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
Figure 1 Enabling the function Updating the gateway associated with a Service Log in to the CCE console and click the cluster name to access the cluster console. In the navigation pane, choose Services & Ingresses.
AI Service Deployment AI Inference Framework Add-on AI Inference Gateway Add-on LeaderWorkerSet Add-on kagent Add-on Parent Topic: Cloud Native AI
AI Inference Gateway Add-on This add-on, developed by the Kubernetes community using the Gateway API, is a solution for managing inference traffic.
Typical Application Scenarios Type Feature Nginx ingress Standard configurations Extensive gateway customization Canary release and blue-green deployment of cloud native applications LoadBalancer ingress Hosted gateway that is highly available and O&M-free Layer 7 high-performance
API gateway flow control may be triggered when APIs are called on scale. For details, see Kubernetes APIs. Parent Topic: API & kubectl FAQs
If a pod tries to access a private CIDR block, the source node will not perform NAT on the pod IP address.
High performance requirements: Cloud Native Network 2.0 uses VPC networks to construct container networks, eliminating the need for tunnel encapsulation or NAT when containers communicate.
If requests pass through the ingress gateway, the grayscale release function of the Service is not affected. Parent Topic: Storage
The server acting as a gateway or proxy receives an invalid response from a remote server. 503 ServiceUnavailable The requested service is invalid.
This includes but is not limited to virtual networks, the OS of virtual machine hosts and guests, virtual firewalls, API Gateway, advanced security services, all types of cloud services, tenant data, identity accounts, and key management.
Timeout setting for waiting for a response from a backend server: If the backend server fails to respond during the timeout duration, the load balancer will stop waiting and return HTTP 504 Gateway Timeout to the client.
Calling Kubernetes APIs Through API Gateway You can call Kubernetes-native APIs through API Gateway using the URL in the format of https://{clusterid}.Endpoint/uri.
If a pod tries to access a private CIDR block, the source node will not perform NAT on the pod IP address.
High performance requirements: Cloud Native Network 2.0 uses VPC networks to construct container networks, eliminating the need for tunnel encapsulation or NAT when containers communicate.
Accessing a public network: Assign an EIP to the node where the workload runs (when a VPC network or tunnel network is used), bind an EIP to the pod IP address (when Cloud Native Network 2.0 is used), or configure SNAT rules through the NAT gateway.
However, if there are service exceptions that cause a sharp increase in response time, abnormal requests can occupy the ingress gateway for an extended period.
However, if there are service exceptions that cause a sharp increase in response time, abnormal requests can occupy the ingress gateway for an extended period.
In containers, enabling this parameter can prevent the bandwidth of TCP connections that have been translated using NAT from being limited.
After a request is forwarded to the backend server, if the backend server does not respond within the duration specified by member_timeout, the load balancer will stop waiting and return HTTP 504 Gateway Timeout. The value ranges from 1 to 300 (in seconds).
If the cluster forwarding mode is iptables, run the iptables -t nat -L command to view the port. If the cluster forwarding mode is IPVS, run the ipvsadm -Ln command to view the port.
