diff --git a/k8s/apps/llm-serving/embeddings.yaml b/k8s/apps/llm-serving/embeddings.yaml index 72a2c5c..b748971 100644 --- a/k8s/apps/llm-serving/embeddings.yaml +++ b/k8s/apps/llm-serving/embeddings.yaml @@ -45,14 +45,13 @@ spec: volumeMounts: - mountPath: /mnt/models name: models - podMetadata: - annotations: - prometheus.io/scrape: "true" - prometheus.io/port: "8080" - prometheus.io/path: "/metrics" - labels: - app.kubernetes.io/name: llm-embeddings - app.kubernetes.io/part-of: llm-serving + annotations: + prometheus.io/scrape: "true" + prometheus.io/port: "8080" + prometheus.io/path: "/metrics" + labels: + app.kubernetes.io/name: llm-embeddings + app.kubernetes.io/part-of: llm-serving maxReplicas: 1 minReplicas: 1 nodeSelector: diff --git a/k8s/apps/llm-serving/reranker.yaml b/k8s/apps/llm-serving/reranker.yaml index 6d45d94..87d5652 100644 --- a/k8s/apps/llm-serving/reranker.yaml +++ b/k8s/apps/llm-serving/reranker.yaml @@ -45,14 +45,13 @@ spec: volumeMounts: - mountPath: /mnt/models name: models - podMetadata: - annotations: - prometheus.io/scrape: "true" - prometheus.io/port: "8080" - prometheus.io/path: "/metrics" - labels: - app.kubernetes.io/name: llm-reranker - app.kubernetes.io/part-of: llm-serving + annotations: + prometheus.io/scrape: "true" + prometheus.io/port: "8080" + prometheus.io/path: "/metrics" + labels: + app.kubernetes.io/name: llm-reranker + app.kubernetes.io/part-of: llm-serving maxReplicas: 1 minReplicas: 1 nodeSelector: