use bigger memory and delete network config

Signed-off-by: JaredforReal <[email protected]>

Author: JaredforReal

.github/workflows/k8s-api-functionality-test.yml

@@ -112,16 +112,16 @@ jobs:
           - patch: |-
               - op: replace
                 path: /spec/template/spec/containers/0/resources/requests/memory
-                value: "512Mi"
+                value: "2Gi"
               - op: replace
                 path: /spec/template/spec/containers/0/resources/requests/cpu
-                value: "250m"
+                value: "500m"
               - op: replace
                 path: /spec/template/spec/containers/0/resources/limits/memory
-                value: "1Gi"
+                value: "4Gi"
               - op: replace
                 path: /spec/template/spec/containers/0/resources/limits/cpu
-                value: "500m"
+                value: "1"
               - op: add
                 path: /spec/template/spec/containers/0/imagePullPolicy
                 value: "IfNotPresent"
@@ -141,7 +141,7 @@ jobs:
           kubectl wait --for=jsonpath='{.status.phase}'=Active namespace/vllm-semantic-router-system --timeout=60s
           kubectl wait --for=jsonpath='{.status.phase}'=Bound pvc/semantic-router-models -n vllm-semantic-router-system --timeout=120s
           kubectl wait --for=condition=Initialized pods -l app=semantic-router -n vllm-semantic-router-system --timeout=600s
-          kubectl wait --for=condition=Ready pods -l app=semantic-router -n vllm-semantic-router-system --timeout=300s
+          kubectl wait --for=condition=Ready pods -l app=semantic-router -n vllm-semantic-router-system --timeout=600s
 
       - name: Run comprehensive API tests
         run: |
@@ -150,9 +150,9 @@ jobs:
           POD_NAME=$(kubectl get pods -n vllm-semantic-router-system -l app=semantic-router -o jsonpath='{.items[0].metadata.name}')
           echo "Pod name: $POD_NAME"
 
-          # Wait for API to be ready
+          # Wait for API to be ready (increased timeout for model loading)
           echo "Waiting for API to be ready..."
-          timeout 120 bash -c 'until kubectl exec -n vllm-semantic-router-system $POD_NAME -- curl -s http://localhost:8080/health > /dev/null 2>&1; do echo "Waiting for API..."; sleep 10; done'
+          timeout 300 bash -c 'until kubectl exec -n vllm-semantic-router-system $POD_NAME -- curl -s http://localhost:8080/health > /dev/null 2>&1; do echo "Waiting for API..."; sleep 10; done'
 
           # Test results tracking
           TESTS_PASSED=0

.github/workflows/k8s-integration-test.yml

@@ -44,37 +44,37 @@ jobs:
   # Step 1: Validate Kubernetes manifests
   validate-manifests:
     uses: ./.github/workflows/k8s-validate-manifests.yml
-    with:
+        with:
       kustomize_version: v5.7.1
 
   # Step 2: Run kind cluster integration test
   kind-integration-test:
     uses: ./.github/workflows/k8s-kind-integration-test.yml
     needs: validate-manifests
-    with:
+        with:
       kind_version: v0.20.0
       kustomize_version: v5.7.1
 
   # Step 3: Run comprehensive API functionality tests
   test-api-functionality:
     uses: ./.github/workflows/k8s-api-functionality-test.yml
     needs: kind-integration-test
-    with:
+        with:
       kind_version: v0.20.0
       kustomize_version: v5.7.1
 
   # Step 4: Test with custom configurations
   test-with-custom-config:
     uses: ./.github/workflows/k8s-config-test.yml
     needs: validate-manifests
-    with:
+        with:
       kustomize_version: v5.7.1
 
   # Step 5: Run security scans
   security-scan:
     uses: ./.github/workflows/k8s-security-scan.yml
     needs: validate-manifests
-    with:
+        with:
       kustomize_version: v5.7.1
 
   # Step 6: Generate test summary

.github/workflows/k8s-kind-integration-test.yml

@@ -193,18 +193,18 @@ jobs:
           - name: ghcr.io/vllm-project/semantic-router/extproc
             newTag: test
 
-          # Patch for CI - reduce resources and set imagePullPolicy
+          # Patch for CI - adjust resources for model loading and set imagePullPolicy
           patches:
           - patch: |-
               - op: replace
                 path: /spec/template/spec/containers/0/resources/requests/memory
-                value: "1Gi"
+                value: "2Gi"
               - op: replace
                 path: /spec/template/spec/containers/0/resources/requests/cpu
                 value: "500m"
               - op: replace
                 path: /spec/template/spec/containers/0/resources/limits/memory
-                value: "2Gi"
+                value: "4Gi"
               - op: replace
                 path: /spec/template/spec/containers/0/resources/limits/cpu
                 value: "1"
@@ -288,9 +288,9 @@ jobs:
             exit 1
           }
 
-          # Wait for main container to be ready
+          # Wait for main container to be ready (increased timeout for model loading)
           echo "Waiting for main container to be ready..."
-          kubectl wait --for=condition=Ready pods -l app=semantic-router -n vllm-semantic-router-system --timeout=300s || {
+          kubectl wait --for=condition=Ready pods -l app=semantic-router -n vllm-semantic-router-system --timeout=600s || {
             echo "❌ Pod did not become ready in time. Showing status and logs..."
             kubectl describe pods -n vllm-semantic-router-system -l app=semantic-router
             kubectl logs -n vllm-semantic-router-system -l app=semantic-router --tail=200 || true

deploy/kubernetes/deployment.yaml

@@ -97,19 +97,6 @@ spec:
           env:
             - name: HF_HUB_CACHE
               value: /tmp/hf_cache
-            # China Mirror
-            - name: HUGGINGFACE_HUB_CACHE
-              value: /tmp/hf_cache
-            - name: HUGGINGFACE_HUB_ENABLE_HF_TRANSFER
-              value: "1"
-            - name: HUGGINGFACE_HUB_DOWNLOAD_TIMEOUT
-              value: "300"
-            - name: HUGGINGFACE_HUB_PROXY_URL
-              value: "https://hf-mirror.com"
-            - name: PIP_INDEX_URL
-              value: https://pypi.tuna.tsinghua.edu.cn/simple
-            - name: NO_PROXY
-              value: localhost,127.0.0.1,10.0.0.0/8,172.16.0.0/12,192.168.0.0/16,.svc,.svc.cluster.local
           # Reduced resource requirements for init container
           resources:
             requests: