CN111427605B - Docker-based distributed code automatic update deployment method and system - Google Patents

Abstract

The invention discloses a distributed code automatic updating and deploying method and system based on docker, wherein the method comprises the following steps: testing the test code; after the tested test codes are sent to the GIT version manager, the test codes are pushed to a first stable code node and a second stable code node; the first stable code node and the second stable code node are respectively mounted to a first production environment docker node and a second production environment docker node; switching the access flow to a second production environment docker node; the first stable code node loads the test code passing the test to a first production environment docker node and updates the test code; switching the access flow to a first production environment docker node; and the second stable code node mounts the test code passing the test to a second production environment docker node and then updates the test code. The invention realizes the perceptive seamless updating switching when the project codes are updated.

Description

Docker-based distributed code automatic update deployment method and system

technical field

The invention relates to the technical field of software code update and deployment, in particular to a docker-based distributed code automatic update and deployment method and system.

Background technique

With the rapid development of cloud computing, container technology is becoming more and more mature, and it is becoming more and more widely used in various fields. In normal development or daily deployment, our traditional project management is often submitted manually through git or svn The code is sent to the server, and then the operation and maintenance personnel manually synchronize the latest version of the code to the production environment. This traditional project management has the following disadvantages: 1. The project code and the production environment are on the same server. As long as the generation server is down, the generation environment will be accessed immediately No; 2. There is manual operation of manually submitting code and updating code, and it is considered that the error rate is relatively high; 3. There will be service interruption when updating code, which will affect the use of the production environment.

SUMMARY OF THE INVENTION

The technical problem to be solved by the present invention is to provide a docker-based distributed code automatic update deployment method and system, introduce docker container technology and distributed storage, and use multi-node update and switch to realize the update of project code. Unaware seamless update switching, and the code is separated from the generation environment server, making the environment more stable and secure.

In order to solve the above-mentioned technical problems, the present invention provides the following technical solutions: a docker-based distributed code automatic update deployment method, comprising the following steps:

S1. Mount the GIT version manager in the first NFS storage unit to the test environment docker node;

S2, the GIT version manager receives the test code and sends the test code to the test environment docker node for testing;

S3. The docker node of the test environment will feed back the test code that has passed the test to the GIT version manager;

S4, the GIT version manager synchronously pushes the test code that has passed the test to the first stable code node and the second stable code node of the second NFS storage unit;

S5, mount the first stable code node and the second stable code node in the second NFS storage unit to the first production environment docker node and the second production environment docker node respectively;

S6. The gateway entrance switches all the access traffic to the second production environment docker node first;

S7, the first stable code node mounts the test code that has passed the test to the first production environment docker node;

S8. After the docker node of the first production environment receives the test code that passes the test and updates it,

The gateway entry switches the access traffic to the first production environment docker node;

S9, the second stable code node mounts the test code that passed the test to the docker node of the second production environment, and the docker node of the second production environment receives the test code that passed the test and updates it;

S10 , the gateway entrance both reopens the first production environment docker node and the second production environment docker node to connect the access traffic, and the test code update and deployment that passes the test ends.

Specifically, the step S3 further includes: the test environment docker node feeds back the test code that fails the test to the GIT version manager;

The GIT version manager synchronously pushes the test code that fails the test to the first stable code node and the second stable code node of the second NFS storage unit;

Mount the first stable code node and the second stable code node in the second NFS storage unit to the first production environment docker node and the second production environment docker node, respectively, and push the test code that fails the test to the first production synchronously For the environment docker node and the second production environment docker node, the test code update deployment that does not pass the test ends.

Specifically, in the step S3, the test environment docker node feeds back the test code that has passed the test to the GIT version manager through the CICD pipeline;

In the S4, the GIT version manager synchronously pushes the test code that has passed the test to the first stable code node and the second stable code node of the second NFS storage unit through the CICD pipeline.

Specifically, the test environment docker node feeds back the test code that fails the test to the GIT version manager through the CICD pipeline;

The GIT version manager synchronously pushes the test code that fails the test to the first stable code node and the second stable code node of the second NFS storage unit through the CICD pipeline;

Mount the first stable code node and the second stable code node in the second NFS storage unit to the docker node of the first production environment and the docker node of the second production environment, respectively, and push the test code that does not pass the test to the CICD pipeline synchronously. The first production environment docker node and the second production environment docker node.

Another object of the present invention is to provide a docker-based distributed code automatic update and deployment system, including a first NFS storage unit, a second NFS storage unit, a test environment unit, an online production environment unit, and a gateway entry unit. An NFS storage unit includes a GIT version manager, the second NFS storage unit includes a first stable code node module and a second stable code node module, the test environment unit includes a test environment docker node module, and the online production environment The unit includes a first production environment docker node module and a second production environment docker node module;

The first NFS storage unit is used to mount its GIT version manager to the test environment docker node module;

The GIT version manager is used to receive the test code, and send the test code to the test environment docker node module of the test environment unit;

The test environment docker node module is used to test the test code and confirm whether the test code passes the test;

The test environment docker node module is also used to feed back the test code that has passed the test to the GIT version manager;

The GIT version manager is further configured to synchronously push the test code that has passed the test to the first stable code node module and the second stable code node module of the second NFS storage unit;

The second NFS storage unit is used to mount the first stable code node module and the second stable code node module to the first production environment docker node module and the second production environment docker node module of the online production environment unit, respectively ;

The gateway entry unit is used to switch all access traffic to the first production environment docker node module or the second production environment docker node module; the gateway entry unit is also used to re-open the first production environment docker node module and the second production environment docker node module together. The production environment docker node module connects the access traffic;

The first stable code node module is used for: when the gateway entry unit switches all access traffic to the second production environment docker node module first, the first stable code node module is used to pass the test of the test. The code is mounted to the first production environment docker node module; the first production environment docker node module is used to receive and update the test code that has passed the test;

The second stable code node module is used for: when the gateway entry unit switches all access traffic to the first production environment docker node module first, the second stable code node module is used to pass the test of the test. The code is mounted to the second production environment docker node module; the second production environment docker node module is used to receive and update the test code that has passed the test.

Specifically, the test environment docker node module is further configured to feed back the test code that fails the test to the GIT version manager, and the GIT version manager is configured to synchronously push the test code that fails the test to the second The first stable code node module and the second stable code node module of the NFS storage unit;

The second NFS storage unit is used to mount the first stable code node module and the second stable code node module to the docker node module of the first production environment and the docker node module of the second production environment, respectively, and store the tests that fail the test. The code is pushed to the docker node module of the first production environment and the docker node module of the second production environment synchronously.

After the above technical solution is adopted, the present invention has at least the following beneficial effects: the present invention completely separates the production environment, the test environment, the stable code and the test code, and mounts it into the docker container of the production environment through the NFS storage technology, and develops The personnel only need to care whether the test passes after the code is submitted. If it passes, it will be automatically released to the stable code node, and they do not need to care about the docker node of the generation environment. The operation and maintenance personnel only need to monitor whether the server is normal, and the entire update and deployment process achieves a high availability. A very stable solution;

The code node of the present invention is completely separated from the production environment node, and the security and stability are greatly improved;

The distributed node update and deployment of the present invention, when one of the nodes fails, the other node can still be used normally, so as to achieve high availability;

The present invention will reopen two nodes at the same time when the code is stable and successfully pushed to the production environment synchronously. In any case, at least one node is maintained to provide external services, and the current update control is interrupted, so as to achieve continuous service update without perception.

Description of drawings

Fig. 1 is a flow chart of the steps of the method for automatic update and deployment of distributed code based on docker of the present invention;

FIG. 2 is a frame diagram of a docker-based distributed code automatic update deployment system of the present invention.

Detailed ways

It should be noted that the embodiments in the present application and the features in the embodiments may be combined with each other without conflict, and the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

Example 1

As shown in Figure 1, the present invention discloses a docker-based distributed code automatic update and deployment method, the steps include:

S11. Mount the GIT version manager in the first NFS storage unit to the test environment docker node;

S12. The GIT version manager receives the test code and sends the test code to the test environment docker node for testing; the test environment docker node judges whether the test code passes the test;

S13. When the test code passes the test, the test environment docker node feeds back the passed test code to the GIT version manager; preferably, the test environment docker node feeds back the passed test code to the GIT version management through the CICD pipeline device;

S14. The GIT version manager synchronously pushes the test code that passed the test to the first stable code node and the second stable code node of the second NFS storage unit; the preferred GIT version manager synchronously pushes the test code that passed the test through the CICD pipeline to the first stable code node and the second stable code node of the second NFS storage unit;

S15, mount the first stable code node and the second stable code node in the second NFS storage unit to the first production environment docker node and the second production environment docker node respectively;

S16. The gateway entrance switches all the access traffic to the second production environment docker node first;

S17, the first stable code node mounts the test code that has passed the test to the first production environment docker node;

S18. After the docker node of the first production environment receives and updates the test code that has passed the test, the gateway entry switches the access traffic to the docker node of the first production environment;

S19, the second stable code node mounts the test code that has passed the test to the second production environment docker node, and the second production environment docker node receives the code and updates it;

In S20, the gateway entrance both re-open the first production environment docker node and the second production environment docker node to connect the access traffic, and the test code update and deployment that has passed the test is completed.

In addition, step S13 further includes: when the test code fails the test, the docker node of the test environment feeds back the test code that fails the test to the GIT version manager; specifically, the docker node of the test environment will not pass the test code of the test Feedback to the GIT version manager through the CICD pipeline;

The GIT version manager synchronously pushes the test code that fails the test to the first stable code node and the second stable code node of the second NFS storage unit; specifically, the GIT version manager synchronizes the test code that fails the test through the CICD pipeline Pushing to the first stable code node and the second stable code node of the second NFS storage unit;

Mount the first stable code node and the second stable code node in the second NFS storage unit to the first production environment docker node and the second production environment docker node, respectively, and push the test code that fails the test to the first production synchronously The environment docker node and the second production environment docker node, the test code that fails the test is updated and deployed; specifically, the second NFS storage unit synchronously pushes the test code that fails the test to the first production environment docker node and the first production environment through the CICD pipeline. Two production environment docker nodes.

Among the above, Docker, NFS, Git and CICD are all commonly used technical means in the development field; among them, Docker is an open source application container engine that allows developers to package their applications and dependencies into a portable container, Then publish it to any popular Linux machine or Windows machine, and it can also be virtualized. The container completely uses the sandbox mechanism and will not have any interface with each other; NFS refers to the network file system, which is called Network File System in English. It is a UNIX presentation layer protocol (presentation layer protocol) developed by SUN, which enables users to access files elsewhere on the network as if they were using their own computer; Git is an open source distributed version control system that can be effective and fast. To handle the version management of small to very large projects, Git is an open source version control software developed to manage Linux kernel development.

The CICD pipeline is a conventional technical means in the development field, which specifically refers to: CI is continuous integration. Through continuous integration, developers can frequently integrate their code into the main branch of the public code repository, and developers can at any time Repositories commit work, rather than developing each feature module independently and committing them one by one at the end of the development cycle; CD is continuous deployment, and continuous deployment extends continuous delivery so that software builds are automatically deployed when they pass all tests, in such a process There is no need to manually decide when and how to put it into the production environment; the last step of the CI and CD system will be automatically deployed when the built components and packages exit the pipeline. Such automatic deployment can be configured to quickly distribute components and function modules to customers Or fix a patch and state exactly what is currently available.

Example 2

This implementation provides a docker-based distributed code automatic update and deployment system based on Embodiment 1, as shown in Figure 2, including a first NFS storage unit, a second NFS storage unit, a test environment unit, and an online production unit. An environment unit and a gateway entry unit, the first NFS storage unit includes a GIT version manager, the second NFS storage unit includes a first stable code node module and a second stable code node module, and the test environment unit includes a test environment docker node module, the online production environment unit includes a first production environment docker node module and a second production environment docker node module;

The first NFS storage unit is used to mount its GIT version manager to the test environment docker node module;

The GIT version manager is used to receive the test code, and send the test code to the test environment docker node module of the test environment unit;

The test environment docker node module is used to test the test code and confirm whether the test code passes the test;

The test environment docker node module is also used to feed back the test code that has passed the test to the GIT version manager;

The GIT version manager is further configured to synchronously push the test code that has passed the test to the first stable code node module and the second stable code node module of the second NFS storage unit;

The second NFS storage unit is used to mount the first stable code node module and the second stable code node module to the first production environment docker node module and the second production environment docker node module of the online production environment unit, respectively ;

The gateway entry unit is used to switch all access traffic to the first production environment docker node module or the second production environment docker node module; the gateway entry unit is also used to re-open the first production environment docker node module and the second production environment docker node module together. The production environment docker node module connects the access traffic;

The first stable code node module is used for: when the gateway entry unit switches all access traffic to the second production environment docker node module first, the first stable code node module is used to pass the test of the test. The code is mounted to the first production environment docker node module; the first production environment docker node module is used to receive and update the test code that has passed the test;

The second stable code node module is used for: when the gateway entry unit switches all access traffic to the first production environment docker node module first, the second stable code node module is used to pass the test of the test. The code is mounted to the second production environment docker node module; the second production environment docker node module is used to receive and update the test code that has passed the test.

The test environment docker node module is further configured to feed back the test code that fails the test to the GIT version manager, and the GIT version manager is configured to synchronously push the test code that fails the test to the second NFS storage unit The first stable code node module and the second stable code node module;

The second NFS storage unit is used to mount the first stable code node module and the second stable code node module to the docker node module of the first production environment and the docker node module of the second production environment, respectively, and store the tests that fail the test. The code is pushed to the docker node module of the first production environment and the docker node module of the second production environment synchronously.

Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various equivalents may be made to these embodiments without departing from the principle and spirit of the invention Changes, modifications, substitutions and alterations, the scope of the invention is defined by the appended claims and their equivalents.

Claims (6)

1. A distributed code automatic updating and deploying method based on docker is characterized by comprising the following steps:

s1, mounting the GIT version manager in the first NFS storage unit to a socket node of the test environment;

s2, the GIT version manager receives the test codes and sends the test codes to a docker node in a test environment for testing;

s3, the testing environment docker node feeds back the testing code passing the test to the GIT version manager;

s4, the GIT version manager synchronously pushes the test codes passing the test to a first stable code node and a second stable code node of a second NFS storage unit;

s5, respectively mounting a first stable code node and a second stable code node in a second NFS storage unit to a first production environment docker node and a second production environment docker node;

s6, switching all access flows to a second production environment docker node by the gateway entrance;

s7, the first stable code node mounts the test code passing the test to a first production environment docker node;

s8, after the first production environment docker node receives and updates the test code passing the test, the gateway inlet switches the access flow to the first production environment docker node;

s9, the second stable code node mounts the test code passing the test to a second production environment docker node, and the second production environment docker node receives and updates the test code passing the test;

and S10, the gateway entrance restarts the connection of the access flow between the first production environment docker node and the second production environment docker node, and the test code updating deployment through the test is finished.

2. The docker-based distributed code automatic update deployment method according to claim 1, wherein the step S3 further comprises: the docker node in the test environment feeds back the test code which does not pass the test to the GIT version manager;

synchronously pushing the test codes which do not pass the test to a first stable code node and a second stable code node of a second NFS storage unit by the GIT version manager;

and respectively mounting the first stable code node and the second stable code node in the second NFS storage unit to the first production environment docker node and the second production environment docker node, and synchronously pushing the test codes which do not pass the test to the first production environment docker node and the second production environment docker node, so that the updating and the deployment of the test codes which do not pass the test are finished.

3. The docker-based distributed code automatic update deployment method according to claim 1, wherein in step S3, the docker node feeds back the test code that passes the test to the GIT version manager through a CICD pipeline;

in S4, the GIT version manager synchronously pushes the test code passing the test to the first stable code node and the second stable code node of the second NFS storage unit through the cic pipeline.

4. The docker-based distributed code automatic update deployment method of claim 2, wherein the docker node feeds back test codes that fail testing to the GIT version manager through a CICD pipeline;

the GIT version manager synchronously pushes the test codes which do not pass the test to a first stable code node and a second stable code node of a second NFS storage unit through a CICD pipeline;

and respectively mounting the first stable code node and the second stable code node in the second NFS storage unit to the first production environment docker node and the second production environment docker node, and synchronously pushing the test codes which do not pass the test to the first production environment docker node and the second production environment docker node through a CICD pipeline.

5. A distributed code automatic updating deployment system based on docker is characterized by comprising a first NFS storage unit, a second NFS storage unit, a test environment unit, an online production environment unit and a gateway entry unit, wherein the first NFS storage unit comprises a GIT version manager, the second NFS storage unit comprises a first stable code node module and a second stable code node module, the test environment unit comprises a test environment docker node module, and the online production environment unit comprises a first production environment docker node module and a second production environment docker node module;

the first NFS storage unit is used for mounting a GIT version manager to a socket node module in a test environment;

the GIT version manager is used for receiving the test codes and sending the test codes to the socket node module of the test environment unit;

the testing environment docker node module is used for testing the testing code and confirming whether the testing code passes the test;

the testing environment docker node module is also used for feeding back a testing code passing the test to the GIT version manager;

the GIT version manager is also used for synchronously pushing the test codes passing the test to a first stable code node module and a second stable code node module of the second NFS storage unit;

the second NFS storage unit is used for respectively mounting the first stable code node module and the second stable code node module to a first production environment docker node module and a second production environment docker node module of the on-line production environment unit;

the gateway entrance unit is used for switching all access flow to a first production environment docker node module or a second production environment docker node module; the gateway inlet unit is also used for starting the connection of the access flow between the first production environment docker node module and the second production environment docker node module together again;

the first stable code node module is to: when the gateway entry unit switches all the access traffic to the second production environment docker node module first, the first stable code node module is used for mounting the test code passing the test to the first production environment docker node module; the first production environment docker node module is used for receiving and updating the test codes passing the test;

the second stable code node module is to: when the gateway entry unit switches all the access traffic to the first production environment docker node module first, the second stable code node module is used for mounting the test code passing the test to the second production environment docker node module; and the second production environment docker node module is used for receiving the test codes passing the test and updating the test codes.

6. The docker-based distributed code automatic update deployment system of claim 5, wherein the test environment docker node module is further configured to feed back test codes that fail the test to the GIT version manager, and the GIT version manager is configured to synchronously push the test codes that fail the test to the first and second stable code node modules of the second NFS storage unit;

the second NFS storage unit is used for respectively mounting the first stable code node module and the second stable code node module to the first production environment docker node module and the second production environment docker node module, and synchronously pushing the test codes which do not pass the test to the first production environment docker node module and the second production environment docker node module.

Images