apiVersion: v1 kind: Secret metadata: name: keda-prom-secret namespace: bj2-dcloud stringData: customAuthHeader: "Authorization" customAuthValue: "Bearer eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJuYW1lc3BhY2UiOiJjcHJvbS11ejJuZ3J6YmpyN243Iiwic2VjcmV0TmFtZSI6IjJjMTI4NGJhZTUxYjRiZjNhMmU2MjM1NTc4NDRjNmQ2IiwiZXhwIjo0OTIzODc3ODIyLCJpc3MiOiJjcHJvbSJ9.idZaTKtQR4WOyr29T_TKZ9cGz2nW9N5Mw-5SBu321p8" --- apiVersion: keda.sh/v1alpha1 kind: TriggerAuthentication metadata: name: keda-prom-creds namespace: bj2-dcloud spec: secretTargetRef: - parameter: customAuthHeader name: keda-prom-secret key: customAuthHeader - parameter: customAuthValue name: keda-prom-secret key: customAuthValue --- apiVersion: keda.sh/v1alpha1 kind: ScaledObject metadata: name: gpu-metrics-scaledobject namespace: bj2-dcloud spec: scaleTargetRef: apiVersion: apps/v1 kind: Deployment name: dstereox minReplicaCount: 1 maxReplicaCount: 2 idleReplicaCount: 0 # 当指标为0时,保持0副本 pollingInterval: 30 # 每30秒检查一次指标 cooldownPeriod: 300 # 扩容后等待5分钟再进行下一次扩容 # initialCooldownPeriod: 0 # 启动后立即检查指标 # fallback: # Optional. Section to specify fallback options # failureThreshold: 3 # Mandatory if fallback section is included # replicas: 6 # Mandatory if fallback section is included # behavior: {kind-of-behavior} # Optional. Default: "static" triggers: # --------------------------------------------------------- # 触发器 1: 基于 volcano 队列的pending任务数 # --------------------------------------------------------- - type: prometheus metadata: serverAddress: https://cprom.bj.baidubce.com/select/prometheus metricName: queuue_pending_tasks query: sum(volcano_queue_pod_group_pending_count{queue_name="dcloud"}) customHeaders: InstanceId=cprom-uz2ngrzbjr7n7 threshold: "1" activationThreshold: "1" authModes: "custom" authenticationRef: name: keda-prom-creds