Method and device for injecting into container

Abstract

The invention provides a method and a device for injecting a container. The method comprises the following steps of: determining a self-defined resource type Advanced Sidecar of an application Kubernetes based on deployment containerization; the method comprises the following steps of: determining a scheduling strategy through an expander based on an Advanced Sidecar; the minimum operation unit Pod of the Kubernetes which needs to be injected into the container Sidecar is labeled, and the minimum operation unit Pod of the Kubernetes needs to be injected into the container Sidecar; the method comprises the following steps of: submitting a container Sidecar description file advanced-Sidecar. Yaml to an API (Application Program Interface) server component, determining a Pod of a node to be injected into the container Sidecar through an extender, and storing the Sidecar description file and the Pod of the node into a memory Etcd; and reading the related information of the Pod of the node from the memory Etcd through a kubelet actuator on the node, and injecting the container Sidecar into the Pod of the node. According to the technical scheme, the Sidecar can be injected without restarting the Pod, the work of the Pod cannot be influenced, and the Pod can correctly process all flows in the injection process.

Description

technical field [0001] The invention relates to the field of computer information technology, in particular to a method and device for filling a container. Background technique [0002] With the development of the information society, the scale of background applications is getting larger and larger, and the calls between applications are becoming more and more complex and frequent. Thousands of services are no longer able to solve the user's access needs with a single physical machine. era. [0003] So there are solutions to solve large-scale software application deployment, such as SOA, distributed, and microservices. Although these solutions solve the horizontal and vertical expansion capabilities of software applications, and can theoretically solve the background application requirements of any concurrent and any number of businesses, they subsequently bring more complex monitoring and alarm requirements than single applications. Based on this, the Istio platform reli...

AI Technical Summary

Problems solved by technology

[0006] However, when sidecar is injected into a large number of namespace Pods, it will bring some problems: it will bring scheduling pressure to the APIServer of Kubernetes, a large number of Pods will be terminated and then restarted, which will consume a lot of time, and the additional properties of new and old Pods are different.

[0011] The additional attributes of the old and new Pods are different. For example, if the IP of the Pod is changed, the service that is whitelisted for the IP is unavailable, etc.

Images