From a2ccad68f761594c7c34b97b199608e89653d21e Mon Sep 17 00:00:00 2001 From: HardMax71 Date: Mon, 19 Oct 2026 03:03:12 +0200 Subject: [PATCH] fix: cert-generator waits up to 300 s for Kueue's controller #402 raised e2e-ready's Kueue wait for the same reason but missed cert-generator, which installs Kueue in every compose stack. In #407's frontend shard the 120 s wait timed out at 00:55:02, set -e stopped setup-k8s.sh before it created the ResourceFlavor, ClusterQueue and LocalQueue, and the stack came up anyway because the entrypoint runs setup-certs.sh after a ';'. Every executor pod was then refused: failed calling webhook mpod.kb.io, no endpoints available for service kueue-webhook-service. --- cert-generator/setup-k8s.sh | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/cert-generator/setup-k8s.sh b/cert-generator/setup-k8s.sh index 17cb9f68..d931f1e1 100644 --- a/cert-generator/setup-k8s.sh +++ b/cert-generator/setup-k8s.sh @@ -98,7 +98,8 @@ curl --proto =https --tlsv1.2 -fsSL -o "$KUEUE_MANIFEST" "https://github.com/kub echo "${KUEUE_MANIFEST_SHA256} ${KUEUE_MANIFEST}" | sha256sum -c - kubectl apply --server-side -f "$KUEUE_MANIFEST" rm -f "$KUEUE_MANIFEST" -kubectl wait --for=condition=Available --timeout=120s \ +# Kueue's cert bootstrap answers /readyz with 404 until the kubelet mounts its new secret, which can take minutes +kubectl wait --for=condition=Available --timeout=300s \ deployment/kueue-controller-manager -n kueue-system # Kueue resources: ResourceFlavor + ClusterQueue + LocalQueue