Service container management method and system, computer device, medium, and product

Abstract

The present application relates to the technical field of cloud computing, and in particular to a service container management method and system, a computer device, a medium, and a product. The service container management method comprises: a network unit constructs an extended network; the network unit creates a first virtual network adapter pair and enables a network space of a container management platform to be communicated with a network space of the extended network; the network unit creates a second virtual network adapter pair and enables a network space of a service container to be communicated with the network space of the extended network, so as to enable the network space of the container management platform to be communicated with a network space of a container processing engine; the container management platform sends a scheduling instruction, wherein the scheduling instruction comprises suspending the service container; and the container processing engine acquires the scheduling instruction, and schedules the service container on the basis of the scheduling instruction.

Description

Method, system, computer device, medium and product for managing service container

[0001] Cross-reference to related applications

[0002] The present application claims priority to the Chinese patent application No. 202410634892.4, filed on May 22, 2024, and entitled “Method, system, computer device, medium and product for managing service container”, the content of which is incorporated herein by reference in its entirety. TECHNICAL FIELD

[0003] The present application relates to a method for managing a service container, a system for managing a service container, a computer device, a non-transitory computer-readable storage medium, and a computer program product. BACKGROUND

[0004] A container management platform such as Kubernetes can effectively utilize resources in a cluster and balance workloads among nodes in the cluster. Kubernetes is an open source platform for managing containerized applications on multiple hosts in a cloud platform.

[0005] However, in some specific scenarios, a single container management platform / container processing engine usually has certain functional defects. For example, a container management platform usually cannot suspend a service container but terminate the service container, and the intermediate state occurring before restarting the service container cannot be recovered. SUMMARY

[0006] According to an embodiment disclosed in the present application, in a first aspect, a method for managing a service container is provided. The method is applied to a system for managing a service container, and the system includes a network element, a container management platform, and a container processing engine. The method includes: the network element constructs an extension network; the network element creates a first pair of virtual network cards, inserts one of the first pair of virtual network cards to the extension network, and inserts the other to a proxy container package, and connects the network space of the container management platform and the extension network; the proxy container package is created by the container management platform and is accommodated in a cluster, and is used to proxy services carried by a service container in the cluster; the network element creates a second pair of virtual network cards, inserts one of the second pair of virtual network cards to the extension network, and inserts the other to the cluster to connect the service container located in the container processing engine, connects the network space of the service container and the extension network, and connects the network space of the container management platform and the container processing engine; the container management platform sends a scheduling instruction; the scheduling instruction includes suspending the service container; the container processing engine acquires the scheduling instruction and schedules the service container based on the scheduling instruction.

[0007] In a second aspect, a computer device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method for managing a service container according to any one of the embodiments when executing the computer program.

[0008] In a third aspect, a non-transitory computer readable storage medium is provided, having stored thereon a computer program, wherein the computer program, when executed by a processor, implements the steps of the method for managing a service container according to any one of the embodiments.

[0009] In a fourth aspect, a computer program product is provided, wherein the computer program product, when executed by a processor, implements the steps of the method for managing a service container according to any one of the embodiments.

[0010] The details of one or more embodiments of the application are set forth in the accompanying drawings and the description below. Other features and advantages of the application will be apparent from the description, the drawings, and the claims. BRIEF DESCRIPTION OF DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the embodiments will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without any creative effort on the basis of these drawings.

[0012] FIG. 1 is a structural schematic diagram of a service container management system according to one or more embodiments of the present application;

[0013] FIG. 2 is another structural schematic diagram of a service container management system according to one or more embodiments of the present application;

[0014] FIG. 3 is a structural schematic diagram of a node according to one or more embodiments of the present application;

[0015] FIG. 4 is a structural schematic diagram of a proxy container package according to one or more embodiments of the present application;

[0016] FIG. 5 is a flow schematic diagram of a method for managing a service container according to one or more embodiments of the present application;

[0017] FIG. 6 is a flow schematic diagram of a method for managing a service container according to another one or more embodiments of the present application;

[0018] FIG. 7 is a flow schematic diagram of suspending a service container according to one or more embodiments of the present application;

[0019] FIG. 8 is a flow schematic diagram of starting a service container according to one or more embodiments of the present application;

[0020] FIG. 9 is a flow diagram illustrating creating a business container according to one or more embodiments of the present application;

[0021] FIG. 10 is a flow diagram illustrating deleting a business container according to one or more embodiments of the present application;

[0022] FIG. 11 is a structural diagram illustrating a computer device according to one or more embodiments of the present application;

[0023] FIG. 12 is a structural diagram illustrating a non-transitory computer-readable storage medium according to one or more embodiments of the present application;

[0024] FIG. 13 is a structural diagram illustrating a computer program product according to one or more embodiments of the present application. DETAILED DESCRIPTION

[0025] In order to make the purposes, technical solutions and advantages of the present application clearer, further detailed description will be given below in combination with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and should not be used to limit the present application.

[0026] To solve the technical problem of defects in container management in the related art, the present application provides a business container management method, a business container management system, a computer device, a non-transitory computer-readable storage medium and a computer program product. The business container management method comprises: a network element constructs an extension network; the network element creates a first virtual network card pair, inserts one of the first virtual network card pair to the extension network, and inserts the other to a proxy container package; connects the network space of the container management platform and the extension network; wherein the proxy container package is created by the container management platform and is accommodated in a cluster, and is used to proxy the services carried by the business container in the cluster; the network element creates a second virtual network card pair, inserts one of the second virtual network card pair to the extension network, and inserts the other to the business container located at the container processing engine in the cluster; connects the network space of the business container and the extension network, and connects the network space of the container management platform and the container processing engine; the container management platform sends a scheduling instruction; wherein the scheduling instruction comprises suspending the business container; the container processing engine acquires the scheduling instruction and schedules the business container based on the scheduling instruction. The detailed technical solutions of the present application are described below.

[0027] Referring to FIG. 1, FIG. 1 is a structural diagram illustrating one or more embodiments of the business container management system of the present application.

[0028] In one or more embodiments, the business container management system comprises a container management platform 11, a container processing engine 12 and a network element 13.

[0029] The container management platform 11 is configured to send a scheduling instruction. The scheduling instruction includes suspending a service container. For example, the container management platform 11 can be Kubernetes (an open source, used to manage containerized applications on multiple hosts in a cloud platform) or the like.

[0030] The container processing engine 12 is configured to obtain the scheduling instruction and schedule the service container based on the scheduling instruction. Optionally, the container processing engine 12 can be Docker (an open source application container engine), Containerd (a lightweight container management tool), Podman (a container engine) or the like.

[0031] The network unit 13 is configured to build an overlay network, create a first pair of virtual network cards, insert one of the first pair of virtual network cards into the overlay network and the other into a proxy container package, and connect the network space of the container management platform 11 and the overlay network. The proxy container package is created by the container management platform 11 and is located in the cluster, and is used to proxy the service carried by the service container in the cluster. The network unit 13 is further configured to create a second pair of virtual network cards, insert one of the second pair of virtual network cards into the overlay network and the other into the service container located in the cluster and connected to the container processing engine 12, and connect the network space of the service container and the overlay network, so as to connect the network space of the container management platform 11 and the container processing engine 12.

[0032] In this way, the container management platform 11 and the container processing engine 12 can be connected through the overlay network, which facilitates the cooperation of the two in service container management, reduces at least part of the limitations when the container management platform 11 and the container management engine are used separately, improves the functionality of service container management, and reduces the risk of losing the intermediate state of the service container. At the same time, the operation on the specific service container can be mapped to the operation on the container management platform 11 / container processing engine 12, which significantly reduces the management difficulty of service container management, improves the flexibility of service container management, and further enriches the application scenarios.

[0033] For example, as shown in FIG. 2, the first pair of virtual network cards can include virtual network card veth1 and virtual network card veth2, which are inserted into the overlay network and the proxy container package, respectively, to connect the network space of the overlay network and the container management platform. The second pair of virtual network cards can include virtual network card veth3 and virtual network card veth4, which are inserted into the overlay network and the service container located in node N2, respectively, to connect the network space of the service container N2 and the overlay network.

[0034] Please refer to FIG. 2 to FIG. 4, FIG. 3 is a structural schematic diagram of one or more embodiments of the node of the present application, and FIG. 4 is a structural schematic diagram of one or more embodiments of the agent container encapsulation of the present application.

[0035] In one or more embodiments, the management system further comprises a scheduler. The cluster is provided with a virtual container encapsulation.

[0036] The scheduling instruction can be in the form of an instruction / request, which is sent to the container processing engine by the container management platform, or the scheduling instruction can be in the form of a scheduling identifier, which can be identified by the container processing engine.

[0037] Specifically, the scheduler can be connected to the container management platform and the virtual container encapsulation, respectively. The scheduler is used to obtain the virtual container encapsulation matched with the scheduling request and inject the scheduling identifier into the annotation of the virtual container encapsulation. So that the container processing engine detects the annotation of the virtual container encapsulation. The specific working principle of the scheduler will be described in detail in the management method of the business container, and will not be repeated here.

[0038] Optionally, the cluster can be composed of a group of independent and interconnected computer nodes, and the multiple computer nodes are managed in the mode of a single system. In other words, the cluster can include multiple nodes, and the nodes are created by the container management platform or the container processing engine.

[0039] The node can be provided with a business service container and / or an agent container encapsulation.

[0040] For example, as shown in FIG. 2, the cluster can include node N1, node N2 and node N3. The node N1 and the node N2 are respectively provided with the business container C1 and the business container C2, and the agent container encapsulation is provided in the node N3. For example, when the service that cannot be directly accessed by the container management platform is started in the business container of the node N1 and / or the node N2, the reverse proxy service can be performed through the agent container encapsulation of the node N3 to proxy the service of the business container of the node N1 and / or the node N2, so that the container management platform can access.

[0041] As shown in FIG. 3, the node N4 can include the business container C3 and the agent container encapsulation.

[0042] It should be noted that FIG. 2 and FIG. 3 are examples of the cluster, the node, the business container and the agent container encapsulation in the present embodiment, and the number of nodes in the cluster, the number of business containers in one node and the number of agent container encapsulations are not strictly limited. For example, the cluster can further include multiple agent container encapsulations, which are respectively used to manage part of the nodes or the business containers, which are not limited here.

[0043] Please continue to refer to FIG. 2 to FIG. 4. In one or more embodiments, the agent container package can include an agent service container.

[0044] The agent service container can receive a service request sent by the container management platform to the agent container package, forward the service request to the corresponding business container, so that the business container responds to the service request. The service request can be generated by the container management platform or sent by the user to the container management platform. As such, the embodiment can implement cross-platform execution of service requests, further improving the convenience and functionality of business container management.

[0045] Further, the agent container can also include a configuration detection container and a configuration file.

[0046] Specifically, the configuration detection container can detect whether there is information change in the configuration file. In response to the information change in the configuration file, the main process number of the agent service container is screened as a second main process number; a loading instruction is sent to the agent service container based on the second main process number; so that the agent service container receives the loading instruction, restarts the reverse proxy service to reload the updated configuration file. The more detailed principles of this part will be described in detail in the management method of the business container later, and will not be repeated here.

[0047] The specific definition of the business container management system can refer to the definition of the management method of the business container later, and will not be repeated here. Each module in the business container management system described above can be realized by software, hardware and their combinations in whole or in part. Each module described above can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory in the computer device in software form, so as to call and execute the operations corresponding to each module by the processor.

[0048] The working principle of the business container management method of the present application is described in detail as follows.

[0049] Please refer to FIG. 5, which is a flowchart of one or more embodiments of the management method of the business container of the present application. The management method of the business container can be applied to the business container management system described above.

[0050] S101: A network element constructs an extended network.

[0051] In the embodiment, the extended network can be an overlay network (overlay network, overlay network) and the like. The extended network can be one or more virtual logical networks constructed on the same underlay network (a kind of bottom layer network) through network virtualization technology.

[0052] S102: The network element creates a first virtual network card pair, inserts one of the first virtual network card pair into the extension network, and inserts the other into the proxy container package, connects the network space of the container management platform and the extension network; wherein the proxy container package is created by the container management platform and accommodated in the cluster, and is used to proxy the service container in the cluster.

[0053] In this embodiment, the proxy container package can be created by the container management platform and accommodated in the cluster, so that the container management platform can proxy the service container through the proxy container package. The specific working principle will be illustrated in the following examples.

[0054] As the name implies, the virtual network card pair can include two virtual network cards. The virtual network card can simulate a network environment with a software network adapter. That is, one of the first virtual network card pair is inserted into the extension network, and the other is inserted into the proxy container package. In this way, the proxy container package can access the extension network.

[0055] S103: The network element creates a second virtual network card pair, inserts one of the second virtual network card pair into the extension network, and inserts the other into the cluster to connect the service container located in the container processing engine, connects the network space of the service container and the extension network, and connects the network space of the container management platform and the container processing engine.

[0056] In this embodiment, one of the second virtual network card pair can be inserted into the extension network; the other can be inserted into at least part of the nodes in the cluster, or can be inserted into the service container, which is not limited here. In this way, the container processing engine can access the extension network. In combination with the implementation of the container management platform accessing the extension network in step S102, the container processing engine can communicate with the container management platform.

[0057] S104: The container management platform sends a scheduling instruction; wherein the scheduling instruction includes suspending the service container; the container processing engine obtains the scheduling instruction and schedules the service container based on the scheduling instruction.

[0058] In this embodiment, in response to the network space of the container management platform and the container processing engine being connected through the extension network, the two can transmit instructions to each other to flexibly manage the service container. The container management platform can send a scheduling instruction, and the container processing engine can obtain the scheduling instruction and schedule the service container based on the scheduling instruction.

[0059] The scheduling instruction can include a suspension instruction to enable the container processing engine to control the service container to suspend, thereby avoiding the limitations of the container management platform in suspending the service container and reducing the risk of state loss, data loss, network connection interruption, application state reset, etc. caused by the container management platform controlling the suspension of the service container.

[0060] Therefore, the container management platform and the container processing engine can be both added to the overlay network, so that the container management platform and the container processing engine are connected, which can facilitate the flexible interaction between the two to jointly manage the business containers in the cluster, improve the functionality of the business container management, and overcome the defects of the separate business container management of the two. In addition, the container management platform can be implemented by including a scheduling instruction for suspending the business container, so that the container processing engine suspends the business container based on the scheduling instruction, which is beneficial to reduce the risk of losing the intermediate state of the business container, thereby improving the reliability of the business container management. At the same time, the embodiment can implement cross-platform execution of service requests, improve the convenience and functionality of the business container management, and thus facilitate the enrichment of application scenarios.

[0061] Please refer to FIG. 6, which is a flowchart of another or more embodiments of the method for managing the business container.

[0062] In the embodiment, the network unit can create a virtual multicast group and a network bridge on each node in the cluster. The network unit connects the virtual device of the virtual multicast group to the network bridge. The network unit starts the network bridge and the virtual device connected to the network bridge to form an overlay network. The network unit plans the network segment information of the overlay network to form the overlay network for the business container management in the embodiment. In this way, the management of the business container can introduce virtual network technology to facilitate the management of the business container through the virtual network, improve the efficiency of the business container management, and enrich the functionality of the business container management.

[0063] Specifically, a VXLAN (Virtual Extensible Local Area Network) device multicast group can be created on each node in the cluster. When creating the VXLAN, the VXLAN id (Identity document), the out-interface, and the port information can be specified. A network bridge device is created, and the created VXLAN device is connected to the network bridge. The network bridge device is started. The VXLAN device connected to the network bridge device is started. In this way, the embodiment constructs an overlay network on the cluster through the VXLAN. At the same time, the entire overlay network segment information can be planned. For example, the network segment information can include the start ip (Internet Protocol), the end ip, the gateway information, and the like.

[0064] Further, the network unit can query the bridge of the overlay network and the network segment information thereof. The network segment information is configured as an Internet Protocol address of one of the first pair of virtual network cards and is inserted into the bridge; a process number of the proxy container package is obtained, a network namespace thereof is obtained based on the process number, and the other of the first pair of virtual network cards is moved to the network namespace of the proxy container package.

[0065] One of the first pair of virtual network cards is inserted into the overlay network, and the other is inserted into the proxy container package, connecting the network spaces of the container management platform and the overlay network.

[0066] In this way, the first pair of virtual network cards can connect the network namespace of the overlay network and the network namespace of the container management platform, so as to connect the network spaces of the container management platform and the overlay network.

[0067] For example, in the application environment of LINUX (an operating system), the scheme of connecting the first pair of virtual network cards to the container management platform and the overlay network is described in detail.

[0068] The first pair of virtual network cards can be created, such as the virtual network card veth1 and the virtual network card veth2 as shown in FIG. 6.

[0069] The instruction “ip link add veth1 type veth peer name veth2” can be executed in LINUX to create the virtual network card veth1 and the virtual network card veth2. The virtual network card veth1 is moved into the network namespace of the pod container; for example, the instruction “ip link set veth2 netns$pod_netns” can be executed in LINUX, which is equivalent to inserting a virtual network card (i.e., the virtual network card veth2) into the pod container. The virtual network card veth2 is started, and the network segment information of the overlay network planned in the foregoing is configured as the IP address of the virtual network card veth2.

[0070] The virtual network card veth1 is inserted into the bridge of the created overlay network; for example, the instruction “ip link set veth1 netns$overlay_netns” can be executed in LINUX to move the virtual network card veth1 to the bridge. The virtual network card veth1 is started, and the IP address of the virtual network card veth1 is configured as the network segment information of the overlay network planned in the foregoing.

[0071] As such, the network space of the proxy container encapsulating the reverse proxy service function of the container management platform such as Kubernetes and the overlay network space are connected, can exchange network data, and are beneficial to improve the efficiency and flexibility of the connection between the container management platform and the container processing engine.

[0072] The bridge of the extension network and the network segment information thereof can be queried, the network segment information is configured as the Internet Protocol address of one of the second virtual network card pair, and the bridge is inserted; the other of the second virtual network card pair is moved to the network namespace where the service container is located.

[0073] As such, one of the second virtual network card pair is inserted into the extension network, and the other is inserted into the service container located in the container processing engine in the cluster, the network spaces of the service container and the extension network are connected, and the network spaces of the container management platform and the container processing engine are connected.

[0074] Similarly, the scheme for connecting the container management platform and the extension network through the second virtual network card pair is described in detail in the LINUX application environment.

[0075] The second virtual network card pair, such as the virtual network card veth3 and the virtual network card veth4, the virtual network card veth5 and the virtual network card veth6 as shown in FIG. 6, can be created to connect the extension network to the nodes N2 and N3 respectively.

[0076] Specifically, the connection between the virtual network card veth3 and the virtual network card veth4 and the node N2 and the extension network is described in detail. After creating the service container on the cluster node through the service logic of the scheme, the network space of the service container needs to be connected to the overlay network space.

[0077] First, the second virtual network card pair, i.e., the virtual network card veth3 and the virtual network card veth4, is created.

[0078] The “ip link add veth3 type veth peer name veth4” instruction is executed in linux to create the virtual network card veth3 and the virtual network card veth4.

[0079] The virtual network card veth4 can be moved to the network namespace where the service container is located, for example, the “ip link set veth4 netns $pod_netns” instruction can be executed in linux. The virtual network card veth4 is started, and the IP address of the virtual network card veth4 is configured as the network segment information of the extension network planned in the foregoing.

[0080] The virtual network card veth3 can be inserted into the bridge of the created overlay network, for example, in Linux, the instruction "ip link set veth4 netns$overlay_netns" is executed. The virtual network card is started, and the IP address of the virtual network card veth3 is configured as the network segment information of the overlay network planned in the foregoing.

[0081] In this way, the network space of the service container is connected with the overlay network space. Moreover, since the network space encapsulated by the proxy container of the container management platform is also connected with the overlay network space, the network space encapsulated by the kubernetes proxy container is also connected with the network space of the service container. Therefore, the network platform of the container management platform is connected with the network platform of the container processing engine.

[0082] The above describes the detailed principle of the management system architecture for building the service container in detail, and the following describes the application process of the management system of the service container, that is, the following describes the management principle of the service container.

[0083] Please continue to refer to FIG. 6. In one or more embodiments, the container processing engine changes the service container in the cluster.

[0084] The container processing engine modifies the service information in the configuration file based on the change. The proxy container encapsulation loads the updated configuration file in response to detecting the change of the service information. In this way, the container management platform can implement the service discovery function, and can automatically update the configuration file of the reverse proxy service when the service is changed, so as to respond to the change of the service in time, improve the timeliness of the service container management, and also help to improve the reliability of the service container management.

[0085] In detail, the service container can carry services, and each service can have a corresponding URL (Uniform Resource Location, network link) access path. The information of the URL can be written into the configuration file of the proxy container encapsulation, and the matching access path information is configured respectively, so that the service carried by the service container can be accessed through the reverse proxy service. It can be considered that the service discovery function is equivalent to registering the service information into the configuration file.

[0086] Please continue to refer to FIG. 6. In one or more embodiments, the container management platform can create a reverse proxy service and a proxy container package. The proxy container package can record resource information, startup script information, and the like used by the reverse proxy service. The reverse proxy service is packaged in the proxy container package, and the proxy container package is deployed to the cluster. Optionally, the container management platform can deploy the proxy container package in the cluster by creating a container package (such as a pod, etc.). In this way, when the proxy container package is running, the reverse proxy service can run to provide related functions to implement the cooperation between the container management platform and the container processing engine to manage business containers and services.

[0087] The container management platform can also create a configuration dictionary, map the configuration dictionary to the proxy container package, and form a configuration file. The configuration dictionary is used to record service information that needs to be proxied. In this way, the container management platform and the container processing engine can detect changes in the cluster through the configuration dictionary and the configuration file, and can flexibly use the configuration file and the configuration dictionary for business container management. Compared with the management mode of calling the APIs of the container management platform and the container processing engine, the container management platform and the container processing engine in this embodiment can improve the flexibility and scalability of business container management, and also help to reduce the difficulty of writing business container management logic, and can improve the efficiency of business container management and maintenance.

[0088] Further, the specific structure and working principle of the proxy container package are described below. Please refer to FIG. 4 and FIG. 6.

[0089] In one or more embodiments, the proxy container package includes a proxy service container. As the name implies, the proxy service container is a container that specifically implements the reverse proxy service and can send a service request to a business container. In this way, the container management platform can schedule business containers and services, and the container management platform can also implement service scheduling of business containers across platforms. In other words, the container management platform can have business container scheduling capabilities and service proxying capabilities.

[0090] Specifically, the container management platform can obtain a service request. The service request can be generated by the container management platform according to services, environments, and the like, or can be sent by a user to the container management platform, which is not limited here.

[0091] The container management platform can send the service request to the proxy container package. The proxy service container can forward the service request to the corresponding business container in response to receiving the service request, so that the business container responds to the service request. In this way, the container management platform can proxy services carried by business containers in the cluster through the proxy container package.

[0092] Still further, the agent container package can further comprise a configuration detection container, which is capable of automatically detecting the change of the configuration file and synchronizing to the container management platform in time, thereby facilitating the consistency of the information detected by the container management platform and the container processing engine, and further facilitating the reliability and functionality of the business container management.

[0093] Specifically, the configuration detection container detects whether there is a change in the information in the configuration file.

[0094] The configuration detection container can filter the main process number of the agent service container as a second main process number in response to the change in the information in the configuration file. The agent service container is sent a loading instruction based on the second main process number. In this way, the agent service container restarts the reverse proxy service to reload the updated configuration file in response to receiving the loading instruction. Optionally, the frequency of updating the configuration file can be identified, and the frequency of restarting the reverse proxy service can be dynamically adjusted based on the updating frequency, which is not limited herein.

[0095] For example, the content included in the configuration dictionary is shown in FIG. 6:

[0096] location / tomcat{

[0097] proxy_pass http: / / 10.22.1.10:8080 / ;

[0098] }

[0099] location / tomcat2{

[0100] proxy_pass http: / / 10.22.1.11:8080;

[0101] }

[0102] The above code example records the address of tomcat, which is a web (World Wide Web) application server.

[0103] Taking the container management platform as Kubernetes as an example, the agent container package created by the Kubernetes is taken as a pod with agent service function. Broadly speaking, the pod can be referred to as a container package or a root container. The pod can include a group of containers, i.e., the pod can include one or more containers.

[0104] The Kubernetes can create a pod with agent service function, and the pod can include at least two containers. One of the containers can be a container responsible for reverse proxy service, i.e., an agent service container; and the other container can be a container capable of listening to a configuration file, i.e., a configuration detection container.

[0105] The kubernetes cluster can create a configuration dictionary, which can record the service information of the service that needs to be proxied, and map the configuration dictionary to a configuration file in the pod. The proxy service container in the pod can read the service information recorded in the configuration file and provide corresponding proxy service.

[0106] Further, please continue to refer to FIG. 6. In one or more embodiments, the container processing engine can create a new business container or add a new service to the business container.

[0107] The container processing engine can obtain the configuration file and write the first information to the configuration file. The first information includes the service information of the newly added business container or service. And / or, the container processing engine can delete the original business container or delete the service in the original business container, take the original business container as the sixth target container, and delete the second information by the container processing engine. The second information includes the service information of the deleted original business container or service. In this embodiment, when the container processing engine manages the business container, the configuration file can be modified accordingly, so that the container management platform can automatically detect the changes of the configuration file and synchronize to the container management platform in time, thereby improving the consistency of the information detected by the container management platform and the container processing engine, and further improving the reliability and functionality of the business container management. In simple terms, when a new business container (or a service that needs to be proxied) is created or deleted, the container processing engine modifies the configuration dictionary accordingly, that is, modifies the configuration file of the proxy service container, to add or delete the service information recorded in the configuration file.

[0108] Further, the proxy container encapsulates traverses the process information of its process namespace, filters the main process number of the reverse proxy service in the process information as the first main process number. The proxy container encapsulates sends a load instruction to the first main process number, triggers the reverse proxy service to reload the configuration file, thereby facilitating the reverse proxy service to learn the changed configuration file in time, improving the consistency of the service information detected by the reverse proxy service and the real service information, and also enabling the reverse proxy service to reload the configuration file without interrupting the ongoing service proxy, thereby reducing the interference to the business proxy process, and further improving the reliability of the business container management. In combination with the implementation mode of the proxy container encapsulation as the pod described in the foregoing, the configuration detection container of the pod can traverse the process information under its process namespace when detecting the change of the configuration file, until the main process number of the reverse proxy service is filtered. The configuration detection container can send a restart signal to the main process number of the reverse proxy service, triggering the daemon process of the reverse proxy service to restart, thereby realizing the reloading of the configuration file.

[0109] The following describes the working principle of the container management platform of the present application in scheduling the container processing engine.

[0110] Please continue to refer to FIG. 6. In one or more embodiments, the container management platform can obtain a creation request, and create a new virtual container package based on the creation request. In other words, the creation request is used to instruct the creation of the virtual container package.

[0111] The scheduler can identify the new virtual container package and reserve resources for the new virtual container package. At the same time, the scheduler can also screen the nodes in the cluster, and the screened nodes are used as the first target nodes associated with the new virtual container package. The node information of the first target node is injected into the annotation of the new virtual container package, thereby realizing the creation of the virtual container package for the node. The virtual container package is created by the container management platform, so the container management platform can schedule the virtual container package, and thus can schedule the node and the business container through the virtual container package. At the same time, the virtual container package can participate in the scheduling process of the scheduler without occupying resources, and operating the virtual container package is equivalent to operating the business container in the container processing engine.

[0112] Specifically, taking the scheduling instruction in the form of an identification (i.e., a scheduling identification) as an example, the container management platform can assign the scheduling identification to the virtual container package located in the cluster.

[0113] The container processing engine can detect the virtual container package and judge whether the virtual container package is assigned with the scheduling identification. In response to the virtual container package being assigned with the scheduling identification, the container processing engine can identify the business container pointed to by the scheduling identification and take it as a target container, and schedule the target container according to the scheduling identification. This is conducive to simplifying the business container management logic. The container management platform can assign the scheduling identification to the virtual container package created by it, and the container processing engine can detect the virtual container package to automatically complete the receiving and execution of the scheduling instruction.

[0114] Further, the container management platform can initiate a scheduling request to the scheduler. The scheduler obtains the virtual container package matched with the scheduling request, and injects the scheduling identification into the annotation of the virtual container package. The container processing engine detects the annotation of the virtual container package. In other words, the processing process of identifying the target virtual container package can be transferred to the scheduler in this embodiment. The scheduling instructions sent by it can all be sent to the scheduler, which can fully utilize the function of the scheduler while also reducing the management burden of the container management platform.

[0115] Optionally, in response to completing the execution of the scheduling instruction, the container processing engine can also obtain a configuration file to update the configuration file, improve the matching of the service information reflected by the configuration file with the actual situation, and thus improve the reliability of the business container management.

[0116] Specifically, the container processing engine can update the configuration file based on the scheduling instruction, trigger the container management platform to reload the updated configuration file. Let the container management platform proxy the service carried by the business container in the cluster based on the updated configuration file.

[0117] In one or more embodiments, the container management platform of the present application can update the resources of the business container through the container processing engine. The specific principles of updating the resources of the business container of the present application are described below.

[0118] Specifically, the scheduler can take the virtual container package as the target virtual package in response to detecting that the virtual container package carries the to-be-updated identifier. The resource usage recorded in the target virtual package is updated, and the target virtual package is given an updated identifier.

[0119] The container processing engine updates the resource data updated by the third target container to the resource information recorded by the target virtual package in response to detecting the updated identifier. For example, the instruction “docker update–cpu $cpu_num $container_name” can be executed to update the CPU number of the third target container. In this way, the embodiment can update the resource information occupied by the business container through the scheduling of the virtual container package, and can improve the functionality of business container management. At the same time, it is also beneficial to macroscopically schedule the resource usage of the business container, improve the flexibility and accuracy of resource allocation, and thus be beneficial to guarantee the reliable completion of the service by the business container.

[0120] For example, the resources occupied by the business container can include CPU (Central Processing Unit), GPU (Graphics Processing Unit), etc., which are not limited here.

[0121] Further, the amount of resources in the cluster can also be considered in the embodiment to reduce the risk of adding resources to the business container when the resource capacity is insufficient, reduce the risk of abnormal operation of the cluster, and improve the reliability of the cluster operation at the business container management level. For example, the scheduler can identify the update target of the resource update. The scheduler checks whether the remaining resources of the second target node associated with the target virtual package are sufficient in response to the update target being to increase resource usage. In response to the remaining resources being insufficient, the target virtual package is given a resource insufficient identifier.

[0122] Meanwhile, the scheduler can identify that the update target is to reduce resource usage, and can execute resource scheduling of the service container of the second target node without checking whether the remaining resources of the target virtual package are sufficient, considering that reducing resource usage does not involve a resource insufficient scenario, thereby improving the efficiency of service container management.

[0123] Further, as described above, the container processing engine can detect the identification carried by the target virtual package. The container processing engine can update the third target container resource to the resource recorded by the target virtual package in response to detecting the updated identification. Meanwhile, after the update is completed, the updated identification carried by the target virtual package can be deleted, reducing the risk of interference with subsequent resource update operations and further improving the reliability of service container management.

[0124] Similarly, the container processing engine can also respond to the detection of the resource insufficient identification, and not perform the mirror update operation on the third target container, and clear the resource insufficient identification in the target virtual package.

[0125] In one or more embodiments, the container management platform of the present application can suspend the service container through the container processing engine. The specific principles of suspending the service container of the present application are described below.

[0126] Please refer to FIG. 7, which is a flowchart of one or more embodiments of the present application for suspending the service container.

[0127] Specifically, the container processing engine can detect the virtual container package arranged in the cluster.

[0128] The container processing engine responds to the virtual container package carrying the suspension identification, and analyzes the first target container pointed by the suspension identification. In other words, the service container that needs to be suspended as indicated by the suspension identification can be analyzed as the first target container. The suspension identification can be sent by the scheduler / container management platform to the virtual container package as described above, and will not be described here.

[0129] The target network card pair is a second virtual network card pair connecting the first target container and the extended network, i.e., the second virtual network card pair connecting the first target container and the extended network is taken as the target network card pair. The container processing engine unbinds one of the target network card pair from the bridge of the extended network, and migrates the other from the network space of the first target container to the default network space, and deletes the target network card pair, thereby realizing the suspension of the virtual container package as the first target container.

[0130] Further, the configuration file in the proxy container package can also be adjusted accordingly. Specifically, the proxy information in the configuration file related to the first target container can be deleted, and the proxy container package / proxy service container / reverse proxy service container is triggered to reload the configuration file, so as to delete the proxy service. In addition, the node information marked in the virtual container package can be linked to the corresponding node, and the business container associated with the node can be stopped, so as to suspend the business container, thereby effectively overcoming the problem that some container management platforms cannot reliably suspend the business container and even easily lose the state. The container processing engine is ingeniously used to reliably suspend the business container, which can significantly improve the reliability of the business container management.

[0131] In one or more embodiments, the container management platform of the present application can start the business container through the container processing engine. The specific principles of starting the suspended business container of the present application are described below.

[0132] Referring to FIG. 8, FIG. 8 is a flowchart of one or more embodiments of starting the business container of the present application.

[0133] Specifically, the container management platform can directly or through the dispatcher inject a start identifier into the annotation of the virtual container package associated with the second target container. The container processing engine starts the second target container in response to detecting the start identifier, and connects the network space of the second target container and the container management platform, so that the second target container in the suspended state can be reliably started, the start and stop of the business container can be reliably realized, the state loss and data loss can be reduced, and the reliability of the business container management strategy can be improved.

[0134] Further, when the second target container is successfully started, the network space of the second target container can be processed to access the extended network. For example, the network namespace of the second target container can be connected to the network space of the container management platform in the manner of constructing the network as described above.

[0135] Further, the service of the second target container can be updated to the service information, the configuration file of the reverse proxy service / proxy container package is updated, and the reverse proxy service is allowed to proxy the service of the second target container. In addition, the virtual container package associated with the second target container can be queried, the node information marked in the virtual container package can be linked to the corresponding node, and the business container associated with the node can be started, so as to reliably start the business container.

[0136] In one or more embodiments, the container management platform of the present application can create a business container through the container processing engine. The specific principles of creating a new business container of the present application are described below.

[0137] Referring to FIG. 9, FIG. 9 is a flowchart of one or more embodiments of the application for creating a service container.

[0138] Specifically, the container processing engine, in response to detecting that the virtual container package carries a new-build identifier, can parse the resource usage indicated by the new-build identifier. A new service container is created based on the parsed resource usage.

[0139] The network unit connects the network space of the new service container and the container management platform.

[0140] For example, a virtual container package can be created, which can be a custom resource type of the container management platform. The virtual container package does not occupy resources at the time of creation, but the scheduler can identify the virtual container package and record the resources applied by the virtual container package. In this way, a certain amount of resources can be reserved for the virtual container package, which is equivalent to reserving resources for the newly created service container. The scheduler can be used to filter the node that is most suitable for carrying the virtual container package, and the node information of the node used to carry the virtual container package can be injected into the annotation of the virtual container package, but the virtual container package will not be truly scheduled as a real container package.

[0141] Further, the service container processing engine listens to the node information in the annotation of the virtual container package. When it is found that the annotation of the virtual container package has node information updates, the service container is created on the corresponding node, and the resource usage of the service container is consistent with the recorded resource usage in the virtual container package.

[0142] When the service container is successfully created, the network namespace of the new service container is connected with the network namespace of the container management platform, so as to access the extended network. The services proxied by the new service container can also be updated in the service information, so as to update the configuration file of the reverse proxy service, so that the reverse proxy server can proxy the newly created service container.

[0143] In one or more embodiments, the container management platform of the application can delete the existing service container in the cluster through the container processing engine. The specific principle of deleting the service container of the application is described below.

[0144] Referring to FIG. 10, FIG. 10 is a flowchart of one or more embodiments of the application for deleting a service container.

[0145] Specifically, the container processing engine, in response to detecting that the virtual container package carries a delete identifier, identifies a fourth target container pointed to by the delete identifier. The fourth target container is unbound from the extended network, and the second virtual network card pair associated with the fourth target container is deleted.

[0146] The container processing engine deletes the proxy information associated with the fourth target container in the configuration file, triggers the proxy container to reload the configuration file. The container processing engine deletes the fourth target container in the node of the cluster.

[0147] For example, when performing the deletion service container process, a deletion identifier can be injected in the annotation of the virtual container package. When the service container processing engine listens to the deletion identifier in the annotation of the virtual container package, one of the previously configured second virtual network card pair can be unbound from the bridge of the extension network, the other one can be moved from the network namespace of the fourth target container to the default network namespace, and the second virtual network card pair associated with the fourth target container can be deleted.

[0148] The proxy information related to the fourth target container in the configuration file of the reverse proxy service can be deleted, and the reverse proxy service can be triggered to reload the configuration file to achieve the deletion of the proxy service. According to the node information marked in the virtual container package, the corresponding node is linked, and the service container associated with it is deleted from the node.

[0149] Further, the deletion of the service container can also be monitored in real time. In response to the successful deletion of the service container, an interface of the container management platform is called to delete the virtual container package resource corresponding to the fourth target container. In this way, the deletion of the fourth target container can be considered to be completed.

[0150] Further, in one or more embodiments, the scheduling instruction can include at least one of a start instruction, an update instruction, a deletion instruction, and a new creation instruction in addition to the pause instruction to enable the pause of the service container. The start instruction can start the service container in the paused state; the update instruction can update the resource usage of the service container; the deletion instruction can delete the service container; and the new creation instruction can create a new service container.

[0151] Optionally, the start instruction, and / or the update instruction, and / or the deletion instruction, and / or the new creation instruction not included in the scheduling instruction can be transmitted and the corresponding functions can be implemented by calling the API of the container processing engine, which will not be described here.

[0152] For example, the scheduling instruction can include a pause instruction, a start instruction, an update instruction, a deletion instruction, and a new creation instruction, so as to basically involve each operation of the service container management, and improve the integrity and reliability of the service container management strategy in the embodiment.

[0153] Specifically, the container processing engine receiving the scheduling instruction can identify the category to which the scheduling instruction belongs.

[0154] In this way, the container processing engine can pause the first target container pointed to by the pause instruction in response to the scheduling instruction being the pause instruction.

[0155] The container processing engine can start a second target container pointed by the start instruction in response to the scheduling instruction being the start instruction.

[0156] The container processing engine can update a third target container pointed by the update instruction in response to the scheduling instruction being the update instruction.

[0157] The container processing engine can delete a fourth target container pointed by the delete instruction in response to the scheduling instruction being the delete instruction.

[0158] The container processing engine can create a new business container based on the new instruction in response to the scheduling instruction being the new instruction.

[0159] The specific implementation manners of suspending the business container, starting the business container, updating the resource used by the business container, deleting the business container, and creating the business container can be as described above, and thus will not be described here again.

[0160] In summary, taking the container management platform Kubernetes as an example, the application can avoid the capability defect that Kubernetes cannot stop the real pod it manages, and use the virtual pod operation mapping to operate the container. In this way, the application can ingeniously use the capability of the container processing engine such as Docker that can start and stop the container, so that the intermediate state of the container will not be lost in the process of starting and stopping the virtual pod, and the business use scenario of Kubernetes is effectively expanded.

[0161] It should be understood that although each step in the flowcharts of FIGS. 5-10 is shown in sequence according to the direction of the arrow, these steps are not necessarily executed in sequence according to the direction of the arrow. Unless otherwise explicitly stated herein, there is no strict order limitation for the execution of these steps, and these steps can be executed in other orders. Moreover, at least a part of the steps in FIGS. 5-10 can include multiple sub-steps or multiple stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution order of these sub-steps or stages is not necessarily sequential, but can be alternately executed with at least a part of other steps or sub-steps or stages of other steps.

[0162] In one or more embodiments, the computer program product provided by the application includes a computer program. The computer program is executed by a processor to implement the steps of the business container management method described above, and thus will not be described here again.

[0163] Please refer to FIG. 11, which is a structural schematic diagram of one or more embodiments of the computer device of the application.

[0164] In one or more embodiments, the computer device can be a server, and an internal structure diagram thereof can be as exemplarily shown in FIG. 11.

[0165] The computer device includes a processor, a memory and a network interface connected through a system bus. The processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for running the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is configured to communicate with an external terminal through a network connection. The computer program is executed by the processor to implement a service container management method.

[0166] Further, the computer device can further include a database, and the database of the computer device can be configured to store data required and / or generated in the service container management method.

[0167] Those skilled in the art can understand that the structure exemplarily shown in FIG. 11 is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0168] In one or more embodiments, a computer device is provided, including a memory, a processor and a computer program stored in the memory and executable on the processor, and the processor implements the following steps when executing the computer program:

[0169] S101: A network element constructs an extension network.

[0170] S102: The network element creates a first pair of virtual network cards, inserts one of the first pair of virtual network cards to the extension network, and inserts the other to a proxy container package. The network element connects the network space of the container management platform and the extension network; wherein the proxy container package is created by the container management platform and accommodated in the cluster, and is configured to proxy the service carried by the service container in the cluster.

[0171] S103: The network element creates a second pair of virtual network cards, inserts one of the second pair of virtual network cards to the extension network, and inserts the other to the cluster to connect the service container located in the container processing engine, so as to connect the network space of the service container and the extension network, and connect the network space of the container management platform and the container processing engine.

[0172] S104: The container management platform sends a scheduling instruction; wherein the scheduling instruction comprises suspending a service container; the container processing engine acquires the scheduling instruction and schedules the service container based on the scheduling instruction.

[0173] In one or more embodiments, the container management platform sends a scheduling instruction; when the container processing engine acquires the scheduling instruction, the processor executing the computer program further implements the following steps: the container management platform assigns a scheduling identifier to a virtual container package located in the cluster; the container processing engine detects the virtual container package and determines whether the virtual container package is assigned the scheduling identifier; in response to the virtual container package being assigned the scheduling identifier, the container processing engine identifies a service container pointed to by the scheduling identifier as a target container and schedules the target container according to the scheduling identifier.

[0174] In one or more embodiments, the container management platform assigns a scheduling identifier to a virtual container package located in the cluster; when the container processing engine detects the virtual container package, the processor executing the computer program further implements the following steps: the container management platform initiates a scheduling request to the scheduler; the scheduler acquires a virtual container package matching the scheduling request and injects a scheduling identifier into an annotation of the virtual container package; the container processing engine detects the annotation of the virtual container package.

[0175] In one or more embodiments, the processor executing the computer program further implements the following steps: the container management platform acquires a creation request and creates a new virtual container package based on the creation request; the scheduler identifies the new virtual container package and reserves resources for the new virtual container package; the scheduler screens nodes in the cluster and selects a node as a first target node associated with the new virtual container package; and the node information of the first target node is injected into the annotation of the new virtual container package.

[0176] In one or more embodiments, after scheduling the service container based on the scheduling instruction, the processor executing the computer program further implements the following steps: in response to completing the execution of the scheduling instruction, the container processing engine acquires a configuration file; the container processing engine updates the configuration file based on the scheduling instruction, triggers the container management platform to reload the updated configuration file; and the container management platform proxies services carried by the service containers in the cluster based on the updated configuration file.

[0177] In one or more embodiments, the container processing engine acquires the scheduling instruction, and when scheduling the service container based on the scheduling instruction, the processor executing the computer program further implements the following steps: the container processing engine detects a virtual container package disposed in the cluster; the container processing engine, in response to the virtual container package carrying a suspension identifier, resolves a first target container pointed to by the suspension identifier; the container processing engine unbinds one of the target network card pair from the bridge of the extended network and migrates the other from the network space of the first target container to the default network space, and deletes the target network card pair; wherein the target network card pair is a second virtual network card pair connecting the first target container and the extended network.

[0178] In one or more embodiments, the scheduling instruction includes starting the service container; the container management platform sends the scheduling instruction to the container processing engine; and when the container processing engine receives the scheduling instruction, the processor executing the computer program further implements the following steps: the container management platform injects a start identifier into an annotation of a virtual container package associated with a second target container; and the container processing engine, in response to detecting the start identifier, starts the second target container and connects network spaces of the second target container and the container management platform.

[0179] In one or more embodiments, the scheduling instruction includes updating resources of the service container; and when scheduling the service container based on the scheduling instruction, the processor executing the computer program further implements the following steps: the scheduler, in response to detecting that the virtual container package carries a to-be-updated identifier, takes the virtual container package as a target virtual package; the scheduler updates a resource usage amount of the target virtual package recorded therein and assigns an updated identifier to the target virtual package; and the container processing engine, in response to detecting the updated identifier, updates resource data updated by a third target container to resource information recorded by the target virtual package.

[0180] In one or more embodiments, when the scheduler updates the resource usage amount of the target virtual package recorded therein, the processor executing the computer program further implements the following steps: the scheduler identifies an update target of the resource update; the scheduler, in response to the update target being an increase in the resource usage amount, checks whether a remaining resource of a second target node associated with the target virtual package is sufficient; and in response to the remaining resource being insufficient, the scheduler assigns a resource insufficient identifier to the target virtual package.

[0181] In one or more embodiments, the scheduling instruction includes creating the service container; and when scheduling the service container based on the scheduling instruction, the processor executing the computer program further implements the following steps: the container processing engine, in response to detecting that the virtual container package carries a new identifier, resolves a resource usage amount indicated by the new identifier; the container processing engine creates a new service container based on the resolved resource usage amount; and a network unit connects network spaces of the new service container and the container management platform.

[0182] In one or more embodiments, the scheduling instruction comprises deleting the service container; when the processor executes the computer program based on the scheduling instruction of the service container, the following steps are further implemented: the container processing engine identifies a fourth target container pointed by the deletion identifier in response to detecting that the virtual container package carries the deletion identifier; the fourth target container is unbound from the extended network, and a second virtual network card pair associated with the fourth target container is deleted; the container processing engine deletes the proxy information associated with the fourth target container in the configuration file, and triggers the proxy container package to reload the configuration file; and the container processing engine deletes the fourth target container in the node of the cluster.

[0183] In one or more embodiments, when the processor executes the computer program based on the scheduling instruction of the service container, the following steps are further implemented: the container processing engine identifies the scheduling instruction; the container processing engine suspends a first target container pointed by the scheduling instruction in response to the scheduling instruction being a suspension instruction; the container processing engine starts a second target container pointed by the scheduling instruction in response to the scheduling instruction being a start instruction; the container processing engine performs resource usage updating on a third target container pointed by the scheduling instruction in response to the scheduling instruction being an updating instruction; the container processing engine deletes a fourth target container pointed by the scheduling instruction in response to the scheduling instruction being a deletion instruction; and the container processing engine creates a new service container based on a new instruction in response to the scheduling instruction being the new instruction.

[0184] In one or more embodiments, when the processor executes the computer program, the following steps are further implemented: the container processing engine changes the service container in the cluster; the container processing engine modifies the service information in the configuration file based on the change; and the proxy container package loads the updated configuration file in response to detecting that the service information changes.

[0185] In one or more embodiments, when the container processing engine modifies the service information in the configuration file based on the change, the processor executes the computer program, and the following steps are further implemented: the container processing engine creates a new service container or adds a new service to the service container; the container processing engine obtains the configuration file and writes first information into the configuration file; wherein the first information comprises service information of the newly added service container or service; and / or the container processing engine deletes an original service container or a service in the original service container, and takes the original service container as a sixth target container; and the container processing engine deletes second information; wherein the second information comprises service information of the deleted original service container or service.

[0186] In one or more embodiments, when the proxy container package loads the updated configuration file, the processor executes the computer program, and the following steps are further implemented: the proxy container package traverses process information of its process namespace, and screens a master process number of the reverse proxy service in the process information as a first master process number; and the proxy container package sends a loading instruction to the first master process number, and triggers the reverse proxy service to reload the configuration file.

[0187] In one or more embodiments, the proxy container package is created by the container management platform and housed in the cluster, and the processor executing the computer program further implements the following steps: the container management platform creates a reverse proxy service and the proxy container package, encapsulates the reverse proxy service in the proxy container package, and deploys the proxy container package to the cluster; the container management platform creates a configuration dictionary, maps the configuration dictionary to the proxy container package, and forms a configuration file; wherein the configuration dictionary is used to record service information that needs to be proxied.

[0188] In one or more embodiments, the proxy container package includes a proxy service container; when the service container in the proxy cluster carries a service, the processor executing the computer program further implements the following steps: the container management platform obtains a service request and sends the service request to the proxy container package; the proxy service container receives the service request and forwards the service request to the corresponding service container to enable the service container to respond to the service request.

[0189] In one or more embodiments, the proxy container package further includes a configuration detection container; the processor executing the computer program further implements the following steps: the configuration detection container detects whether there is information change in the configuration file; the configuration detection container, in response to the information change in the configuration file, screens a main process number of the proxy service container as a second main process number; sends a load instruction to the proxy service container based on the second main process number; the proxy service container receives the load instruction and restarts the reverse proxy service to reload the updated configuration file.

[0190] In one or more embodiments, when the extended network is constructed, the processor executing the computer program further implements the following steps: the network unit creates a virtual multicast group and a bridge in each node in the cluster; the network unit connects the virtual device of the virtual multicast group to the bridge; the network unit starts the bridge and the virtual device connected to the bridge to form the extended network; and the network unit plans the network segment information of the extended network.

[0191] In one or more embodiments, one of the first pair of virtual network cards is inserted into the extended network and the other is inserted into the proxy container package, and the processor executing the computer program further implements the following steps: querying the bridge and the network segment information of the extended network, configuring the network segment information as the Internet Protocol address of one of the first pair of virtual network cards, and inserting it into the bridge; obtaining the process number of the proxy container package, obtaining its network namespace based on the process number, and moving the other of the first pair of virtual network cards to the network namespace of the proxy container package.

[0192] In one or more embodiments, when one of the second pair of virtual network cards is plugged into the stretch network and the other is plugged into a service container located in the container processing engine within the cluster, the processor executing the computer program further implements the following steps: querying the bridge of the stretch network and its network segment information, configuring the network segment information as the Internet Protocol address of one of the second pair of virtual network cards, and plugging into the bridge; moving the other of the second pair of virtual network cards to the network namespace where the service container is located.

[0193] In one or more embodiments, a non-transitory computer-readable storage medium is provided, having stored thereon a computer program, which, when executed by a processor, implements the following steps:

[0194] S101: The network unit constructs a stretch network.

[0195] S102: The network unit creates a first pair of virtual network cards, plugs one of the first pair of virtual network cards into the stretch network, and plugs the other into a proxy container package created by the container management platform and located in the cluster, and connects the network space of the container management platform and the stretch network; wherein the proxy container package is created by the container management platform and located in the cluster, and is used to proxy the service carried by the service container in the cluster.

[0196] S103: The network unit creates a second pair of virtual network cards, plugs one of the second pair of virtual network cards into the stretch network, and plugs the other into the service container located in the container processing engine within the cluster, connects the network space of the service container and the stretch network, and connects the network space of the container management platform and the container processing engine.

[0197] S104: The container management platform sends a scheduling instruction; wherein the scheduling instruction includes pausing the service container; the container processing engine acquires the scheduling instruction and schedules the service container based on the scheduling instruction.

[0198] In one or more embodiments, the computer program, when executed by the processor, further implements the following steps:

[0199] In one or more embodiments, the container management platform sends a scheduling instruction; when the container processing engine acquires the scheduling instruction, the computer program, when executed by the processor, further implements the following steps: the container management platform assigns a scheduling identifier to the virtual container package located in the cluster; the container processing engine detects the virtual container package and judges whether the virtual container package is assigned the scheduling identifier; the container processing engine, in response to the virtual container package being assigned the scheduling identifier, identifies the service container pointed to by the scheduling identifier as a target container and schedules the target container according to the scheduling identifier.

[0200] In one or more embodiments, the container management platform assigns a scheduling identifier to a virtual container package located in the cluster; when the container processing engine detects the virtual container package, the computer program executed by the processor further implements the following steps: the container management platform initiates a scheduling request to the scheduler; the scheduler acquires the virtual container package matching the scheduling request, and injects the scheduling identifier into the annotation of the virtual container package; the container processing engine detects the annotation of the virtual container package.

[0201] In one or more embodiments, when the computer program executed by the processor further implements the following steps: the container management platform acquires a creation request, and creates a new virtual container package based on the creation request; the scheduler identifies the new virtual container package, and reserves resources for the new virtual container package; the scheduler screens nodes in the cluster, and takes the screened nodes as first target nodes associated with the new virtual container package; and the node information of the first target nodes is injected into the annotation of the new virtual container package.

[0202] In one or more embodiments, after the service container is scheduled based on the scheduling instruction, the computer program executed by the processor further implements the following steps: the container processing engine acquires a configuration file in response to completing the execution of the scheduling instruction; the container processing engine updates the configuration file based on the scheduling instruction, triggers the container management platform to reload the updated configuration file; and the container management platform proxies the services carried by the service container in the cluster based on the updated configuration file.

[0203] In one or more embodiments, when the container processing engine acquires the scheduling instruction and schedules the service container based on the scheduling instruction, the computer program executed by the processor further implements the following steps: the container processing engine detects a virtual container package located in the cluster; the container processing engine analyzes a first target container pointed by a suspension identifier in response to the virtual container package carrying the suspension identifier; the container processing engine unbinds one of the target network card pair from the bridge of the extended network, and migrates the other of the target network card pair from the network space of the first target container to the default network space, and deletes the target network card pair; wherein the target network card pair is a second virtual network card pair connecting the first target container and the extended network.

[0204] In one or more embodiments, the scheduling instruction includes starting the service container; the container management platform sends the scheduling instruction to the container processing engine; when the container processing engine receives the scheduling instruction, the computer program executed by the processor further implements the following steps: the container management platform injects a start identifier into the annotation of the virtual container package associated with the second target container; and the container processing engine starts the second target container in response to detecting the start identifier, and connects the network spaces of the second target container and the container management platform.

[0205] In one or more embodiments, the scheduling instruction comprises updating the resource of the service container; when the service container is scheduled based on the scheduling instruction, the computer program is further implemented by the processor to perform the following steps: the scheduler identifies the target virtual package as the target virtual package carrying the to-be-updated identifier is detected; the scheduler updates the resource usage of the target virtual package recorded therein and assigns the target virtual package with the updated identifier; the container processing engine updates the updated resource data of the third target container to the resource information recorded in the target virtual package in response to detecting the updated identifier.

[0206] In one or more embodiments, when the scheduler updates the resource usage of the target virtual package recorded therein, the computer program is further implemented by the processor to perform the following steps: the scheduler identifies the update target of the resource update; the scheduler checks whether the remaining resource of the second target node associated with the target virtual package is sufficient in response to the update target being the increase of the resource usage; and the target virtual package is assigned with the resource insufficient identifier in response to the remaining resource being insufficient.

[0207] In one or more embodiments, the scheduling instruction comprises creating the service container; when the service container is scheduled based on the scheduling instruction, the computer program is further implemented by the processor to perform the following steps: the container processing engine identifies the resource usage indicated by the new identifier in response to the virtual container package carrying the new identifier; the container processing engine creates a new service container based on the parsed resource usage; and the network unit connects the network space of the new service container and the container management platform.

[0208] In one or more embodiments, the scheduling instruction comprises deleting the service container; when the service container is scheduled based on the scheduling instruction, the computer program is further implemented by the processor to perform the following steps: the container processing engine identifies the fourth target container pointed by the deletion identifier in response to the virtual container package carrying the deletion identifier; the fourth target container is unbound from the extended network, and the second virtual network card pair associated with the fourth target container is deleted; the container processing engine deletes the proxy information associated with the fourth target container in the configuration file, triggers the proxy container package to reload the configuration file; and the container processing engine deletes the fourth target container in the node of the cluster.

[0209] In one or more embodiments, when the computer program is executed by the processor, the following steps are further implemented: the container processing engine identifies the scheduling instruction; the container processing engine pauses a first target container pointed by the scheduling instruction in response to the scheduling instruction being a pause instruction; the container processing engine starts a second target container pointed by the scheduling instruction in response to the scheduling instruction being a start instruction; the container processing engine updates resource usage of a third target container pointed by the scheduling instruction in response to the scheduling instruction being an update instruction; the container processing engine deletes a fourth target container pointed by the scheduling instruction in response to the scheduling instruction being a delete instruction; and the container processing engine creates a new service container based on a new instruction in response to the scheduling instruction being the new instruction.

[0210] In one or more embodiments, when the computer program is executed by the processor, the following steps are further implemented: the container processing engine changes the service containers in the cluster; the container processing engine modifies the service information in the configuration file based on the change; and the proxy container package loads the updated configuration file in response to detecting a change in the service information.

[0211] In one or more embodiments, when the container processing engine modifies the service information in the configuration file based on the change, the computer program executed by the processor further implements the following steps: the container processing engine creates a new service container or adds a new service to the service container; the container processing engine obtains the configuration file and writes first information to the configuration file; wherein the first information includes service information of the newly added service container or service; and / or the container processing engine deletes an original service container or a service in the original service container and takes the original service container as a sixth target container; and the container processing engine deletes second information; wherein the second information includes service information of the deleted original service container or service.

[0212] In one or more embodiments, when the proxy container package loads the updated configuration file, the computer program executed by the processor further implements the following steps: the proxy container package traverses process information of processes in a process namespace of the proxy container package, and filters a master process number of the reverse proxy service in the process information as a first master process number; and the proxy container package sends a loading instruction to the first master process number to trigger the reverse proxy service to reload the configuration file.

[0213] In one or more embodiments, the proxy container package is created by the container management platform and is accommodated in the cluster, and the computer program executed by the processor further implements the following steps: the container management platform creates the reverse proxy service and the proxy container package, encapsulates the reverse proxy service in the proxy container package, and deploys the proxy container package to the cluster; and the container management platform creates a configuration dictionary and maps the configuration dictionary to the proxy container package to form the configuration file; wherein the configuration dictionary is used to record service information that needs to be proxied.

[0214] In one or more embodiments, the proxy container package comprises a proxy service container; when the service container in the proxy cluster carries a service, the computer program, when executed by the processor, further implements the following steps: the container management platform obtains a service request, and sends the service request to the proxy container package; the proxy service container receives the service request, forwards the service request to the corresponding service container, and enables the service container to respond to the service request.

[0215] In one or more embodiments, the proxy container package further comprises a configuration detection container; the computer program, when executed by the processor, further implements the following steps: the configuration detection container detects whether there is information change in the configuration file; the configuration detection container, in response to the information change in the configuration file, screens a main process number of the proxy service container as a second main process number; sends a loading instruction to the proxy service container based on the second main process number; the proxy service container receives the loading instruction, restarts the reverse proxy service, and reloads the updated configuration file.

[0216] In one or more embodiments, when the extended network is constructed, the computer program, when executed by the processor, further implements the following steps: the network unit creates a virtual multicast group and a bridge in each node in the cluster; the network unit connects the virtual device of the virtual multicast group to the bridge; the network unit starts the bridge and the virtual device connected to the bridge to form the extended network; and the network unit plans the network segment information of the extended network.

[0217] In one or more embodiments, when one of the first pair of virtual network cards is inserted into the extended network and the other is inserted into the proxy container package, the computer program, when executed by the processor, further implements the following steps: querying the bridge and the network segment information of the extended network, configuring the network segment information as the Internet Protocol address of one of the first pair of virtual network cards, and inserting it into the bridge; obtaining the process number of the proxy container package, obtaining its network namespace based on the process number, and moving the other of the first pair of virtual network cards to the network namespace of the proxy container package.

[0218] In one or more embodiments, when one of the second pair of virtual network cards is inserted into the extended network and the other is inserted into the service container located in the container processing engine in the cluster, the computer program, when executed by the processor, further implements the following steps: querying the bridge and the network segment information of the extended network, configuring the network segment information as the Internet Protocol address of one of the second pair of virtual network cards, and inserting it into the bridge; and moving the other of the second pair of virtual network cards to the network namespace where the service container is located.

[0219] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by a computer program instructing relevant hardware, and the computer program can be stored in a non-volatile computer readable storage medium. When the computer program is executed, the processes of the above-mentioned embodiments of each method can be included. Any reference to memory, storage, database or other medium used in each embodiment provided by the present application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration but not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0220] Any combination of the technical features of the above embodiments can be made, and in order to make the description simple, all possible combinations of each technical feature in the above embodiments are not described, however, as long as the combination of the technical features does not exist, it should be considered as the scope of the present application.

[0221] The above embodiments only express several implementation manners of the present application, and the description is more specific and detailed, but it should not be understood as a limitation on the scope of the patent. It should be pointed out that for ordinary skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are within the scope of the present application.

Claims

1. A management method of a service container, characterized by, The management method of the service container is applied to a management system of the service container, and the management system comprises a network element, a container management platform, and a container processing engine. The management method comprises: The network element constructs an extension network; The network element creates a first virtual network card pair, inserts a first card in the first virtual network card pair into the extension network, and inserts a second card into a proxy container package, and connects the network space of the container management platform and the extension network; wherein the proxy container package is created by the container management platform and is accommodated in a cluster, and is used for proxying services carried by the service container in the cluster; The network element creates a second virtual network card pair, inserts a first card in the second virtual network card pair into the extension network, and inserts a second card into the cluster and connects the service container located in the container processing engine, connects the network space of the service container and the extension network, and connects the network space of the container management platform and the network space of the container processing engine; The container management platform sends a scheduling instruction; wherein the scheduling instruction comprises suspending the service container; and The container processing engine receives the scheduling instruction and schedules the service container according to the scheduling instruction.

2. The management method according to claim 1, characterized in that, The container management platform sends a scheduling instruction; and the container processing engine receives the scheduling instruction, comprising: The container management platform configures a scheduling identifier to a virtual container package located in a cluster; The container processing engine detects the virtual container package, judges whether the virtual container package is endowed with the scheduling identifier; and The container processing engine responds to the virtual container package being configured with the scheduling identifier, identifies a service container pointed by the scheduling identifier and takes the service container as a target container, and schedules the target container according to the scheduling identifier.

3. The management method according to claim 2, characterized in that, The container management platform configures a scheduling identifier to a virtual container package located in a cluster; The container processing engine detects the virtual container package, comprising: The container management platform initiates a scheduling request to a scheduler; The scheduler acquires a virtual container package matched with the scheduling request, inserts the scheduling identifier into an annotation of the virtual container package; and The container processing engine detects the annotation of the virtual container package.

4. The management method according to claim 2, characterized by, The management method further comprises: The container management platform acquires a creation request, and creates a new virtual container package according to the creation request; The scheduler identifies the new virtual container package, and reserves resources for the new virtual container package; The scheduler screens nodes in a cluster, and takes the screened nodes as first target nodes associated with the new virtual container package; and Inserts node information of the first target nodes into the annotation of the new virtual container package.

5. The management method according to claim 1, characterized by, After the service container is scheduled based on the scheduling instruction, the method further comprises: The container processing engine acquires a configuration file in response to the scheduling instruction being executed; and The container processing engine acquires a configuration file in response to the scheduling instruction being executed; and The container processing engine updates the configuration file according to the scheduling instruction, triggers the container management platform to reload the updated configuration file, so that the container management platform proxies services carried by the service containers in the cluster according to the updated configuration file.

6. The management method according to claim 1, characterized by, The container processing engine receives the scheduling instruction and schedules the service container according to the scheduling instruction, including: The container processing engine detects a virtual container package configured in the cluster; The container processing engine analyzes the suspension identifier to obtain a first target container pointed by the suspension identifier in response to the virtual container package carrying the suspension identifier; The container processing engine unbinds a first card of a target network card pair from a bridge of the extension network, migrates a second card of the target network card pair from a network space of the first target container to a default network space, and deletes the target network card pair; and The target network card pair connects the first target container and the second virtual network card pair of the extension network.

7. The management method according to claim 1, characterized by, The scheduling instruction includes starting the service container; and the container management platform sends the scheduling instruction to the container processing engine; The container processing engine receives the scheduling instruction, including: The container management platform inserts a start identifier into an annotation of a virtual container package associated with a second target container; And The container processing engine starts the second target container in response to detecting the start identifier, and connects a network space of the second target container and a network space of the container management platform.

8. The management method according to claim 1, characterized by, The scheduling instruction includes updating resources of the service container; and the scheduling of the service container according to the scheduling instruction includes: The scheduler takes the virtual container package as a target virtual package in response to detecting that the virtual container package carries a to-be-updated identifier; The scheduler updates the recorded resource usage of the target virtual package and configures an updated identifier for the target virtual package; and The container processing engine updates the updated resource data of the third target container to the recorded resource information of the target virtual package in response to detecting the updated identifier.

9. The management method according to claim 8, characterized in that, The scheduler updates the recorded resource usage of the target virtual package, including: The scheduler identifies an update target of resource update; The scheduler checks whether the remaining resources of a second target node associated with the target virtual package are sufficient in response to the update target being an increase in resource usage; and In response to the remaining resources being insufficient, the scheduler configures an insufficient resource identifier for the target virtual package.

10. The management method according to claim 1, characterized by, The scheduling instruction includes creating the service container; and the scheduling of the service container according to the scheduling instruction includes: The container processing engine analyzes the new identifier to obtain resource usage indicated by the new identifier in response to detecting that the virtual container package carries the new identifier; The container processing engine creates a new service container according to the analyzed resource usage; and The network unit connects a network space of the new service container and a network space of the container management platform.

11. The management method according to claim 1, characterized by, The scheduling instruction includes deleting the service container; and the scheduling of the service container according to the scheduling instruction includes: The container processing engine identifies the fourth target container to which the deletion identifier points in response to detecting that the virtual container package carries the deletion identifier; unbinds the fourth target container from the extension network, and deletes a second virtual network card pair associated with the fourth target container; The container processing engine deletes proxy information associated with the fourth target container in a configuration file, and triggers the proxy container package to reload the configuration file; and The container processing engine deletes the fourth target container in a node of the cluster.

12. The management method according to any one of claims 1 to 11, characterized by, The scheduling of the service container according to the scheduling instruction comprises: The container processing engine identifies the scheduling instruction; The container processing engine suspends a first target container to which the scheduling instruction points in response to the scheduling instruction being a suspension instruction; starts a second target container to which the scheduling instruction points in response to the scheduling instruction being a start instruction; performs resource usage updating on a third target container to which the scheduling instruction points in response to the scheduling instruction being an updating instruction; deletes a fourth target container to which the scheduling instruction points in response to the scheduling instruction being a deletion instruction; and creates a new service container based on a new creation instruction in response to the scheduling instruction being the new creation instruction.

13. The management method according to claim 1, characterized by, The management method further comprises: The container processing engine responds to a change in a service container in the cluster; The container processing engine modifies service information in a configuration file according to the change; and The proxy container package loads the updated configuration file in response to detecting that the service information changes.

14. The management method according to claim 13, characterized by, The container processing engine modifies service information in a configuration file according to the change, comprising: The container processing engine creates a new service container or adds a new service to the service container; The container processing engine obtains a configuration file, and writes first information into the configuration file; wherein the first information comprises service information of the newly added service container or service; and / or, The container processing engine deletes an original service container or a service in the original service container, and takes the original service container as a sixth target container; and The container processing engine deletes second information; wherein the second information comprises service information of the deleted original service container or service.

15. The management method according to claim 13, characterized by, The loading of the updated configuration file comprises: The proxy container package traverses process information of a process namespace, and screens a master process number of a reverse proxy service in the process information as a first master process number; The proxy container package sends a loading instruction to the first master process number, and triggers the reverse proxy service to reload the configuration file.

16. The management method according to claim 13, wherein The proxy container package is created by the container management platform and accommodated in the cluster, comprising: The container management platform creates a reverse proxy service and a proxy container package, encapsulates the reverse proxy service in the proxy container package, and deploys the proxy container package to the cluster; and The container management platform creates a configuration dictionary, maps the configuration dictionary to the proxy container package, and forms the configuration file; wherein the configuration dictionary is used to record service information that needs to be proxied.

17. The management method according to claim 1, characterized by, The proxy container package comprises a proxy service container; the proxy service container encapsulates services carried by the business container in the cluster, comprising: The container management platform acquires a service request and sends the service request to the proxy container package; and The proxy service container receives the service request and forwards the service request to the corresponding business container to enable the business container to respond to the service request.

18. The management method according to claim 16, wherein, The proxy container package further comprises a configuration detection container; the management method further comprises: The configuration detection container detects whether there is information change in the configuration file; The configuration detection container, in response to the information change in the configuration file, screens a main process number of the proxy service container as a second main process number, sends a loading instruction to the proxy service container according to the second main process number; and The proxy service container receives the loading instruction and restarts the reverse proxy service to reload the updated configuration file.

19. The management method according to claim 1, characterized by, The construction of the extension network comprises: The network unit creates a virtual multicast group and a bridge in each node in the cluster; The network unit connects the virtual device of the virtual multicast group to the bridge; The network unit starts the bridge and the virtual device connected to the bridge to form the extension network; and The network unit configures the network segment information of the extension network.

20. The management method according to claim 1, characterized by, The insertion of the first card in the first virtual network card pair into the extension network and the second card into the proxy container package comprises: Obtaining the bridge of the extension network and the network segment information of the bridge, configuring the network segment information as the Internet Protocol address of the first card in the first virtual network card pair, and inserting the first card into the bridge; and Obtaining the process number of the proxy container package and obtaining the network namespace of the proxy container package according to the process number, and moving the second card in the first virtual network card pair to the network namespace of the proxy container package.

21. The management method according to claim 1, characterized by, The insertion of the first card in the first virtual network card pair into the extension network and the second card into the business container located in the container processing engine in the cluster comprises: Obtaining the bridge of the extension network and the network segment information of the bridge, configuring the network segment information as the Internet Protocol address of the first card in the second virtual network card pair, and inserting the first card into the bridge; and Moving the second card in the second virtual network card pair to the network namespace where the business container is located.

22. A management system of a service container, characterized by, The management system comprises: A container management platform configured to send a scheduling instruction; wherein the scheduling instruction comprises suspending the business container; and A container processing engine configured to acquire the scheduling instruction and schedule the business container based on the scheduling instruction. The network unit is configured to build an extension network, create a first virtual network card pair, plug a first card in the first virtual network card pair into the extension network, plug a second card into a proxy container package, and connect network spaces of the container management platform and the extension network; the proxy container package is created by the container management platform and accommodated in a cluster, and is configured to proxy services carried by a service container in the cluster; create a second virtual network card pair, plug a first card in the second virtual network card pair into the extension network, plug a second card into the cluster and into the service container located in the container processing engine, connect network spaces of the service container and the extension network, and connect network spaces of the container management platform and the container processing engine.

23. The management system of claim 22, wherein, The management system further comprises a scheduler, and the cluster is provided with a virtual container package; the scheduler is connected to the container management platform and the virtual container package. The scheduler is configured to obtain a virtual container package matched with a scheduling request, insert a scheduling identifier into an annotation of the virtual container package, and enable the container processing engine to detect the annotation of the virtual container package. The proxy service container is configured to receive a service request sent by the container management platform to the proxy container package, forward the service request to a corresponding service container, and enable the service container to respond to the service request.

24. The management system of claim 22, wherein, The proxy container further comprises a configuration detection container and a configuration file. The configuration detection container is configured to detect whether information in the configuration file changes, in response to the information in the configuration file changing, filter a main process number of the proxy service container as a second main process number, and send a loading instruction to the proxy service container according to the second main process number.

25. The management system of claim 24, wherein, The proxy service container is configured to receive the loading instruction, restart a reverse proxy service, and reload an updated configuration file. The cluster comprises a plurality of nodes. The nodes are created by the container management platform or the container processing engine.

26. The management system of claim 22, wherein, The nodes are provided with service containers and / or proxy container packages. The processor executes the computer program to implement the steps of the management method of the service container in any one of claims 1 to 21. The computer program is executed by the processor to implement the steps of the management method of the service container in any one of claims 1 to 21.

27. A computer device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the computer program comprises the steps of: The computer program is executed by the processor to implement the steps of the management method of the service container in any one of claims 1 to 21. 28.A non-transitory computer-readable storage medium having stored thereon a computer program, characterized in that, ​ 29. A computer program product comprising a computer program, characterized in that, ​

Images