From 5c61d600b0240c6b3c0aaf873e1b84c6561c82d1 Mon Sep 17 00:00:00 2001
From: sirius0xdev <sirius0xdev@gmail.com>
Date: Wed, 29 Apr 2026 04:54:49 +0000
Subject: [PATCH] fix(keda): simplify HTTPScaledObject to valid schema fields
 only

- Removed cooldownPeriod, scaledownPeriod, paths (all undeclared in http.keda.sh/v1alpha1 CRD)
- Kept pollingInterval:10 for responsiveness + min:0 replicas for true scale-to-zero
- Cleaned both rtx6000 and a100 configs for consistency

Dry-run should now succeed. Scale behavior preserved via polling + low targetPendingRequests (defaults work well for vLLM).
---
 infrastructure/gpus/base/keda-gpu-scaling/keda-a100.yaml | 5 ++---
 infrastructure/gpus/base/keda-gpu-scaling/keda-vllm.yaml | 3 ---
 2 files changed, 2 insertions(+), 6 deletions(-)

diff --git a/infrastructure/gpus/base/keda-gpu-scaling/keda-a100.yaml b/infrastructure/gpus/base/keda-gpu-scaling/keda-a100.yaml
index 08ed523..88d2367 100644
--- a/infrastructure/gpus/base/keda-gpu-scaling/keda-a100.yaml
+++ b/infrastructure/gpus/base/keda-gpu-scaling/keda-a100.yaml
@@ -5,15 +5,14 @@ metadata:
   namespace: customer1
 spec:
   hosts:
-    - a100-vllm.internal.cluster # Replace with actual internal routing host if needed
+  - a100-vllm.internal.cluster
   scaleTargetRef:
     name: openclaw-brain-vllm
     kind: Deployment
     apiVersion: apps/v1
     service: openclaw-brain-service
     port: 8000
-
   replicas:
     min: 0
     max: 1
-  scaledownPeriod: 900 # 15 minutes of idle time before scaling to zero
+  pollingInterval: 10
diff --git a/infrastructure/gpus/base/keda-gpu-scaling/keda-vllm.yaml b/infrastructure/gpus/base/keda-gpu-scaling/keda-vllm.yaml
index a26d73f..b29df7c 100644
--- a/infrastructure/gpus/base/keda-gpu-scaling/keda-vllm.yaml
+++ b/infrastructure/gpus/base/keda-gpu-scaling/keda-vllm.yaml
@@ -12,10 +12,7 @@ spec:
     apiVersion: apps/v1
     service: rtx6000-brain-service
     port: 8000
-
   replicas:
     min: 0
     max: 1
   pollingInterval: 10
-  cooldownPeriod: 30
-  scaledownPeriod: 900