diff --git a/.github/workflows/deploy.yml b/.github/workflows/deploy.yml
index ce5f209..16ce710 100644
--- a/.github/workflows/deploy.yml
+++ b/.github/workflows/deploy.yml
@@ -176,10 +176,14 @@ jobs:
             "deployment/${K8S_DEPLOYMENT}" \
             "${K8S_CONTAINER}=${IMAGE}" \
             -n "${K8S_NAMESPACE}"
+          # 2026-05-30: bumped 180s → 300s after a real api rollout took
+          # ~240s (image pull + readiness on a fresh node) and the workflow
+          # timed out even though the pod came up healthy. Same change in
+          # api + provisioner deploy.yml. False-failure forces a manual rerun.
           kubectl rollout status \
             "deployment/${K8S_DEPLOYMENT}" \
             -n "${K8S_NAMESPACE}" \
-            --timeout=180s
+            --timeout=300s
 
       - name: Verify rolled-out image tag matches built version
         # The worker has no public ingress, so we can't curl an external