Automatic deployment method based on GitLab and Harbor

Abstract

The invention discloses an automatic deployment method based on GitLab and Harbor. The method comprises the following steps: S1) setting a source code warehouse and a management warehouse by using GitLab; s2) after the source code warehouse receives the submitted code, a GitLab Runner automatically triggers a continuous construction function to compile a package container mirror image; s3) performing mirror image management on the Docker container by using Harbor; one version of the container mirror image is submitted to the mirror image management service every time the container mirror image is successfully constructed, and the same mirror image is used in subsequent different environments; and S4) releasing and deploying the mirror image in the mirror image management service to a test environment or a production environment. According to the method, the problem that the same code is constructed for multiple times in a multi-branch environment and is inconsistent in the same environment can be solved, most errors in software are exposed as soon as possible during automatic construction, the communication cost among teams is reduced, and guarantee is provided for on-time delivery of products.

Description

technical field [0001] The invention relates to a software development and deployment method, in particular to an automatic deployment method based on GitLab and Harbor. Background technique [0002] In the software development process, the traditional application publishing and deployment mode is that after the R&D personnel complete the function development, they submit the code to the code warehouse, and the operation and maintenance personnel pull the code from the code warehouse to package and build it, and deploy it to the server in the test environment after completion. Testers intervene in the test after the deployment of the test server is completed, and the version that fails the test is rejected, and the version that passes the test is released to the production environment. The release process is as follows: figure 1 shown. [0003] The traditional application release deployment mode has the following disadvantages: [0004] 1. Errors are not found in time, and...

AI Technical Summary

Problems solved by technology

[0007] The technical problem to be solved by the present invention is to provide an automated deployment method based on GitLab and Harbor, which can solve the problem of multiple builds of the same code in a multi-branch environment and inconsistent performance in the same environment, and expose the bugs in the software as soon as possible during automatic builds. Most of the errors are timely fed back to the R&D personnel; the cost of communication between teams is reduced, human resources are saved, and products are delivered on time.

Images