How to shield problematic pods?

1. Registration center blocking

Take the pod offline at the Nacos level

2. K8S Service level control

In Kubernetes, you can use Service objects and Readiness probes to manage Pod access and health.

Service Object: Create a Service object responsible for routing traffic to one of these three Pods. The Service object can select these three Pods using the label selector.

Readiness Probe: In a Deployment or Pod configuration, use the Readiness probe to check whether the Pod is healthy. If a pod is unhealthy, Kubernetes will not route traffic to that pod.

3. K8S Scheduling

It may also be a problem with the machine. Label the machine to stop scheduling, delete the pod and let it be scheduled to other nodes

4. Use ISTO

Istio is a powerful service mesh platform that can be used to achieve fine-grained control over traffic, including stopping traffic if a pod is unhealthy.

Define DestinationRule: Using Istio's DestinationRule resource, you can define policies for your services, including traffic load balancing, connection timeout, and more. You can configure load balancing policies in DestinationRule to ensure that traffic is only routed to healthy Pods.

Guess you like

Origin blog.csdn.net/summer_fish/article/details/132624089