检测到您已登录华为云国际站账号，为了您更好的体验，建议您访问国际站服务网站 https://www.huaweicloud.com/intl/zh-cn
不再显示此消息
检测到您已登录华为云国际站账号，为了您更好的体验，建议您访问国际站服务网站 https://www.huaweicloud.com/intl/zh-cn
不再显示此消息
Install the cloud native plug-in, select local storage, and enable custom metric collection. For details, see Creating an HPA Policy with Custom Metrics. Create external APIServices and use kubectl apply to apply the configurations to the Kubernetes cluster. a.
API Gateway checks the time format and compares the time with the time when API Gateway receives the request. If the time difference exceeds 15 minutes, API Gateway will reject the request. Obtaining an AK/SK Pair If an AK/SK pair is already available, skip this step.
Due to the limitation of API Gateway, the prediction duration of each request does not exceed 40 seconds. Procedure To access a real-time service through a VPC channel, follow these steps: Obtain the ModelArts VPC endpoint service address.
Due to the limitation of API Gateway, the prediction duration of each request does not exceed 40 seconds. Calling a WebSocket Real-Time Service WebSocket itself does not require additional authentication.
Long prediction time The following error is reported: {"error_code": "ModelArts.4503", "error_msg": "Failed to find backend service because response timed out, please confirm your service is able to process the request without timeout. "} Due to the limitation of API Gateway, the
Due to the limitation of API Gateway, the prediction duration of each request does not exceed 40 seconds. Only the services deployed in a dedicated resource pool support high-speed access through VPC peering.
Due to the limitation of API Gateway, the prediction duration of each request does not exceed 40 seconds. Only the services deployed in a dedicated resource pool support high-speed access through VPC peering.
request_id": "d0ddda0fcdd0cc23a1588fafe426****" The API URL is wrong or does not exist. 405 N/A "detail":"Method Not Allowed" The request method is incorrect. 429 APIG.0308 "error_msg": "The throttling threshold has been reached: policy ip over ratelimit,limit:5,time:1 minute" The API Gateway
ModelArts registers an inference API with API Gateway for you to access the service.
Due to the limitation of API Gateway, the duration of a single prediction cannot exceed 40s. This feature is used for commissioning. Use API calling for actual production.
Due to the limitation of API Gateway, the prediction duration of each request does not exceed 40 seconds. Prerequisites A ModelArts model in the Normal state is available. The account is not in arrears to ensure available resources for service running.
For example, in subnet 10.0.0.0/24, 10.0.0.1 is the gateway address, 10.0.0.253 is the system interface address, 10.0.0.254 is used by DHCP, and 10.0.0.255 is the broadcast address. The subnet CIDR block cannot be too large, either.
Due to the limitation of API Gateway, the duration of a single prediction in ModelArts cannot exceed 40s. The model inference code must be logically clear and concise for satisfactory inference performance.
Due to the limitation of API Gateway, the prediction duration of each request does not exceed 40 seconds. Prerequisites You have obtained a user token, local path to the inference file, URL of the real-time service, and input parameters of the real-time service.
The IP address and gateway of the RoCE NIC cannot be configured. Snt9b Snt9b23 telescope: 2.7.5.3 2.7.5.9 or later Major The npu-smi is unavailable. Check if the NPU driver is normal. NPUs cannot be used.
