensating control, not
+# hygiene: vLLM v0.11.0 is frozen on Volta and will never receive patches for
+# several remote/unauthenticated advisories, so it must not be broadly reachable.
+#
+# Without this label Cilium DROPS the packets rather than refusing them, so the
+# symptom is a request that hangs until the client's timeout — not a connection
+# error. /v1/models still worked while this was missing, because
+# request-termination answers inside Kong and never touches an upstream.
+podLabels:
+  llm-client: "true"
+
 env:
   # DB-less. Config comes from Kubernetes objects via the ingress controller, so
   # git stays the source of truth. A Postgres-backed Kong would put live routing