CN118092979B

CN118092979B - Cluster application reconstruction method and medium

Info

Publication number: CN118092979B
Application number: CN202410458250.3A
Authority: CN
Inventors: 郑伟波; 柯桓; 邬柏; 刘建华; 仪思奇
Original assignee: Inspur General Software Co Ltd
Current assignee: Inspur General Software Co Ltd
Priority date: 2024-04-17
Filing date: 2024-04-17
Publication date: 2024-07-30
Anticipated expiration: 2044-04-17
Also published as: CN118092979A

Abstract

The invention relates to the technical field of computers, and is used for solving the problems that a large amount of time and human resources are likely to be consumed in the conventional cluster application reconstruction process, and meanwhile, system shutdown and service interruption are likely to be caused, and the embodiment of the specification discloses a cluster application reconstruction method and medium, wherein the method comprises the following steps: storing and managing application data through a resource module, wherein the application data comprises application images, application deployment files and a prefabricated updating strategy, and the application deployment files comprise existing deployment files and newly added deployment files of the original cluster application; connecting a cluster environment through a deployment module, and deploying application data in the original cluster application to generate a reconstructed cluster application; the method comprises the steps that a monitoring module collects relevant application indexes in the reconstructed cluster application and sends the relevant application indexes to an analysis module; and analyzing whether the related application indexes meet the requirements or not according to the updating strategy by an analysis module, and if not, modifying the deployment parameters of the reconfiguration cluster application according to the updating strategy.

Description

Cluster application reconstruction method and medium

Technical Field

The present disclosure relates to the field of computer technologies, and in particular, to a cluster application reconstruction method and a medium.

Background

Along with the continuous improvement of the utilization rate of the container cluster deployment architecture, a large number of internet companies choose to deploy self products or application systems in a container environment, and the normal operation of the system is ensured by exerting the characteristics of high availability, robustness and the like of the container clusters. However, as the application product is iteratively upgraded and the access amount is increased, the complexity of the system is gradually increased, the cluster size is also continuously enlarged, and the cluster application needs to be reconfigured in many times.

In conventional cluster application reconfiguration processes, to update the application system and cope with the ever-increasing amount of access, it is often necessary to reconfigure the entire application system in a second environment. In the reconstructed second environment, all container clusters, services and related components need to be rebuilt and configured. First, the application code and related resources in the original environment need to be copied into the second environment. Then, the infrastructure of the second environment, including the operating system and the container runtime, needs to be reinstalled and configured. Next, the container orchestration tool needs to be reconfigured. And finally, migrating the configuration files and the data in the original environment to ensure that the application system in the second environment can normally run.

It can be seen that, in the conventional cluster application reconfiguration process, a lot of time and manpower resources may be consumed, and meanwhile, problems of system shutdown and service interruption may be caused.

Disclosure of Invention

One or more embodiments of the present disclosure provide a cluster application reconstruction method and medium, which are used to solve the technical problems set forth in the background art.

One or more embodiments of the present disclosure adopt the following technical solutions:

One or more embodiments of the present disclosure provide a cluster application reconfiguration method, including:

Storing and managing application data through a resource module, wherein the application data comprises an application mirror image, an application deployment file and a prefabricated updating strategy, and the application deployment file comprises an existing deployment file and a newly added deployment file of an original cluster application;

the cluster environment is connected through a deployment module, and the application data is deployed in the original cluster application to generate a reconstructed cluster application;

the method comprises the steps that a monitoring module collects relevant application indexes in the reconfiguration cluster application and sends the relevant application indexes to an analysis module;

and analyzing whether the related application indexes meet the requirements or not according to the updating strategy by an analysis module, and if not, modifying the deployment parameters of the reconfiguration cluster application according to the updating strategy.

It should be noted that, the following beneficial effects may be achieved by the embodiments of the present disclosure through the foregoing description:

Time and resource cost reduction: according to the cluster application reconstruction method, application data are stored in the resource module, and the deployment module is connected with the cluster environment for deployment, so that the process of reconstructing the whole application system is avoided. Therefore, compared with the traditional migration reconstruction method, a great amount of time and manpower resources are saved.

Reduction of system outages and service outages: since it is not necessary to redeploy the entire application system in the second environment, the problems of system downtime and service disruption can be avoided. By deploying application data in the original cluster application, the reconstruction can be performed under the condition of not interrupting the existing service, and the stability and reliability of the system are improved.

Optimization of monitoring and analysis: and acquiring related application indexes in the reconstructed cluster application through a monitoring module, and sending the indexes to an analysis module. The analysis module analyzes whether the indexes meet the requirements according to a prefabricated updating strategy and modifies the deployment parameters of the reconfiguration cluster application if necessary. Therefore, the running state of the system can be timely monitored and adjusted, and the performance and the expandability of the application system are optimized.

In summary, the cluster application reconstruction method reduces the risks of system shutdown and service interruption while saving time and resource cost, optimizes the monitoring and analysis process, and improves the stability and reliability of the system.

Further, in one or more embodiments of the present disclosure, the related application index includes a connection specification database in a reconfiguration cluster application running process;

Analyzing, by an analysis module, whether the related application indexes meet requirements according to the update policy, and if not, modifying deployment parameters of the reconfiguration cluster application according to the update policy, including:

simulating the reconfiguration cluster application to access the specified database by using an analysis module according to a database link detection container deployed in the updating strategy;

and if the reconfiguration cluster application cannot access the designated database, modifying the database connection address of the reconfiguration cluster application to a standby database according to the updating strategy.

Runtime connection monitoring: and monitoring the connection condition of the reconfiguration cluster application and the appointed database in the running process through an analysis module. This can help identify if the database connection of the reconfiguration cluster application is normal, if there is a connection problem or delay, etc.

And (5) timely adjusting deployment parameters: according to a preset updating strategy, the analysis module analyzes the related application indexes and judges whether the related application indexes meet the requirements. If the related index of the connection appointed database is found to be unsatisfactory, namely the appointed database cannot be accessed, the analysis module correspondingly modifies the deployment parameters of the reconfiguration cluster application according to the updating strategy.

Setting a standby database: according to the update strategy, the analysis module deploys a database link detection container to simulate access of the reconfiguration cluster application to the specified database. If the specified database cannot be accessed in the running process, the analysis module modifies the database connection address of the reconfiguration cluster application into a standby database according to the updating strategy. In this way, the reconfiguration cluster application can still operate normally and access the standby database when the database access problem occurs.

Further, in one or more embodiments of the present disclosure, the related application index includes a number of copies and memory resources in a running process of the reconfiguration cluster application;

Acquiring memory resources and the number of copies in the running process of the reconfiguration cluster application through an analysis module, judging whether the memory resources of the reconfiguration cluster application are preset upper limits or not according to the updating strategy, and judging that the number of copies of the reconfiguration cluster application are preset upper limits;

If yes, increasing the memory resources or the copy number of the reconfiguration cluster application according to the updating strategy.

Optimizing memory resources: and monitoring the use condition of the memory resources of the reconfiguration cluster application in the running process through an analysis module. According to the update strategy, the analysis module can judge whether the memory resource of the reconstruction cluster application exceeds a preset upper limit. If the memory resources exceed the upper limit, the analysis module may modify the deployment parameters of the reconfiguration cluster application accordingly, increasing the available memory resources. Therefore, the performance, stability and expandability of the system can be improved, and the application system can be ensured to normally run when facing heavy loads and high concurrency.

Copy number adjustment: and obtaining copy number information of the reconstructed cluster application through an analysis module. According to the update strategy, the analysis module can judge whether the copy number of the reconstruction cluster application exceeds a preset upper limit. If the number of copies exceeds the upper limit, the analysis module can correspondingly modify the deployment parameters of the reconfiguration cluster application and increase the number of copies. The fault tolerance and availability of the application system can be improved by increasing the number of the copies, so that when a single copy fails, other copies can take over the work, and the normal operation of the system is maintained.

In summary, by monitoring and adjusting the memory resources and the copy number of the reconstruction cluster application, the resource utilization can be optimized, and the performance and stability can be improved. According to the preset upper limit, the memory resources and the copy number are automatically adjusted, so that the application system can be ensured to operate efficiently under different load conditions, the fault-tolerant capability is achieved, and stable and reliable service is provided.

Further, in one or more embodiments of the present disclosure, the relevant application index includes a storage volume usage of the reconfiguration cluster application;

Acquiring the storage volume usage of the reconfiguration cluster application through an analysis module, and judging whether the storage volume usage of the reconfiguration cluster application exceeds a preset threshold according to the updating strategy;

If yes, expanding the capacity of the storage volume applied by the reconstruction cluster according to the updating strategy.

storage volume optimization: and monitoring the storage volume usage of the reconstruction cluster application through an analysis module. According to the update strategy, the analysis module can judge whether the storage volume quantity applied by the reconstruction cluster exceeds a preset threshold value. If the storage volume usage exceeds the threshold, the analysis module may modify deployment parameters of the reconfiguration cluster application accordingly to expand the storage volume. Thus, the application system can be ensured to have enough storage space to store data and resources, and the problem of overflow or shortage of storage is avoided.

In summary, by monitoring and adjusting the usage of the storage volumes of the cluster application, the cluster application reconstruction method can optimize the utilization and management of storage resources, and ensure that the storage space meets the requirements of an application system. By timely expanding the storage volume, the problems of data loss, service interruption and the like can be avoided, and the reliability and stability of the system are improved.

Further, in one or more embodiments of the present disclosure, the expanding the storage volume of the reconfiguration cluster application includes:

accessing the capacity-expansion storage volume resources of the storage device interface and the container cluster interface;

And expanding the capacity of the storage volume used by the reconstruction cluster application according to the capacity-expanding storage volume resource.

Storage capacity promotion: by expanding the capacity of the expanded storage volume resources accessing the storage device interface and the container cluster interface, the storage capacity of the reconfiguration cluster application may be increased. Therefore, the data storage requirements of the application system which are continuously increased can be met, the sufficient storage space is ensured, and the problems caused by insufficient storage space, such as data loss or service interruption, are avoided.

Further, in one or more embodiments of the present disclosure, the related application index includes a request mapping access of a specified version of the application of a specified type of network in the running process of the reconfiguration cluster application;

Acquiring a network request mapping access appointed version application of a current type in the running process of the reconfiguration cluster application through an analysis module, and judging whether the network request of the current type is mapped to access the appointed version application according to the updating strategy;

And if not, mapping the network request of the current type to access the corresponding version application according to the updating strategy.

The availability of cluster applications is improved: the analysis module is used for monitoring and analyzing the related application indexes, so that problems in the application running process, such as incorrect or unsatisfactory network request mapping access, can be found in time, correction and adjustment can be performed in time, and the usability of the application can be improved.

The performance of cluster applications is improved: the analysis module is used for analyzing the related application indexes, so that whether the current type of network request is mapped to access the application of the appointed version can be judged, if not, correction is carried out, and the current type of network request is mapped to access the corresponding version application, so that the performance of the application is improved.

Reliability of cluster application is improved: the analysis module is used for analyzing the related application indexes, so that whether the current network request mapping meets the requirements or not can be judged, if not, the deployment parameters of the reconfiguration cluster application are modified according to the update strategy, the correctness of the network request mapping is ensured, and the reliability of the application is improved.

The manual intervention and management cost is reduced: and the analysis module is used for automatically monitoring and analyzing related application indexes, and automatically correcting and adjusting according to the updating strategy, so that the workload and the cost of manual intervention and management are reduced.

In a word, through analyzing and correcting related application indexes, the availability, performance and reliability of cluster application can be improved, and meanwhile, the manual intervention and management cost is reduced.

Further, in one or more embodiments of the present disclosure, determining whether the current type of network request maps to access a specified version of application according to the update policy includes:

And judging whether the current type of network request is mapped to access the appointed version application according to a preset mapping access strategy in the updating strategy, wherein the mapping access strategy is used for mapping access of the corresponding version application for each type of network request.

Stability of cluster application is improved: by updating the preset mapping access strategy in the strategy, whether the current type of network request is mapped to access the application of the appointed version can be judged. Therefore, different types of network requests can be ensured to be correctly mapped to the corresponding version application, confusion and wrong mapping are avoided, and the stability of the application is improved.

Simplifying network request management: through presetting the mapping access strategy, mapping management can be carried out on each type of network request and corresponding version application. Therefore, the management operation of the network request can be simplified, the workload of manual intervention is reduced, and the management efficiency and accuracy are improved.

Flexibility of application deployment is improved: by presetting the mapping access policy, whether the current type of network request is mapped to access the application of the appointed version can be judged according to the updating policy. Therefore, the mapping strategy can be flexibly adjusted according to specific service requirements and actual conditions, the requirements of different types of network requests on the application of the appointed version are met, and the flexibility of application deployment is improved.

In a word, whether the current type of network request is mapped to access the appointed version application is judged according to the updating strategy, and management is carried out according to the preset mapping access strategy, so that stability of the cluster application can be improved, management of the network request is simplified, and flexibility of application deployment is improved.

Further, in one or more embodiments of the present disclosure, the related application index includes a network mapping policy deployed in a reconfiguration cluster application;

Acquiring a network mapping strategy deployed in the reconfiguration cluster application through an analysis module, and judging whether the network mapping strategy deployed in the reconfiguration cluster application is normal or not according to the updating strategy;

if not, rolling back the network mapping strategy to the previous version according to the updating strategy.

Stability of cluster application is improved: the network mapping strategy deployed in the reconstruction cluster application is monitored and analyzed through the analysis module, so that whether the mapping strategy meets the requirement or not can be found in time. And if the configuration parameters do not meet the requirements, modifying the deployment parameters of the reconstruction cluster application according to the update strategy, including rollback of the network mapping strategy. Therefore, the normal operation of the network mapping strategy can be ensured, and the application unavailability or abnormality caused by the wrong mapping strategy is avoided.

Reliability of cluster application is improved: and acquiring a network mapping strategy deployed in the reconfiguration cluster application through an analysis module, and judging whether the mapping strategy is normal or not according to the updating strategy. If not, the mapping strategy is rolled back to the previous version according to the updating strategy. Therefore, the wrong mapping strategy can be corrected in time, the network request can be ensured to correctly access the corresponding application, and the reliability of the application is improved.

Simplifying maintenance and management work: and automatically monitoring and analyzing related application indexes through an analysis module, and automatically modifying deployment parameters according to the update strategy, including rollback of the network mapping strategy. Therefore, the workload of manual intervention and management can be reduced, maintenance and management work is simplified, and the working efficiency is improved.

Reducing risk and failure impact: by analyzing and correcting the network mapping strategy which does not meet the requirements in time, the risk and the fault influence caused by the wrong mapping strategy can be reduced. The mapping strategy rolled back to the previous version in time can quickly restore normal application access, and potential problem expansion and influence on other parts are avoided.

In a word, the analysis module analyzes the related application indexes according to the update strategy, and modifies the deployment parameters of the reconstructed cluster application according to the result, including rollback of the network mapping strategy, so that the stability and reliability of the cluster application can be improved, maintenance and management work can be simplified, and risks and fault influence can be reduced.

Further, in one or more embodiments of the present disclosure, the related application index includes a mirrored download address of the reconstructed cluster application;

Acquiring a designated mirror image downloading address in the reconfiguration cluster application through an analysis module, and judging whether the designated mirror image downloading address in the reconfiguration cluster application is normal or not according to the updating strategy;

And if not, changing the appointed mirror image downloading address into a standby mirror image downloading address according to the updating strategy.

the availability of cluster applications is improved: and monitoring and analyzing related application indexes by an analysis module, wherein the monitoring and analyzing comprise reconstructing the mirror image downloading address of the cluster application. And if the image download address does not meet the requirement, modifying the deployment parameters of the reconfiguration cluster application according to the update strategy. By changing the designated image download address to the standby image download address, it is possible to ensure that the application can smoothly download the required image and to improve the usability of the application.

The deployment efficiency of cluster applications is improved: and acquiring a designated mirror image downloading address in the reconstructed cluster application through an analysis module, and judging whether the designated mirror image downloading address is normal or not according to an updating strategy. If not, the image download address is changed to a standby image download address. Therefore, deployment delay and errors caused by unavailable or unstable assigned mirror image download addresses can be avoided, and the deployment efficiency of the application is improved.

Enhancing application security: an alternative choice of mirroring may be provided by the update policy changing the specified mirror download address to a standby mirror download address. This may enhance the security of the application to prevent malicious or problematic mirrored download addresses from potentially risking the security of the application.

Simplifying maintenance and management work: and automatically monitoring and analyzing related application indexes through an analysis module, and automatically modifying deployment parameters according to an update strategy, wherein the deployment parameters comprise the modification of the mirror image download address. Therefore, the workload of manual intervention and management can be reduced, maintenance and management work is simplified, and the working efficiency is improved.

In a word, the analysis module analyzes the related application indexes according to the update strategy, and modifies the deployment parameters of the reconfiguration cluster application according to the result, including the modification of the mirror image download address, so that the availability, deployment efficiency and safety of the cluster application can be improved, and meanwhile, the maintenance and management work is simplified.

One or more embodiments of the present description provide a non-volatile computer storage medium storing computer-executable instructions that, when executed by a computer, enable:

The above-mentioned at least one technical scheme that this description embodiment adopted can reach following beneficial effect:

Drawings

In order to more clearly illustrate the embodiments of the present description or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described below, it being obvious that the drawings in the following description are only some of the embodiments described in the present description, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art. In the drawings:

Fig. 1 is a flow diagram of a cluster application reconstruction method according to one or more embodiments of the present disclosure.

Detailed Description

The embodiment of the specification provides a cluster application reconstruction method and a medium.

In order to make the technical solutions in the present specification better understood by those skilled in the art, the technical solutions in the embodiments of the present specification will be clearly and completely described below with reference to the drawings in the embodiments of the present specification, and it is obvious that the described embodiments are only some embodiments of the present specification, not all embodiments. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments herein without making any inventive effort, shall fall within the scope of the present disclosure.

Fig. 1 is a schematic flow diagram of a cluster application reconfiguration method according to one or more embodiments of the present disclosure, where the flow may be executed by a cluster application reconfiguration system. Some input parameters or intermediate results in the flow allow for manual intervention adjustments to help improve accuracy.

The method flow steps of the embodiment of the present specification are as follows:

S102, storing and managing application data through a resource module, wherein the application data comprises application images, application deployment files and a prefabricated updating strategy, and the application deployment files comprise existing deployment files and newly-added deployment files of the original cluster application.

In the embodiment of the present disclosure, regarding the above storage and management of application data through the resource module, the following specific implementation steps may be performed:

Creating a resource module: a resource module needs to be created for storing and managing application material. This may be a cloud storage service such as Amazon S3 or Google Cloud Storage, or a folder on a local server.

Collecting application data: and collecting required application data, including application images, application deployment files and pre-made update strategies. Application mirroring may be used to create containers or virtual machine instances. The application deployment files comprise existing deployment files of the original cluster application and newly added deployment files, and can be used for defining the configuration and the dependency relationship of the application.

Organizing application data: the collected application data is organized and stored in the resource module according to a certain directory structure or naming rule. The application can be classified according to application name, version number or other identifiers, so that subsequent management and searching are facilitated.

Managing application data: the application data is maintained and updated periodically, including updating images, updating deployment files, updating policies, and the like. This may be accomplished by replacing an existing application profile file or adding a new file.

S104, connecting the cluster environment through a deployment module, deploying the application data in the original cluster application, and generating a reconstructed cluster application.

In the embodiment of the present specification, the following specific implementation steps may be used for the above description:

Creating a deployment module: there is a need to create a deployment module in a clustered environment that can be used to connect the clustered environment and perform deployment operations for application material. Existing deployment tools, such as Kubernetes or Docker Swarm, may be used, or a deployment module may be developed on its own.

Connection cluster environment: the deployment module is used for connecting to the original cluster environment, so that communication and interaction with the existing cluster management tool or platform can be ensured. This may involve configuration-related authentication information, network settings, etc.

Preparing application data: the application data collected before is obtained, including application images, application deployment files and pre-made update policies. Ensuring that these materials are already stored in the resource module in the manner described in the previous step and are accessible through the deployment module.

Deploying application data: and deploying the application data into the original cluster application by using a deployment module. This may include creating new container instances or virtual machine instances and applying corresponding application images and deployment files to those instances.

Generating a reconfiguration cluster application: after deployment is completed, the original cluster application will be reconfigured to contain the new version of the application material. These materials define new application configurations and dependencies and can be updated according to preset update policies.

Testing and verifying: and testing and verifying the reconstructed cluster application, ensuring that the application can normally run and meeting the expected requirements and performance requirements. This may include using automated test tools, monitoring tools, or manual test methods.

Monitoring and maintaining: the running state of the reconstructed cluster application is monitored regularly, and maintenance and updating are carried out timely. This may include monitoring performance metrics, logs, and error reports of the application, and expanding, optimizing, or repairing as needed.

S106, collecting related application indexes in the reconfiguration cluster application through a monitoring module, and sending the related application indexes to an analysis module.

creating a monitoring module: a monitoring module needs to be deployed in the reconstructed cluster application for collecting relevant application indexes. Existing monitoring tools, such as Prometheus, grafana, may be used.

And (3) configuring a monitoring module: and configuring the monitoring module to ensure that the monitoring module can be connected to the reconstructed cluster application and collect related application indexes. This may involve configuring the data source of the monitoring module, the index collection frequency, the monitored object, etc.

And (3) collecting application indexes: and starting the monitoring module and starting to collect related application indexes. These metrics may include performance metrics of the application (e.g., CPU utilization, memory usage), health of the container or virtual machine, response time of the service, etc.

And a configuration analysis module: the monitoring module is configured to send the collected application index to the analysis module for further processing and analysis. The metrics may be sent to the API endpoint of the analysis module or data may be transferred by way of a message queue or the like.

Analyzing application indexes: an analysis module is deployed for receiving and processing the application index from the monitoring module. This module may be responsible for storing the metrics in a database, performing real-time analysis, generating reports, or performing predictive analysis, etc.

And (3) implementing real-time monitoring: and connecting the monitoring module with the analysis module to realize the aim of real-time monitoring. The monitoring module sends the acquired application index to the analysis module, and the analysis module processes and analyzes the acquired application index in real time after receiving the index.

Data visualization and report generation: the monitoring data may be visualized using a visualization tool (e.g., grafana) to allow a user to easily view and understand the behavior and performance of the application. Reports may be generated and provided to an associated team or management layer for decision making and optimization.

S108, analyzing whether the related application indexes meet the requirements or not according to the updating strategy through an analysis module, and if not, modifying the deployment parameters of the reconfiguration cluster application according to the updating strategy.

And a configuration analysis module: the analysis module is ensured to be capable of receiving and processing the application index from the monitoring module and has a corresponding analysis function. This may include setting analysis rules, defining thresholds and rule sets, configuring alarm mechanisms, and so forth.

Defining an update strategy: and according to the updating strategy, defining the application index to be evaluated and the threshold value of the requirement. These indicators may be performance indicators, capacity indicators, service availability indicators, or other key indicators, defined according to actual needs.

Analyzing application indexes: and analyzing the acquired application index by using an analysis module, and comparing the difference between the index and the target threshold. And judging whether the application index meets the requirements or not according to the updating strategy, and identifying the application index which does not meet the requirements.

Modifying deployment parameters: and if the analysis result finds that certain application indexes are not satisfactory, determining deployment parameters to be modified according to the updating strategy. This may include adjusting container resource limitations, optimizing network configuration, modifying service configuration, etc., to improve performance, availability, or extensibility of the application.

And (3) application parameter modification: and applying the modified deployment parameters to the reconstructed cluster application. This may be accomplished through the use of orchestration tools or call APIs. Ensuring that the modified parameters are validated when the application is started or restarted.

Redeployment application: and redeploying the application to apply the new deployment parameters according to the functions of the deployment module. This may include performing a rolling update, forced restart, or lateral expansion, etc., to ensure that the configuration update of the application is in effect.

Monitoring and analysis again: after restarting the application, monitoring and analyzing are carried out again, and whether the application index meets the requirement of the updating strategy is confirmed. Repeating the analysis of the application index until the application index meets the requirement.

Further, in one or more embodiments of the present disclosure, the related application index includes a connection specification database in a reconfiguration cluster application running process; s108, analyzing whether the related application indexes meet requirements or not according to the updating strategy through an analysis module, and if not, simulating the reconfiguration cluster application to access the appointed database through a database link detection container deployed in the updating strategy through the analysis module in the process of modifying the deployment parameters of the reconfiguration cluster application according to the updating strategy; if the reconfiguration cluster application cannot access the appointed database, modifying a database connection address of the reconfiguration cluster application to a standby database according to the updating strategy; if the reconfiguration cluster application can access the designated database, the database connection address of the reconfiguration cluster application does not need to be modified to a standby database according to the update strategy.

It should be noted that, in the process of reconstructing the cluster application, the database needs to be normally connected, and if the access fails, the access is switched to the standby database. The embodiment of the specification can supplement and deploy a database link detection container simulation application access database when deploying an application container group (the application container group refers to a product of application deployment in a container cluster, namely an actually running application program), so that whether an application environment can be connected with the database normally or not can be accurately detected. Upon detecting that the container finds an anomaly, for example, that ten consecutive attempts to link the database failed, a message is sent to the designation tool, which considers the application to be unable to connect to the database, and the database address in the application deployment file is modified to a slave or standby environment address and the application is redeployed.

It should be noted that the assignment tool may be an operation and maintenance automation tool or an orchestration tool, such as Kubernetes, docker Compose, or Ansible. These tools may provide functionality for automating the deployment and management of clustered environments and applications. The Kubernetes is a popular container arranging tool, has functions of automatic expansion, self-healing, fault transfer and the like, and can monitor application states and automatically adjust. If Kubernetes is used, the redeployment and address modification of the application may be triggered by its built-in monitoring and automation mechanism.

The assignment tool may also be a custom monitoring and control tool for receiving messages sent by the probe container and detecting the connection status of the application to the database according to predefined rules. Upon detection of an anomaly, the tool may trigger a corresponding operation, such as modifying an application deployment file, redeploying an application, and so forth. The custom tool may be a script, a stand-alone application, or a consumer of a message queue.

Further, in one or more embodiments of the present disclosure, the related application index includes a number of copies and memory resources in a running process of the reconfiguration cluster application; the step S108 is to analyze whether the related application indexes meet the requirements according to the updating strategy through an analysis module, if not, the analysis module is used for acquiring the memory resources and the number of copies in the running process of the reconfiguration cluster application, and judging whether the memory resources of the reconfiguration cluster application are preset upper limits according to the updating strategy and the number of copies of the reconfiguration cluster application is preset upper limits in the process of modifying the deployment parameters of the reconfiguration cluster application according to the updating strategy; if yes, increasing the memory resources or the copy number of the reconfiguration cluster application according to the updating strategy; if not, the memory resources or the copy number of the reconfiguration cluster application do not need to be increased according to the updating strategy.

It should be noted that if the memory resource consumed in the running process of the reconfiguration cluster application reaches the upper limit set by the original rule and the number of copies reaches the expansion upper limit set by the original rule, the running pressure of the reconfiguration cluster application is still large, and then the upper limit parameter of the available resource of the application deployment needs to be modified. The resources consumed by the reconstructed cluster application can be obtained in real time by accessing the cluster-related APIs (the related rights have been obtained by the resource module), once the resources and the number of copies are found to reach the upper limit and the application container responds slowly. Then the upper memory limit or the upper copy number limit may begin to be increased gradually, for example, the memory is increased by 2G each time or one copy upper limit is increased, then the reconfiguration application deployment application in the cluster checks whether the application is running normally, and if not, the increasing steps are repeated until normal.

Further, in one or more embodiments of the present disclosure, the relevant application index includes a storage volume usage of the reconfiguration cluster application; the step S108 is to analyze whether the related application indexes meet the requirements according to the update strategy through an analysis module, if not, the analysis module can acquire the storage volume usage of the reconfiguration cluster application in the process of modifying the deployment parameters of the reconfiguration cluster application according to the update strategy, and judge whether the storage volume usage of the reconfiguration cluster application exceeds a preset threshold according to the update strategy; if yes, expanding the capacity of the storage volume applied by the reconstruction cluster according to the updating strategy; if not, the storage volume of the reconfiguration cluster application is expanded without a new strategy.

Further, in one or more embodiments of the present disclosure, when expanding a storage volume of the reconfiguration cluster application, an expanded storage volume resource of a storage device interface and a container cluster interface may be accessed; and expanding the capacity of the storage volume used by the reconstruction cluster application according to the capacity-expanding storage volume resource.

The above description relates to an automatic volume expansion application storage volume, and if an application in a container cluster needs to store persistent data, a storage device mounted in the cluster needs to apply for a storage volume with a specified capacity in advance. However, with the continuous use of the storage volume capacity of the application, there is a risk of exhaustion, and manual intervention is typically required to manually expand the storage volume. Based on this, the monitoring policy for the storage resource may be configured in advance, including: the method comprises the steps of applying an initial application storage volume capacity upper limit, a residual capacity trigger threshold (such as a warning is triggered when 90% of the storage volume is used), a resource expansion range (such as expanding 10G hard disk storage when the threshold is triggered each time), and a storage device docking interface or mode (used for inquiring whether the storage device has residual resources which can be expanded and applying for new storage, and sending a notification to an operation and maintenance if the resources are insufficient). And once the monitoring module finds that the application storage volume usage exceeds the threshold, the analysis module accesses the storage volume resources of the storage device interface and the container cluster interface according to the prefabricated strategy. If the storage device itself does not meet the capacity expansion condition, the operation and maintenance personnel can be notified of the intervention.

Further, in one or more embodiments of the present disclosure, the related application index includes a request mapping access of a specified version of the application of a specified type of network in the running process of the reconfiguration cluster application; s108, analyzing whether the related application indexes meet requirements or not according to the updating strategy through an analysis module, if not, acquiring a current type of network request mapping access appointed version application in the running process of the reconfiguration cluster application through the analysis module in the process of modifying the deployment parameters of the reconfiguration cluster application according to the updating strategy, and judging whether the current type of network request mapping access the appointed version application according to the updating strategy; if not, mapping the current type of network request to access the corresponding version application according to the updating strategy; if yes, mapping access to the corresponding version application of the current type of network request is not required according to the updating strategy.

It should be noted that, the foregoing is related to automatic/emergency switching of gray level release, where gray level release refers to that the same set of applications deploys different versions in a container cluster, and accesses the applications of different versions through controlling network traffic and mapping policies. The module tool can automatically switch the specified network request to different version applications by means of monitoring and preset strategies. When a certain interface fault occurs in the running process of the website, the monitoring module can modify the network mapping components (such as ingress, istio network configuration tools) in the cluster after finding, and the network request is switched to another set of version application in an emergency mode under the condition that the application service target is not influenced, or the interface request is intercepted briefly and the service is not processed in maintenance is prompted temporarily, and the operation and maintenance personnel are informed to detect the fault in time, so that damage and pollution to service data are prevented. Or when certain holidays with specific dates are reached, the tool forwards the webpage access to the designated resource path according to the preset time strategy, and the operation and maintenance personnel do not need to update the operation manually.

Further, in one or more embodiments of the present disclosure, the related application index includes a network mapping policy deployed in a reconfiguration cluster application; the step S108 is to analyze whether the related application indexes meet the requirements according to the updating strategy through an analysis module, if not, whether the network mapping strategy deployed in the reconfiguration cluster application is normal or not can be judged according to the updating strategy by acquiring the network mapping strategy deployed in the reconfiguration cluster application through the analysis module in the process of modifying the deployment parameters of the reconfiguration cluster application according to the updating strategy; if not, rolling back the network mapping strategy to the previous version according to the updating strategy; if yes, the network mapping strategy does not need to be rolled back to the previous version according to the updating strategy.

It should be noted that, the above process and the link tracking monitoring are related to network policy reset rollback, after the deployment of the reconfiguration cluster application is successful, the network mapping policy needs to be deployed in the container cluster, including configuration such as domain name and path forwarding, and the monitoring module may also configure the network domain name dial testing policy for the application access scene to check whether the application access is normal. However, the gradual accumulation or change of network policies in the container clusters has the probability of causing failure of the network policies that have been operating properly. For example, in preparation for deploying a new application network policy or modifying an existing application network policy, the tool will immediately perform a network dial-up check to see if it is clear after the update is completed. And once the network policy is found to have faults, immediately rolling back the network policy to the last normal version, reconstructing deployment in the cluster to take effect, and recording new policy error reporting information to inform operation and maintenance personnel of processing.

The network domain name dialing test is a common network detection means, and refers to that whether an address responds normally or not is confirmed by accessing a designated address according to a return value, for example, if an a website domain name is accessed, if an a website home page can be displayed normally, the a website is proved to be accessed normally, and similarly, a plurality of designated ports or sub-paths can be dialed and tested to meet the requirement of detecting a designated function.

Further, in one or more embodiments of the present disclosure, the related application index includes a mirrored download address of the reconstructed cluster application; s108, analyzing whether the related application indexes meet requirements or not according to the updating strategy through an analysis module, if not, acquiring a designated mirror image download address in the reconfiguration cluster application through the analysis module in the process of modifying the deployment parameters of the reconfiguration cluster application according to the updating strategy, and judging whether the designated mirror image download address in the reconfiguration cluster application is normal or not according to the updating strategy; if not, changing the appointed mirror image downloading address into a standby mirror image downloading address according to the updating strategy; if yes, the appointed mirror image downloading address is not required to be changed into a standby mirror image downloading address according to the updating strategy.

It should be noted that, the above content is related to container cluster mirror image pulling, and in the deployment process of the reconfiguration cluster application, the mirror image file needs to be downloaded from the mirror image warehouse to the cluster node, so that the mirror image warehouse address and the warehouse login account password need to be configured in the application deployment file. But if the warehouse access is not enabled, the card cannot be deployed in the pull failure state. The tool can support prefabricating a plurality of standby warehouse addresses and account passwords in advance, and if the failure of accessing the currently configured mirror warehouse address is detected, the application mirror image downloading address can be modified and redeployed according to the standby warehouse address until the mirror image is downloaded normally.

When various containerized frameworks are used for deploying the application, only one mirror image download address is configured for supporting the application to maintain consistency, so that once a mirror image warehouse fails, the application deployment fails, an operation and maintenance person manually adjusts parameters or copies the mirror image file offline, and a plurality of download sources can be supported and configured through tools. But this also changes the way tools control the individual sources must be identical or consistency of application containerization is compromised.

It should be noted that, the update policy in the embodiment of the present disclosure may also be related to update of a self-visa, and the self-visa may be created for internal communication encryption during the running process of the reconfiguration cluster application, but the default requirement of the certificate for configuring the validity time based on the security specification is that the certificate is expired annually, and after the expiration, the certificate information needs to be updated, otherwise, the cluster cannot be used any more. The certificate updating mode comprises restarting the cluster frame of each node, executing specified operation instructions and the like, and particularly needs to refer to the cluster version. The tool can monitor the cluster interface to obtain the expiration time of the certificate, pre-prepare a certificate reconstruction instruction set in advance according to the cluster environment, reconstruct the cluster certificate immediately before expiration and import the cluster environment, and the cluster certificate does not need to be updated manually, namely, the related application indexes comprise the expiration time of the certificate of the reconstructed cluster application.

It should be noted that, the update policy in the embodiments of the present disclosure may also relate to application forced migration, where in a container cluster, an application container image and a console log generated in real time during operation may be stored in a server node, and there is a risk of gradually occupying server disk resources with increasing time. The tool can modify the server and the application deployment affinity configuration by configuring a machine disk monitoring threshold, when the disk overflow risk exists, the tool immediately invokes the container cluster interface, immediately creates copies of related applications at other server nodes, and closes an application program of the risk server to clean disk garbage such as a log.

Affinity as referred to herein may refer to which cluster nodes a restructured cluster application may specify at installation time to take precedence or must run on to meet the effects of migrating restructures on a specified machine. The affinity policy is also to modify the configuration in the reconfiguration cluster application deployment data and the machine configuration file, and the configuration needs to be restarted in place to be effective after modification.

IT should be noted that, along with the promotion of the cloud platform and the continuous improvement of the utilization rate of the container cluster deployment architecture, a large number of IT companies choose to deploy their products or application systems in the container environment, so as to ensure the normal operation of the system by exerting the characteristics of high availability, robustness and the like of the container clusters. However, as the application product is continuously and iteratively upgraded and the access amount is increased, the complexity of the system is gradually increased, and the cluster scale is also continuously enlarged, which also brings great operation difficulty to operation and maintenance personnel.

Situational awareness may assist an application system in collecting some information and parameters in the environment, such as output logs of applications, clustered network parameter information, and so forth. However, after collection, operation and maintenance personnel are often required to intervene in checking the log to analyze the specific problem, so if a set of analysis standard adapting to the current application or a method for assisting the application to automatically analyze and collect information and operate, the stability and the standardization of the whole deployment and reconstruction of the application can be greatly optimized.

The embodiment of the specification discloses an application automatic reconstruction model design method and tool based on situation awareness, wherein the method automatically optimizes and adjusts a parameter formula or strategy prefabricated in the tool by collecting environment monitoring data conditions, and edits and adjusts application deployment file contents by using a new strategy, wherein the application deployment instruction, variable parameters, resource indexes, service exposure discovery modes and the like are included. After the deployment file is updated, the tool sends a deployment instruction to the current cluster or other environments to redeploy the reconstructed application, and changes the current network load path to complete the dynamic reconstruction and automatic release of the application.

The tool mainly solves the following problems: the traditional migration reconstruction essentially comprises the steps of redeploying a set of new environment in a second environment, and modifying the deployment parameters of deployed applications in the current environment, so as to realize in-situ reconstruction updating of the applications; the reconfiguration release of the application is automatically carried out by functions and tools, and excessive personnel maintenance is not required to be additionally input; all reconstruction modification operations are carried out according to a given formula or strategy, so that errors caused by manual modification are avoided; for newly built micro-service applications, the tool can create new applications and configure service registration discovery according to policies, and complete automatic release of the applications without manual registration by operation and maintenance personnel.

The main flow steps are as follows:

A. The tool is divided into a monitoring module, an analysis module, a deployment module and a resource module.

A) The resource module comprises application management and cluster management, wherein the application management is responsible for maintaining application deployment files, and the cluster management is responsible for maintaining cluster environments.

B) The deployment module is responsible for installing the applications in the resource modules into the cluster.

C) The monitoring module is responsible for monitoring application data information in the acquisition cluster.

D) The analysis module is responsible for compiling an index monitoring strategy, monitoring the data change of the monitoring module, and modifying the parameter reconstruction application after triggering the strategy.

B. The operation and maintenance personnel use tools to deploy applications in the clusters, and configure monitoring indexes in an analysis module, and configure acquisition strategies in a monitoring module.

C. After the application is operated, the monitoring module collects application data, the analysis module checks the data according to the preset strategy, if the related strategy is triggered, the application deployment file data parameters are modified according to the preset rule, a new application system is deployed in the cluster, and the history application access address is switched to the new application system.

For step A:

the resource module is mainly responsible for storing and managing application data, including application images, application deployment files, prefabricated update policies and the like. The updating strategy means that the application developer through test or, besides the application initial data, the application developer also comprises the reconstructed application deployment data, and relevant copy data is stored after each time the strategy is triggered to execute data updating.

The deployment module is responsible for connecting to access the cluster environment and deploying applications in the designated cluster using the application profiles in the resource module, including applying the current cluster environment or other alternate cluster environments.

The monitoring module is responsible for monitoring application related indexes in the acquisition cluster and providing data to the analysis module for judgment. The collection mode can be accessed to a third party collection tool or a custom collection mode.

The analysis mode is a tool core and is mainly used for analyzing the data collected by the monitoring module according to the application predefined strategy, checking whether the triggering prefabrication strategy exists or not and modifying the application deployment parameters according to the prefabrication strategy to update the application locally.

Aiming at the step B, the prefabricated strategy is firstly from the test arrangement of the product personnel on the application itself, and then the related strategy is optimized according to the actual deployment scene feedback condition of each application. Furthermore, the tool acquisition module and the analysis module can integrate big data or an AI analysis engine to calculate or even predict the actual situation of the deployment environment, so as to optimize the reconstruction strategy.

Aiming at the step C, when the tool reconstructs the application, the main operation is to modify the deployment file of the application according to the strategy, and then load the modified file information into the appointed cluster to finish the application reconstruction update.

The method is mainly aimed at presetting related processing strategies in advance for some foreseeable practices and problems. Once the related situation occurs, the tool can automatically perform application adjustment reconstruction according to the prefabrication strategy without waiting for the intervention of operation and maintenance personnel to repair the operation and maintenance personnel. Therefore, the problem repairing efficiency of the field environment can be greatly improved, and meanwhile, the probability of manual operation errors can be reduced through standardized tool or instruction operation, so that the situations of secondary problems and the like are avoided.

The above objects, features and advantages of the present invention will become more apparent and understood from the following detailed description of the preferred embodiments of the present invention.

Introduction to example environment: in the k8s cluster operation and maintenance scene of a certain large company, a website system is deployed in a container cluster, a database is deployed outside the cluster and is configured in a master-slave mode, and the link address is required to be switched in time when the access of the master database is not enabled.

The application mode is as follows:

firstly, importing network station deployment data in a tool, pre-configuring a database access connection address and a monitoring strategy, and setting a database link address in an environment variable of a website server as a main database address;

secondly, configuring the survival monitoring of the address of the main database and the monitoring strategy of the log of the website access database, setting that once the monitoring strategy is abnormal and the main database cannot be normally accessed, modifying the address of the website access database as a standby database, and updating and deploying website application.

Thirdly, deploying website application in the cluster and setting the website application as 3 server copies, so that the other two servers can still respond normally when one of the application servers is updated;

fourthly, once the monitoring module finds that the access of the main database address is not feasible or database access errors occur in the website log, the website deployment file is immediately modified according to the strategy, the database address is modified into a standby database, and an instruction is sent to the container cluster to require updating application;

And fifthly, loading the new deployment file to sequentially close the 3 server copies according to the rolling update mode, and creating new application according to new parameters, namely closing the server copy 1, creating a new server copy 1, closing the server copy 2, creating a new server copy 2, closing the server copy 3 and creating a new server copy 3. Thus, the updating of the database access address is completed under the condition of ensuring that the website application is totally alive.

Another: in addition to the rolling update approach, an adjustment policy that modifies service discovery may also be used: after the address of the database is updated, a new version of website application is directly deployed in the cluster, then the tool changes the service discovery address of the application bound with the previous version into the new version of application according to the service discovery modification strategy, and the gateway routing mapping is adjusted to forward the subsequent access to the new version of application, so that the website application is updated.

In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. In particular, for non-volatile computer storage media embodiments, the description is relatively simple, as it is substantially similar to method embodiments, with reference to the section of the method embodiments being relevant.

The foregoing describes specific embodiments of the present disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims can be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

The foregoing is merely one or more embodiments of the present description and is not intended to limit the present description. Various modifications and alterations to one or more embodiments of this description will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, or the like, which is within the spirit and principles of one or more embodiments of the present description, is intended to be included within the scope of the claims of the present description.

Claims

1. A method for cluster application reconstruction, the method comprising:

the method comprises the steps that a monitoring module collects relevant application indexes in the reconfiguration cluster application and sends the relevant application indexes to an analysis module, wherein the relevant application indexes comprise performance indexes, capacity indexes and service availability indexes;

2. The method of claim 1, wherein the associated application metrics include a connection specification database during the operation of the reconfiguration cluster application;

3. The method of claim 1, wherein the related application index comprises a number of copies and memory resources in a process of reconstructing the cluster application;

4. The method of claim 1, wherein the associated application index comprises a volume usage of the reconstructed cluster application;

5. The method of claim 4, wherein expanding the storage volume of the reconstructed cluster application comprises:

6. The method of claim 1, wherein the associated application index comprises a specified version of application for mapping access by a specified type of network request during operation of the reconfiguration cluster application;

7. The method of claim 6, wherein said determining whether the current type of network request maps to access the specified version of the application according to the update policy comprises:

8. The method of claim 1, wherein the associated application metrics include a network mapping policy deployed in a reconfiguration cluster application;

9. The method of claim 1, wherein the associated application index comprises a mirrored download address of a reconstructed cluster application;

10. A non-transitory computer storage medium storing computer executable instructions that when executed by a computer enable: