Distributed deployment system and method of components

Abstract

The invention discloses a distributed deployment system and method of components. The method comprises the following steps of: step 202, generating a component deployment package by a configured component project, and storing deployment information, wherein the deployment information comprises the information of the component project, the information of at least one component in the component project, the association information of a plurality of components and the information of the component deployment package; and step 204, scanning the component deployment package and obtaining state attribute of the component in the component deployment package, if the state attribute is static, deploying the static component at a corresponding server node, and if the state attribute is dynamic, deploying the dynamic component at the corresponding server node, and storing the corresponding relation between the component and the server node. With the technical scheme of the invention, the components with complex dependence relation can be conveniently deployed and instances can be managed finely in a distributed environment, simultaneously, needs-based creation and startup can be realized, andinfluences on the entire system can be avoided.

Description

technical field [0001] The present invention relates to component deployment technology, in particular, to a distributed deployment system and method of components. Background technique [0002] Component-based programming has always been an important method to enhance software reusability, scalability, and improve software development efficiency and quality. However, different languages ​​and platforms often require different component models, and components of different implementation models cannot interact directly. The purpose of SCA is to enable users to build enterprise applications without directly facing the level of specific technical details, but to build applications through service components. This method also makes the customer's enterprise application have a good layered architecture, which can well separate the business logic and IT logic of the application, which is not only easy to build the application, but also easy to change and deploy the application. T...

AI Technical Summary

Problems solved by technology

During specific configuration, the physical deployment package (contribution), component assembly configuration file (composite), running node (Node), component assembly configuration file deployment configuration (Cloud), etc. must be manually registered, which not only increases the operation and maintenance cost of the software , it is also prone to errors, that is, the actual running state and configuration of the server are prone to inconsistencies

In addition, in terms of component instance management, there is a problem in the existing technology, that is, once the component assembly configuration file is registered to a specific running node through the domain manager, when the node starts, the corresponding component instance will be created instead of based on the inter-component Dependencies to dynamically create and start corresponding component instances, which will cause some components that are not referenced at all to be created and start services, resulting in a waste of valuable server resources

Similarly, when a component stops running, its single-line dependent component instance cannot automatically stop the service and destroy the instance, and those components that rely on this component in reverse cannot handle it accordingly (such as adopting a service stop strategy)

In addition, in a distributed environment, the configuration of server nodes is sometimes modified according to security and management needs. When dealing with such problems, the current technical framework not only requires manual restart of all components deployed on the server nodes with modified configurations, but also All other components that reference these components must be manually restarted. These components may be distributed on different server nodes. When there are complex dependencies between components, this is a very difficult task, and the management cost is high, causing administrators to Tends to reboot all servers, disrupting other otherwise functioning services

Images