Scheduling method, scheduling system, and container cluster

By coordinating the monitoring and scheduling components within the container cluster, network fault detection and migration of container groups are achieved, resolving the issue of service unavailability in the container cluster and improving the stability and disaster recovery capabilities of the container cluster.

WO2024230294A1PCT designated stage expired Publication Date: 2024-11-14HANGZHOU ALICLOUD FEITIAN INFORMATION TECH CO LTD

Patent Information

Application Number
PCT/CN2024/080636
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2023-05-06
Filing Date
2024-03-07
Publication Date
2024-11-14

AI Technical Summary

Technical Problem

In existing technologies, when a container cluster fails, service availability cannot be guaranteed even after restarting the container cluster, resulting in poor disaster recovery capabilities of the container cluster.

Method used

By setting up monitoring and scheduling components in the container cluster, the network status on the nodes can be monitored, container groups with network failures can be identified, and they can be scheduled to work nodes with normal network conditions, thus enabling timely migration of container groups.

Benefits of technology

It improves the stability and disaster recovery capabilities of container clusters, reduces project interruptions caused by network issues, and provides more stable service capabilities.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN2024080636_14112024_PF_FP_ABST
    Figure CN2024080636_14112024_PF_FP_ABST
Patent Text Reader

Abstract

Embodiments of the present description provide a scheduling method, a scheduling system, and a container cluster. The scheduling system comprises monitoring components and a scheduling component; the monitoring components are configured at working nodes in a container cluster, and the scheduling component is configured at a control node in the container cluster. Each monitoring component is used for monitoring a network over which pods on the working node where the monitoring component is located communicates with the outside to obtain network state information of the working node; and determining a pod to be scheduled having a network fault on the working node according to the network state information, and sending to the scheduling component pod information of the pod to be scheduled. The scheduling component is used for scheduling, according to the pod information sent by the monitoring component on each working node, the pod to be scheduled to the working node having a normal network. A network fault in the container cluster is accurately found by means of the monitoring component, and after the network fault occurs, the pod to be scheduled is scheduled by the scheduling component to the working node having the normal network in a timely manner, thereby improving the stability and disaster recovery capability of the container cluster.
Need to check novelty before this filing date? Find Prior Art

Description

Scheduling method, scheduling system, and container cluster

[0001] This application claims priority to the Chinese patent application filed with the China Patent Office on May 6, 2023, with application number 202310512174.5 and invention name “Scheduling Method, Scheduling System and Container Cluster”, the entire contents of which are incorporated by reference into this application. Technical Field

[0002] This specification relates to the field of computer technology, and more particularly to a scheduling system. One or more embodiments of this specification also relate to a scheduling method, a container cluster, a computing device, a computer-readable storage medium, and a computer program. Background Art

[0003] With the development of cloud-native technologies, more and more applications are deployed in container clusters (K8S, Kubernetes) using containers to provide external services. When a pod in a container cluster fails, it can cause errors in the service processes within the pod, leading to unavailability of the project services on the container cluster.

[0004] Currently, when a service process within a container group encounters an error, the group is typically considered to have failed, and the node is restarted to re-run the group. However, even with this approach, restarting the container group still doesn't guarantee the availability of services within the group. Therefore, a solution that improves cluster disaster recovery capabilities is urgently needed.

[0005] Summary of the Invention

[0006] In view of this, embodiments of this specification provide a scheduling system. One or more embodiments of this specification also involve a scheduling method, a container cluster, a computing device, a computer-readable storage medium, and a computer program to address technical deficiencies in the prior art.

[0007] According to a first aspect of an embodiment of this specification, a scheduling system is provided, including: a monitoring component and a scheduling component, wherein the monitoring component is provided on each working node in a container cluster, and the scheduling component is provided on a control node in the container cluster; the monitoring component is configured to monitor the network in which the container group on the working node communicates with the outside world and obtain network status information of the working node; based on the network status information, determine a container group to be scheduled due to a network failure on the working node, and send the container group information of the container group to be scheduled to the scheduling component; and the scheduling component is configured to schedule the container group to be scheduled to a working node with a normal network based on the container group information sent by the monitoring component on each working node.

[0008] According to a second aspect of an embodiment of this specification, a scheduling method is provided, which is applied to a scheduling component, where the scheduling component is set at a control node in a container cluster. The method includes: receiving container group information of a container group to be scheduled sent by a monitoring component on each working node, wherein the container group to be scheduled is a container group with a network failure, and the container group information is obtained by the monitoring component based on network status information, where the network status information is obtained by the monitoring component monitoring the network in which the container group on the working node communicates with the outside world; and scheduling the container group to be scheduled to a working node with a normal network based on the container group information of the container group to be scheduled sent by the monitoring component on each working node.

[0009] According to a third aspect of the embodiments of this specification, a scheduling method is provided, which is applied to a monitoring component, where the monitoring component is set on each working node in a container cluster. The method includes: monitoring the network on which the container group on the working node communicates with the outside world to obtain network status information of the working node; determining, based on the network status information, a container group to be scheduled due to a network failure on the working node; and sending the container group information of the container group to be scheduled to a scheduling component in the container cluster, so that the scheduling component schedules the container group to be scheduled to a working node with a normal network according to the container group information sent by the monitoring component on each working node.

[0010] According to a fourth aspect of an embodiment of this specification, a container cluster is provided, including: a control node and multiple working nodes, wherein the control node is provided with a scheduling component, and the working nodes are provided with a monitoring component and a container group; the monitoring component is configured to determine a container group to be scheduled due to a network failure on the working node according to any scheduling method applied to the monitoring component provided in the embodiments of this specification; and the scheduling component is configured to schedule the container group to be scheduled to a working node with a normal network according to any scheduling method applied to the scheduling component provided in the embodiments of this specification.

[0011] According to a fifth aspect of the embodiments of this specification, a scheduling device is provided, which is applied to a scheduling component, where the scheduling component is provided at a control node in a container cluster. The device includes: a receiving module configured to receive container group information of a container group to be scheduled, sent by a monitoring component on each working node, wherein the container group to be scheduled is a container group with a network failure, and the container group information is obtained by the monitoring component based on network status information, which is obtained by the monitoring component monitoring the network communication between the container group on the working node where the container group is located and the outside world; and a scheduling module configured to schedule the container group to be scheduled to a working node with a normal network based on the container group information of the container group to be scheduled, sent by the monitoring component on each working node.

[0012] According to a sixth aspect of the embodiments of this specification, a scheduling device is provided, which is applied to a monitoring component, where the monitoring component is provided on each working node in a container cluster. The device includes: a monitoring module configured to monitor the network on which the container group on the working node communicates with the outside world, and obtain network status information of the working node; a determination module configured to determine, based on the network status information, a container group to be scheduled that has a network failure on the working node; and a sending module configured to send the container group information of the container group to be scheduled to the scheduling component in the container cluster, so that the scheduling component schedules the container group to be scheduled to a working node with a normal network according to the container group information sent by the monitoring component on each working node.

[0013] According to the seventh aspect of the embodiments of this specification, a computing device is provided, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions, which, when executed by the processor, implement the steps of the above-mentioned scheduling method.

[0014] According to an eighth aspect of the embodiments of this specification, a computer-readable storage medium is provided, which stores computer-executable instructions, and when the instructions are executed by a processor, the steps of the above-mentioned scheduling method are implemented.

[0015] According to a ninth aspect of the embodiments of this specification, a computer program is provided, wherein when the computer program is executed in a computer, the computer is caused to execute the steps of the above-mentioned scheduling method. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] FIG1 is an architecture diagram of a scheduling system provided by one embodiment of this specification;

[0017] FIG2 is a schematic diagram of a container cluster provided by one embodiment of this specification;

[0018] FIG3 is a flowchart of a scheduling method provided by one embodiment of this specification;

[0019] FIG4 is a flowchart of another scheduling method provided by an embodiment of this specification;

[0020] FIG5 is a flowchart of reporting network status information in a scheduling method provided in one embodiment of this specification;

[0021] FIG6 is a flowchart of a taint marking method in a scheduling method provided by an embodiment of this specification;

[0022] FIG7 is a flowchart of a processing process of a scheduling method provided by one embodiment of this specification;

[0023] FIG8 is a schematic structural diagram of a scheduling device provided by one embodiment of this specification;

[0024] FIG9 is a schematic structural diagram of another scheduling device provided by an embodiment of this specification;

[0025] FIG10 is a structural block diagram of a computing device provided in one embodiment of this specification. DETAILED DESCRIPTION

[0026] The following description sets forth many specific details to facilitate a thorough understanding of this specification. However, this specification can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the scope of this specification. Therefore, this specification is not limited to the specific implementations disclosed below.

[0027] The terms used in one or more embodiments of this specification are for the purpose of describing specific embodiments only and are not intended to limit one or more embodiments of this specification. The singular forms "a," "the," and "the" used in one or more embodiments of this specification and the appended claims are also intended to include plural forms unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used in one or more embodiments of this specification refers to and includes any or all possible combinations of one or more associated listed items.

[0028] It should be understood that although the terms first, second, etc. may be used to describe various information in one or more embodiments of this specification, such information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of one or more embodiments of this specification, the first may also be referred to as the second, and similarly, the second may also be referred to as the first. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0029] In addition, it should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in one or more embodiments of this specification are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0030] First, the terms involved in one or more embodiments of this specification are explained.

[0031] Container cluster: A container cluster is an open source system for automatically deploying, scaling, and managing containerized applications.

[0032] Container group: A container group (pod) is the basic scheduling unit of a container cluster. A container group represents a process running in a container cluster. It encapsulates one or more closely related containers. Therefore, containers run in the container cluster as container groups.

[0033] Virtual network: A virtual network (Overlay) is a virtual network built on an existing physical network with independent control and forwarding planes. The physical network is transparent to terminal devices connected to the virtual network, thus achieving the separation of the bearer network and services.

[0034] Physical network: The physical network (Underlay) refers to a network that uses a driver to directly expose the underlying network interface of the host machine to the container for use.

[0035] Cluster address: The cluster address (clusterip, cluster internet protocol) is the address assigned by the container cluster for the service proxy.

[0036] Node Port: A node port is a node port used for proxy services. Each node proxies this port (the same port number on each node) to the corresponding service.

[0037] Service Component: The service component (Kubelet) is responsible for the operational status of each node, ensuring that all containers on the node are running normally. The service component handles starting, stopping, and maintaining application containers as instructed by the control panel.

[0038] Network Proxy: Kube-proxy is a network proxy and load balancing implementation that supports service abstraction and other network operations. The network proxy can forward traffic to a specified container based on the address and port of the incoming request.

[0039] With the development of cloud-native technologies, more and more applications are deployed in container clusters using containers to provide external services. Container groups within a container cluster often require network connectivity and access from outside the cluster. The network capabilities of a container cluster depend on the physical network capabilities of the cluster machines, as well as the cluster's container network interface plug-ins, the cluster's components, and the machine's network configuration (such as routing and kernel parameters). A network failure on a worker node within a container cluster can directly cause errors in the service processes within the container group on that machine, leading to unavailability of services within the container cluster. Therefore, improving the stability of container cluster networks has become a research priority.

[0040] Currently, when a service process within a container group encounters an error, the container group is typically considered to have failed. This can be restarted on the worker node to re-run the container group. However, even with this solution, the service availability within the container group cannot be guaranteed after the container group is restarted, resulting in poor disaster recovery capabilities for the container cluster.

[0041] In one or more embodiments of this specification, the network communication status on each working node in the container cluster is abstracted, and the expulsion capability of the monitoring component in the container cluster and the scheduling capability of the scheduling component (kube-schdeluer) in the control node are combined. When a container cluster failure occurs, the container group is scheduled to a working node that can meet the normal network requirements of the container group based on the container group information of the container group, thereby reducing project interruptions caused by network problems, improving the disaster recovery capability within the container cluster and the stability and robustness of the projects in the container cluster, enabling the container cluster to provide more stable service capabilities, and further improving the overall service quality and user experience.

[0042] Specifically, an embodiment of this specification proposes a scheduling system, which includes a monitoring component and a scheduling component. The monitoring component is set on each working node in the container cluster, and the scheduling component is set on the control node in the container cluster; the monitoring component is used to monitor the network in which the container group on the working node communicates with the outside world, and obtain the network status information of the working node; based on the network status information, the container group to be scheduled due to network failure on the working node is determined, and the container group information of the container group to be scheduled is sent to the scheduling component; the scheduling component is used to schedule the container group to be scheduled to the working node with normal network according to the container group information sent by the monitoring component on each working node. By using the monitoring component to monitor the network in which the container group on the working node communicates with the outside world, the container group to be scheduled due to network failure on the working node is determined, the network failure in the container cluster is accurately discovered, and after the network failure occurs, the scheduling component is used to promptly schedule the container group to be scheduled to the working node with normal network, thereby improving the stability and disaster recovery capability of the container cluster.

[0043] In this specification, a scheduling system is provided. This specification also involves a scheduling method, a container cluster, a computing device, a computer-readable storage medium, and a computer program, which are described in detail one by one in the following embodiments.

[0044] Referring to FIG. 1 , FIG. 1 shows an architecture diagram of a scheduling system provided by one embodiment of this specification. The scheduling system may include a monitoring component 102 and a scheduling component 104 . The monitoring component 102 is set on each working node in a container cluster, and the scheduling component 104 is set on a control node in the container cluster.

[0045] Monitoring component 102 is used to monitor the network in which the container group on the working node communicates with the outside world and obtain network status information of the working node. Based on the network status information, it determines the container group to be scheduled due to network failure on the working node and sends the container group information of the container group to be scheduled to scheduling component 104.

[0046] The scheduling component 104 is configured to schedule the container group to be scheduled to a working node with a normal network according to the container group information sent by the monitoring component 102 on each working node.

[0047] By applying the solution of the embodiments of this specification, a monitoring component is used to monitor the network in which the container group on the working node communicates with the outside world, and the container group to be scheduled due to network failure on the working node is determined. Network failures in the container cluster are accurately discovered. Moreover, after a network failure occurs, the scheduling component is used to promptly schedule the container group to be scheduled to a working node with normal network, thereby improving the stability and disaster recovery capabilities of the container cluster.

[0048] 2 , which shows a schematic diagram of a container cluster provided by one embodiment of this specification. The container cluster includes a control node 202 and multiple worker nodes 204 . The control node 202 is provided with a scheduling component 2022 , and the worker nodes 204 are provided with a monitoring component 2042 and a container group 2044 .

[0049] The monitoring component 2042 is configured to determine a container group to be scheduled due to a network failure on the working node 204 according to any scheduling method applied to the monitoring component provided in the embodiments of this specification;

[0050] The scheduling component 2022 is configured to schedule the container group to be scheduled to a working node with a normal network according to any scheduling method applied to the scheduling component provided in the embodiments of this specification.

[0051] It's important to note that the control node manages and controls the entire container cluster; worker nodes are workload nodes, assigned workloads by the control node. If a worker node fails, the control node automatically transfers its workload to other worker nodes. A worker node can run one or more container groups.

[0052] By applying the solution of the embodiments of this specification, the monitoring component monitors the network between the container group on the working node and the outside world, and determines the container group to be scheduled due to network failure on the working node. This can accurately detect network failures in the container cluster. Moreover, after the network failure occurs, the scheduling component will promptly schedule the container group to be scheduled to the working node with normal network, thereby improving the stability and disaster recovery capabilities of the container cluster.

[0053] Referring to FIG3 , FIG3 shows a flowchart of a scheduling method provided by an embodiment of this specification. The scheduling method is applied to a scheduling component, which is set at a control node in a container cluster. Specifically, the scheduling method includes the following steps:

[0054] Step 302: Receive container group information of the container group to be scheduled, sent by the monitoring component on each working node. The container group to be scheduled is the container group with the network failure. The container group information is obtained by the monitoring component based on network status information. The network status information is obtained by the monitoring component monitoring the network communication between the container group on the working node and the outside world.

[0055] In one or more embodiments of this specification, the scheduling component in the control node can receive container group information of the container group to be scheduled from the monitoring component on each working node based on its scheduling capability, and schedule the container group to be scheduled to the working node with normal network.

[0056] Specifically, network status information is used to describe network communication conditions. Network status information can be connectivity information indicating whether the network is connected, or network bandwidth information. The specific selection depends on the actual situation and is not limited in this embodiment of the present specification. Container group information includes, but is not limited to, the container group identifier, the network communication requirements for the container group to communicate with the outside world, and the container group's taint tolerance information. The container group identifier is used to uniquely identify the container group, such as the container group number. The taint tolerance information indicates that the container group allows the scheduling component to schedule it to a working node carrying the corresponding taint label. For example, container group A only requires network communication with the virtual network and can tolerate other taints except virtual network failure taints. Container group A will only be scheduled to a working node if the working node is not tainted with a virtual network inaccessibility taint.

[0057] The network communication requirements for container groups to communicate with the outside world include, but are not limited to, network communication requirements between container groups, network communication requirements between container groups and nodes, network communication requirements between container groups and external networks, and network communication requirements between container groups and nodes and services within the container cluster via cluster and node ports. The networks used by container groups to communicate with the outside world can be divided into virtual networks (overlay) and physical networks (underlay), and these two networks often coexist in a container cluster.

[0058] A network failure refers to a network communication failure, which includes but is not limited to a network access path being unavailable, network bandwidth not meeting a preset bandwidth threshold, and no network status information being received within a preset time period. Network communication requirements correspond to different network access paths, and different network access paths depend on different network components within the container cluster. For example, access between virtual network container groups (overlay pods) may only rely on the node's physical network and the container cluster's Container Network Interface (CNI) plug-in, while accessing services through a cluster address (clusterip) may only rely on the node's physical network and the network proxy component (kube-proxy). The impact path and scope of failures in different network components vary. When a failure occurs in a part of the container cluster, causing a network path to be unavailable, the failure may be on certain nodes within the cluster, or on certain network paths on certain nodes.

[0059] Step 304: Schedule the container group to be scheduled to a working node with a normal network according to the container group information of the container group to be scheduled sent by the monitoring component on each working node.

[0060] In one or more embodiments of this specification, after receiving container group information of a container group to be scheduled from a monitoring component on each working node, the scheduling component may further schedule the container group to be scheduled to a working node with a normal network according to the container group information of the container group to be scheduled from the monitoring component on each working node.

[0061] Specifically, a normal network means that network communication is normal, and normal network communication includes but is not limited to the connectivity of a network access path and the network bandwidth meeting a preset bandwidth threshold.

[0062] It should be noted that the scheduling component can determine a working node with a normal network before scheduling the container group to be scheduled to a working node with a normal network based on the container group information of the container group to be scheduled sent by the monitoring component on each working node. There are many ways to determine a working node with a normal network, and the specific selection is based on the actual situation. The embodiments of this specification do not impose any restrictions on this. In the first possible implementation of this specification, the working node where all network access paths are connected can be regarded as a working node with a normal network. In the second possible implementation of this specification, the working node where the network access path corresponding to the network communication demand of the container group to be scheduled is connected can be regarded as a working node with a normal network. In the third possible implementation of this specification, the working node with a normal network can be determined based on the connectivity of the network access path and the preset bandwidth threshold.

[0063] By applying the solution of the embodiments of this specification, by utilizing the scheduling capabilities of the scheduling component, when a container cluster failure occurs, the container group is scheduled to a working node that can meet the normal network requirements of the container group based on the container group information of the container group, thereby reducing project interruptions caused by network problems, improving the disaster recovery capabilities within the container cluster and the stability and robustness of the projects in the container cluster, enabling the container cluster to provide more stable service capabilities, and further improving the overall service quality and user experience.

[0064] In an optional embodiment of the present specification, a working node with a normal network condition can be determined based on the taint label carried by each working node. That is, scheduling the container group to be scheduled to a working node with a normal network condition based on the container group information of the container group to be scheduled sent by the monitoring component on each working node can include the following steps:

[0065] Find working nodes in the container cluster that do not carry taint labels;

[0066] According to the container group information of the container group to be scheduled sent by the monitoring component on each working node, the container group to be scheduled is scheduled to the working node without the taint label.

[0067] Specifically, taints are used to describe container groups that are excluded by worker nodes in a container cluster. A taint label can indicate a network failure between a container group on a worker node and the outside world, for example, a taint label of "virtual network failure." A taint label can also indicate a container group that a worker node does not support. For example, a taint label of "virtual network container group access unavailable" indicates that the worker node corresponding to the taint label does not support virtual network container groups. If a worker node does not carry a taint label, it indicates that the worker node's network is normal and supports all container groups in the container cluster and all network access paths within the container cluster.

[0068] In actual applications, there are many ways to find working nodes that do not carry taint labels in a container cluster. The specific method is selected according to the actual situation. The embodiments of this specification do not impose any restrictions on this. In one possible implementation of this specification, each working node in the container cluster can be traversed to determine the working nodes that do not carry taint labels. In another possible implementation of this specification, each taint label can be determined from the taint label repository, and the working nodes corresponding to each taint label can be excluded from the container cluster to determine the working nodes that do not carry taint labels.

[0069] Furthermore, after finding a working node in the container cluster that does not carry a taint label, the container group to be scheduled with a network failure can be scheduled to a working node with a normal network that does not carry a taint label.

[0070] It should be noted that since the scheduling component in the control node can receive the container group information sent by the monitoring component on each working node in the container cluster, that is, there may be multiple container groups to be scheduled that the scheduling component needs to schedule. Therefore, when scheduling the container group to be scheduled to the working node that does not carry the taint label, the corresponding container group to be scheduled can be determined according to the container group information, avoiding scheduling the container group with normal network that has no scheduling requirements, and realizing accurate scheduling of the container group to be scheduled.

[0071] By applying the solution of the embodiments of this specification, a working node without a taint label is found in the container cluster. Based on the container group information of the container group to be scheduled sent by the monitoring component on each working node, the container group to be scheduled is scheduled to the working node without a taint label, thereby achieving accurate scheduling of the container group to be scheduled.

[0072] In another optional embodiment of the present specification, the container group information includes taint tolerance information of the container group to be scheduled. When a working node carries a taint label, the working node may only have certain network access paths that are faulty, while still having network access paths that are connected to the network. However, the container group has its own network communication requirements and taint tolerance information and may not use all network access paths when communicating with the outside world. Therefore, a working node with a normal network condition can be determined based on the taint tolerance information of the container group to be scheduled. That is, scheduling the container group to be scheduled to a working node with a normal network condition based on the container group information of the container group to be scheduled sent by the monitoring component on each working node may include the following steps:

[0073] Get the taint labels of the working nodes in the container cluster;

[0074] Compare the taint labels with the taint tolerance information, and use the working nodes corresponding to the taint labels that meet the taint tolerance information as normal working nodes of the network;

[0075] Schedule the container group to be scheduled to a working node with normal network connection.

[0076] Specifically, taint tolerance information indicates whether a container group allows the scheduling component to schedule it to a worker node carrying the corresponding taint label. This taint tolerance information can include both tolerable and intolerable taint labels. For example, if container group A only requires network communication with the virtual network, its taint tolerance information could be "tolerance for network failures other than virtual network failures" or "intolerance for virtual network failures." The specific representation of taint tolerance information depends on the specific situation and is not limited in this specification.

[0077] It should be noted that there are multiple ways to obtain the taint labels of the working nodes in the container cluster, and the specific selection is based on the actual situation. The embodiments of this specification do not impose any restrictions on this. In one possible implementation of this specification, the scheduling component can traverse each working node in the container cluster and sense whether each working node carries a taint label. If the working node carries a taint label, the taint label of the working node is directly obtained. In another possible implementation of this specification, the taint labels of each working node in the container cluster can be obtained directly from the taint label repository.

[0078] Furthermore, after obtaining the taint labels of each worker node in the container cluster, the taint labels of each worker node can be compared with the taint tolerance information of each container group to be scheduled, received by the scheduling component. For each container group to be scheduled, the scheduling component determines the worker node corresponding to the taint label that meets the taint tolerance information of the container group to be scheduled. This worker node is designated as the worker node with normal network connectivity corresponding to the container group to be scheduled, and the container group to be scheduled is then scheduled to the worker node with normal network connectivity.

[0079] For example, assume that a container cluster includes worker nodes 1, 2, and 3. Worker node 1 carries taint label 1, which indicates "virtual network failure," worker node 2 carries taint label 2, which indicates "physical network failure," and worker node 3 carries taint label 3, which indicates "virtual network failure." The scheduling component receives container group information 1 for container group 1 to be scheduled, sent by monitoring component 1 on worker node 1, which indicates "container group identifier 1, taint tolerance information 1 indicating intolerance to virtual network failures." Monitoring component 2 on worker node 2 receives container group information 2 for container group 2 to be scheduled, which indicates "container group identifier 2, taint tolerance information 2 indicating intolerance to physical network failures." After comparing taint labels 1, 2, and 3 with taint tolerance information 1, 2, and 3, the scheduling component determines that the taint label that satisfies taint tolerance information 1 is taint label 2, and the taint labels that satisfy taint tolerance information 2 are taint labels 1 and taint label 3. Therefore, container group 1 to be scheduled can be scheduled to worker node 2, and container group 2 to be scheduled can be scheduled to either worker node 1 or worker node 3.

[0080] It is worth noting that, referring to the above example, the three working nodes in the container cluster all have faulty network access paths. By comparing the taint labels and the taint tolerance information, the working nodes corresponding to the taint labels that meet the taint tolerance information are regarded as working nodes with normal network. This allows us to find the working nodes with normal network conditions for the scheduled containers among the above three working nodes, thereby improving the disaster recovery capability of the container cluster.

[0081] By applying the solution of the embodiments of this specification, the taint labels of the working nodes in the container cluster are obtained; the taint labels are compared with the taint tolerance information, and the working nodes corresponding to the taint labels that meet the taint tolerance information are used as working nodes with normal network conditions; the container group to be scheduled is scheduled to the working nodes with normal network conditions, thereby improving the disaster recovery capability of the container cluster.

[0082] Referring to FIG4 , FIG4 shows a flowchart of another scheduling method provided by an embodiment of this specification. The scheduling method is applied to a monitoring component, and the monitoring component is set on each working node in the container cluster. Specifically, the following steps are included:

[0083] Step 402: Monitor the network through which the container group on the working node communicates with the outside world, and obtain network status information of the working node.

[0084] In one or more embodiments of this specification, the monitoring component in each working node can monitor the network in which the container group on the working node communicates with the outside world based on its monitoring capability, and obtain network status information of the working node.

[0085] It should be noted that the monitoring component monitors the network used by the container group on its worker node to communicate with the outside world. When obtaining network status information for the worker node, it can use the Internet Control Message Protocol (ICMP) and Transmission Control Protocol (TCP) to detect whether it can access monitoring components on other worker nodes. Furthermore, while monitoring the network used by the container group on its worker node to communicate with the outside world, the monitoring component must also provide accessibility to meet the monitoring needs of monitoring components on other worker nodes.

[0086] In an optional embodiment of the present specification, the monitoring component includes a monitoring container group and a service component, and the monitoring container group is constructed based on a container network interface in the container cluster;

[0087] Monitor container groups, which monitor the connectivity of the network access path between the container group on the working node and the outside world, obtain network status information of the working node, and report the network status information to the service component;

[0088] The service component is used to determine the container group to be scheduled due to network failure on the working node based on the network status information, and send the container group information of the container group to be scheduled to the scheduling component.

[0089] It should be noted that the number of monitoring container groups can be one or more, depending on the actual situation, and this specification does not impose any restrictions on this. When building a monitoring container group on each working node in the container cluster, the communication network supported by the container network interface in the container cluster can be obtained, and the monitoring container group can be built on each working node in the container cluster based on the communication network supported by the container network interface.

[0090] For example, assume that container network interface 1 supports virtual networks, container network interface 2 supports physical networks, and container network interface 3 supports both virtual and physical networks. Therefore, a virtual network monitoring container group can be built in a worker node based on container network interface 1; a physical network monitoring container group can be built in a worker node based on container network interface 2; and a virtual network monitoring container group and a physical network monitoring container group can be built in a worker node based on container network interface 3.

[0091] Furthermore, when the monitoring container group reports network status information to the service component, the network status information can be reported to the service component in real time, and the network status information can also be reported to the service component periodically so as to adjust the taint label of the working node in time according to the network status information.

[0092] By applying the solution of the embodiments of this specification, at least one monitoring container group is constructed in each working node based on the container network interface in the container cluster. The connectivity of the network access path between the container group on the working node and the outside world is monitored by the monitoring container group, the network status information of the working node is obtained, and the network status information is reported to the service component. The service component is used to determine the container group to be scheduled due to the network failure on the working node based on the network status information, and the container group information of the container group to be scheduled is sent to the scheduling component, thereby achieving accurate monitoring of each working node. Further, when a container cluster failure occurs, the scheduling component schedules the container group to be scheduled to a working node that can meet the normal network requirements of the container group based on the container group information of the container group to be scheduled, thereby reducing project interruptions caused by network problems, improving the disaster recovery capability within the container cluster, and improving the stability and robustness of the projects in the container cluster, so that the container cluster provides more stable service capabilities.

[0093] Referring to FIG5 , FIG5 shows a flowchart for reporting network status information in a scheduling method provided in one embodiment of the present specification. As shown in FIG5 , the working node 1 includes a container group A1, a container group B2, and a service component. In the working node 1, a monitoring container group M and a monitoring container group N are constructed based on the container network interface in the container cluster to monitor the network access path. The monitoring container group M is a physical network monitoring container group, and the monitoring container group N is a virtual network monitoring container group. The monitoring container group M can monitor the connectivity of each container group on the working node 1 with the physical network for communicating with the outside world, obtain the network status information of the working node 1, and report the network status information to the service component; the monitoring container group N can monitor the connectivity of each container group on the working node 1 with the virtual network for communicating with the outside world, obtain the network status information of the working node 1, and report the network status information to the service component.

[0094] In an optional embodiment of the present specification, in order to facilitate the scheduling component to determine the working node with normal network status corresponding to the container group to be scheduled, after monitoring the network of the container group on the working node communicating with the outside world and obtaining the network status information of the working node, the following steps may be further included:

[0095] According to the network status information, a taint label is marked for the corresponding working node, wherein the taint label is used to mark the network where the container group on the working node fails to communicate with the outside world.

[0096] In actual applications, there are multiple ways to mark the taint labels for the corresponding working nodes according to the network status information. The specific selection is based on the actual situation, and the embodiments of this specification do not impose any limitation on this.

[0097] In one possible implementation of this specification, the network where the container group fails to communicate with the outside world can be directly used as the taint label of the working node. For example, if the virtual network used for communication between the container group and the outside world fails, the taint label can be "virtual network failure."

[0098] In another possible implementation of this specification, the working node can determine that the container group is not supported by the container group based on the network failure of the container group communicating with the outside world. The information that the container group is not supported by the working node and cannot be accessed is used as a taint label for the working node. For example, if the virtual network failure of the container group communicating with the outside world determines that the working node does not support the virtual network container group, the taint label is "Virtual network container group cannot be accessed."

[0099] For example, referring to FIG6 , FIG6 shows a flowchart of a taint marking method provided by an embodiment of the present specification. Work node 1 includes container group A1, container group B2, and a service component. In work node 1, a monitoring container group M and a monitoring container group N are constructed based on the container network interface in the container cluster, wherein monitoring container group M is a physical network monitoring container group and monitoring container group N is a virtual network monitoring container group. The connectivity of the network access path between container group A1 and container group B2 and the outside world is monitored by monitoring container group M and monitoring container group N respectively. Monitoring container group N determines that the network status information of work node 1 is "virtual network failure" and reports the network status information to the service component. Based on the network status information "virtual network failure", the service component marks the work node 1 with a taint label representing "virtual network failure".

[0100] By applying the solution of the embodiments of this specification, taint labels are marked for corresponding working nodes based on network status information, preparing for the subsequent scheduling component to schedule the container group to be scheduled, so that the scheduling component can determine the working node with normal network based on the taint label of each working node, and schedule the container group to be scheduled to the working node with normal network, reducing project interruptions caused by network problems, improving the disaster recovery capability within the container cluster and the project stability and robustness in the container cluster, so that the container cluster can provide more stable service capabilities.

[0101] Step 404: Determine the container group to be scheduled due to the network failure on the working node based on the network status information.

[0102] In one or more embodiments of this specification, a monitoring component in each working node monitors the network in which the container group on the working node communicates with the outside world. After obtaining network status information of the working node, the monitoring component can further determine, based on the network status information, a container group to be scheduled due to a network failure on the working node.

[0103] In actual applications, there are multiple ways to determine the container group to be scheduled due to network failure on the working node based on network status information. The specific method is selected according to the actual situation, and the embodiments of this specification do not impose any limitations on this.

[0104] In a possible implementation of this specification, after obtaining the network status information of the working node, the working node with the network failure can be determined, and all container groups on the working node with the network failure can be determined as the container groups to be scheduled due to the network failure.

[0105] In another possible implementation of this specification, after determining a network failure of a working node based on the network status information of the working node, since the working node may only have certain network access paths that fail, and still have network access paths that are connected to the network, and the container group has its own network communication requirements and taint tolerance information, the network access path of the network failure on the working node may not be used when communicating with the outside world. Therefore, the container group to be scheduled due to the network failure can be determined based on the network status information and the taint tolerance information of each container group on the working node. That is, the above-mentioned determination of the container group to be scheduled due to the network failure on the working node based on the network status information may include the following steps:

[0106] Get the taint tolerance information of the container group on the working node;

[0107] Compare network status information and taint tolerance information to determine the container group to be scheduled for network failure on the working node.

[0108] In actual applications, there are many ways to obtain the taint tolerance information of the container group on the working node. The specific method is selected according to the actual situation. The embodiments of this specification do not impose any limitations on this.

[0109] In a possible implementation of this specification, the taint tolerance information configured by the configuration personnel for each container group can be directly obtained.

[0110] In another possible implementation of this specification, the network communication requirements configured by the configuration personnel on the working node for each container group according to the project requirements can be obtained, and the taint tolerance information of each container group can be further determined based on the network communication requirements of each container group and the network access path supported by the container cluster.

[0111] For example, assuming that the network communication requirement of the container group is "virtual network", the network access paths supported by the container cluster include virtual network and physical network. According to the network communication requirement "virtual network" and the network access paths supported by the container cluster, the taint tolerance information of the container group can be determined as "tolerating physical network failures" or "tolerating network failures other than virtual network failures".

[0112] Furthermore, after obtaining the taint tolerance information of the container group on the working node, the network status information and the taint tolerance information can be compared to determine the container group to be scheduled due to the network failure on the working node.

[0113] For example, assuming that the network status information of the working node is "virtual network failure", the working node includes container group A and container group B, the taint tolerance information A of container group A is "tolerance of physical network failure", and the taint tolerance information B of container group B is "tolerance of virtual network failure", then by comparing the network status information "virtual network failure" with the taint tolerance information A "tolerance of physical network failure", it is determined that container group A is the container group to be scheduled for network failure; by comparing the network status information "virtual network failure" with the taint tolerance information B "tolerance of virtual network failure", it is determined that container group B is not the container group to be scheduled.

[0114] By applying the solution of the embodiments of this specification, the taint tolerance information of the container group on the working node is obtained; the network status information and the taint tolerance information are compared to determine the container group to be scheduled for the network failure on the working node. This eliminates the need to determine all container groups on the working node with the network failure as the container group to be scheduled for the network failure, thereby improving the disaster recovery capability of the container cluster.

[0115] Step 406: Send the container group information of the container group to be scheduled to the scheduling component in the container cluster, so that the scheduling component schedules the container group to be scheduled to the working node with normal network according to the container group information sent by the monitoring component on each working node.

[0116] In one or more embodiments of this specification, a monitoring component in each working node monitors the network through which the container group on the working node communicates with the outside world, obtains network status information of the working node, and determines, based on the network status information, a container group to be scheduled that has a network failure on the working node. Furthermore, the monitoring component can send the container group information of the container group to be scheduled to a scheduling component in the container cluster, so that the scheduling component schedules the container group to be scheduled to a working node with a normal network based on the container group information sent by the monitoring component on each working node.

[0117] By applying the solution of the embodiments of this specification, a monitoring component is used to monitor the network through which the container group on the working node communicates with the outside world, and the container group to be scheduled due to network failure on the working node is determined, and the network failure in the container cluster is accurately discovered. Moreover, after the network failure occurs, the scheduling component is used to promptly schedule the container group to be scheduled to the working node with normal network, thereby reducing project interruptions caused by network problems, improving the disaster recovery capability within the container cluster, and the stability and robustness of the projects in the container cluster, so that the container cluster can provide more stable service capabilities.

[0118] Referring to Figure 7, Figure 7 shows a flowchart of the processing process of a scheduling method provided by one embodiment of this specification. As shown in Figure 7, working node 1 includes container group A1, container group B2, monitoring container group M, monitoring container group N, and service components. Monitoring container group M is a physical network monitoring container group, while monitoring container group N is a virtual network monitoring container group. The taint tolerance information A1 of container group A1 indicates "tolerance of virtual network failures," while the taint tolerance information B2 of container group B2 indicates "tolerance of physical network failures."

[0119] After monitoring the network access path of worker node 1 by monitoring container group M and monitoring container group N, it is determined that the network status information of worker node 1 is "virtual network failure." The network status information is reported to the service component, and the service component marks worker node 1 with a taint label representing "virtual network failure" based on the network status information "virtual network failure." After worker node 1 is tainted, container group B2, which has network communication requirements for the virtual network, will be expelled from worker node 1 because worker node 1's current network capabilities are not met. Furthermore, when container group B2 is scheduled by the container cluster after expulsion, the scheduling component will filter out worker node 1 due to worker node 1's taint label when selecting the worker node corresponding to container group B2, and will schedule container group B2 to worker node 2, which has a normal network connection.

[0120] It is worth noting that before the network failure of worker node 1 is recovered, all container groups that cannot tolerate the tainted labels of worker node 1 will not be scheduled to worker node 1. As shown in Figure 7, worker node 2 includes two container groups B2, monitoring container group M, monitoring container group N, and service components.

[0121] By applying the solutions of the embodiments of this specification, a container group scheduling strategy based on the network connectivity of the working nodes is implemented. Monitoring components are established on the working nodes in the container cluster to detect the different network capabilities of the container groups on the working nodes to communicate with the outside world. Network failures are described using taint labels, and container groups that cannot tolerate such failures are scheduled to other working nodes with normal network conditions, thereby improving cluster stability and disaster recovery capabilities. Furthermore, cluster utilization is improved by refining connectivity capabilities for different network scenarios.

[0122] Corresponding to the above-mentioned scheduling method embodiment applied to the scheduling component, this specification also provides a scheduling device embodiment. Figure 8 shows a schematic diagram of the structure of a scheduling device provided by one embodiment of this specification. As shown in Figure 8, the device is applied to the scheduling component, and the scheduling component is set on the control node in the container cluster. The device includes:

[0123] Receiving module 802 is configured to receive container group information of a container group to be scheduled, sent by a monitoring component on each working node. The container group to be scheduled is a container group with a network failure. The container group information is obtained by the monitoring component based on network status information. The network status information is obtained by the monitoring component monitoring the network communication between the container group on the working node and the outside world.

[0124] The scheduling module 804 is configured to schedule the container group to be scheduled to a working node with a normal network according to the container group information of the container group to be scheduled sent by the monitoring component on each working node.

[0125] Optionally, the scheduling module 804 is further configured to search for a work node in the container cluster that does not carry a taint label; and schedule the container group to be scheduled to a work node that does not carry a taint label based on the container group information of the container group to be scheduled sent by the monitoring component on each work node.

[0126] Optionally, the container group information includes taint tolerance information of the container group to be scheduled; the scheduling module 804 is further configured to obtain the taint label of the working node in the container cluster; compare the taint label with the taint tolerance information, and use the working node corresponding to the taint label that meets the taint tolerance information as the working node with normal network; and schedule the container group to be scheduled to the working node with normal network.

[0127] By applying the solution of the embodiments of this specification, by utilizing the scheduling capabilities of the scheduling component, when a container cluster failure occurs, the container group is scheduled to a working node that can meet the normal network requirements of the container group based on the container group information of the container group, thereby reducing project interruptions caused by network problems, improving the disaster recovery capabilities within the container cluster and the stability and robustness of the projects in the container cluster, enabling the container cluster to provide more stable service capabilities, and further improving the overall service quality and user experience.

[0128] The above is a schematic scheme of a scheduling device of this embodiment. It should be noted that the technical scheme of the scheduling device and the technical scheme of the scheduling method applied to the scheduling component described above are based on the same concept. For details not described in detail in the technical scheme of the scheduling device, please refer to the description of the technical scheme of the scheduling method applied to the scheduling component described above.

[0129] Corresponding to the above-mentioned scheduling method embodiment applied to the monitoring component, this specification also provides a scheduling device embodiment. Figure 9 shows a schematic diagram of the structure of another scheduling device provided by one embodiment of this specification. As shown in Figure 9, the device is applied to the monitoring component, and the monitoring component is set on each working node in the container cluster. The device includes:

[0130] The monitoring module 902 is configured to monitor the network between the container group on the working node and the outside world, and obtain network status information of the working node;

[0131] The determination module 904 is configured to determine, based on the network status information, a container group to be scheduled due to a network failure on a working node;

[0132] The sending module 906 is configured to send the container group information of the container group to be scheduled to the scheduling component in the container cluster, so that the scheduling component schedules the container group to be scheduled to the working node with normal network according to the container group information sent by the monitoring component on each working node.

[0133] Optionally, the monitoring component includes a monitoring container group and a service component. The monitoring container group is constructed based on the container network interface in the container cluster; the monitoring container group is used to monitor the connectivity of the network access path between the container group on the working node and the outside world, obtain the network status information of the working node, and report the network status information to the service component; the service component is used to determine the container group to be scheduled due to network failure on the working node based on the network status information, and send the container group information of the container group to be scheduled to the scheduling component.

[0134] Optionally, the device further includes: a marking module configured to mark a taint label for the corresponding working node according to the network status information, wherein the taint label is used to mark the network where the container group on the working node has a communication failure with the outside world.

[0135] Optionally, the determination module 904 is further configured to obtain taint tolerance information of the container group on the working node; compare the network status information with the taint tolerance information, and determine the container group to be scheduled due to the network failure on the working node.

[0136] By applying the solution of the embodiments of this specification, a monitoring component is used to monitor the network through which the container group on the working node communicates with the outside world, and the container group to be scheduled due to network failure on the working node is determined, and the network failure in the container cluster is accurately discovered. Moreover, after the network failure occurs, the scheduling component is used to promptly schedule the container group to be scheduled to the working node with normal network, thereby reducing project interruptions caused by network problems, improving the disaster recovery capability within the container cluster, and the stability and robustness of the projects in the container cluster, so that the container cluster can provide more stable service capabilities.

[0137] The above is a schematic scheme of a scheduling device of this embodiment. It should be noted that the technical scheme of the scheduling device and the technical scheme of the scheduling method applied to the monitoring component described above are based on the same concept. For details not described in detail in the technical scheme of the scheduling device, please refer to the description of the technical scheme of the scheduling method applied to the monitoring component described above.

[0138] Figure 10 shows a block diagram of a computing device according to one embodiment of the present disclosure. Components of the computing device 1000 include, but are not limited to, a memory 1010 and a processor 1020. The processor 1020 is connected to the memory 1010 via a bus 1030, and a database 1050 is used to store data.

[0139] The computing device 1000 also includes an access device 1040 that enables the computing device 1000 to communicate via one or more networks 1060. Examples of such networks include a public switched telephone network (PSTN), a local area network (LAN), a wide area network (WAN), a personal area network (PAN), or a combination of communication networks such as the Internet. The access device 1040 may include one or more of any type of network interface (e.g., a network interface card (NIC)) whether wired or wireless, such as an IEEE 802.11 wireless local area network (WLAN) wireless interface, a World Interoperability for Microwave Access (Wi-MAX) interface, an Ethernet interface, a universal serial bus (USB) interface, a cellular network interface, a Bluetooth interface, a near field communication (NFC) interface, and the like.

[0140] In one embodiment of the present specification, the aforementioned components of the computing device 1000 and other components not shown in FIG10 may also be connected to each other, for example, via a bus. It should be understood that the computing device structure block diagram shown in FIG10 is for illustrative purposes only and does not limit the scope of this specification. Those skilled in the art may add or replace other components as needed.

[0141] Computing device 1000 may be any type of stationary or mobile computing device, including a mobile computer or mobile computing device (e.g., a tablet computer, personal digital assistant, laptop computer, notebook computer, netbook computer, etc.), a mobile phone (e.g., a smartphone), a wearable computing device (e.g., a smartwatch, smart glasses, etc.), or other types of mobile devices, or a stationary computing device such as a desktop computer or personal computer (PC). Computing device 1000 may also be a mobile or stationary server.

[0142] The processor 1020 is configured to execute the following computer-executable instructions, which implement the steps of the above-mentioned scheduling method when executed by the processor.

[0143] The above is a schematic solution of a computing device of this embodiment. It should be noted that the technical solution of the computing device and the technical solution of the above-mentioned scheduling method belong to the same concept. For details not described in detail in the technical solution of the computing device, please refer to the description of the technical solution of the above-mentioned scheduling method.

[0144] An embodiment of the present specification further provides a computer-readable storage medium storing computer-executable instructions, which implement the steps of the above-mentioned scheduling method when executed by a processor.

[0145] The above is a schematic scheme of a computer-readable storage medium of this embodiment. It should be noted that the technical solution of the storage medium and the technical solution of the scheduling method described above are of the same concept. For details not described in detail in the technical solution of the storage medium, please refer to the description of the technical solution of the scheduling method described above.

[0146] An embodiment of the present specification further provides a computer program, wherein when the computer program is executed in a computer, the computer is caused to execute the steps of the above-mentioned scheduling method.

[0147] The above is an illustrative solution of a computer program of this embodiment. It should be noted that the technical solution of the computer program and the technical solution of the scheduling method described above are of the same concept. For details not described in detail in the technical solution of the computer program, please refer to the description of the technical solution of the scheduling method described above.

[0148] The foregoing description of this specification describes specific embodiments. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in an order different from that described in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order shown or the sequential order to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0149] The computer instructions include computer program code, which may be in source code form, object code form, executable file, or some intermediate form. The computer-readable medium may include any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a mobile hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunications signal, and a software distribution medium.

[0150] It should be noted that for the aforementioned method embodiments, for the sake of simplicity of description, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the embodiments of this specification are not limited by the order of the actions described, because according to the embodiments of this specification, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in this specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the embodiments of this specification.

[0151] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0152] The preferred embodiments disclosed above are intended only to help illustrate this specification. The optional embodiments do not exhaustively describe all details, nor do they limit the invention to the specific embodiments described. Obviously, many modifications and variations can be made based on the content of the embodiments of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the embodiments of this specification, so that those skilled in the art can better understand and utilize this specification. This specification is limited only by the claims and their full scope and equivalents.

Claims

1. A scheduling system, comprising: A monitoring component and a scheduling component, wherein the monitoring component is arranged on each working node in the container cluster, and the scheduling component is arranged on a control node in the container cluster; The monitoring component is used to monitor the network in which the container group on the working node communicates with the outside world, and obtain the network status information of the working node; determine the container group to be scheduled due to the network failure on the working node according to the network status information, and send the container group information of the container group to be scheduled to the scheduling component; The scheduling component is used to schedule the container group to be scheduled to the working node with normal network according to the container group information sent by the monitoring component on each working node.

2. A scheduling method, applied to a scheduling component, wherein the scheduling component is set at a control node in a container cluster, the method comprising: Receive container group information of a container group to be scheduled sent by a monitoring component on each working node, wherein the container group to be scheduled is a container group with a network failure, and the container group information is obtained by the monitoring component according to network status information, and the network status information is obtained by the monitoring component monitoring a network in which the container group on the working node communicates with the outside world; According to the container group information of the container group to be scheduled sent by the monitoring component on each working node, the container group to be scheduled is scheduled to a working node with a normal network.

3. The method according to claim 2, wherein scheduling the container group to be scheduled to a working node with a normal network according to the container group information of the container group to be scheduled sent by the monitoring component on each working node comprises: Find a working node in the container cluster that does not carry a tainted label; According to the container group information of the container group to be scheduled sent by the monitoring component on each working node, the container group to be scheduled is scheduled to the working node that does not carry the taint label.

4. The method according to claim 2, wherein the container group information includes taint tolerance information of the container group to be scheduled; The step of scheduling the container group to be scheduled to a working node with a normal network according to the container group information of the container group to be scheduled sent by the monitoring component on each working node includes: Obtaining the taint labels of the working nodes in the container cluster; Comparing the taint label with the taint tolerance information, and taking the working node corresponding to the taint label that satisfies the taint tolerance information as a normal working node of the network; The container group to be scheduled is scheduled to the working node where the network is operating normally.

5. A scheduling method, applied to a monitoring component, wherein the monitoring component is set at each working node in a container cluster, the method comprising: Monitor the network in which the container group on the working node communicates with the outside world, and obtain network status information of the working node; Determine, according to the network status information, a container group to be scheduled due to a network failure on the working node; The container group information of the container group to be scheduled is sent to the scheduling component in the container cluster, so that the scheduling component schedules the container group to be scheduled to the working node with normal network according to the container group information sent by the monitoring component on each working node.

6. The method according to claim 5, wherein the monitoring component comprises a monitoring container group and a service component, and the monitoring container group is constructed based on a container network interface in the container cluster; The monitoring container group monitors the connectivity of the network access path between the container group on the working node and the external world, obtains the network status information of the working node, and reports the network status information to the service component; The service component determines, according to the network status information, a container group to be scheduled due to a network failure on the working node, and sends container group information of the container group to be scheduled to the scheduling component.

7. The method according to claim 5, wherein after monitoring the network of the container group on the working node communicating with the outside world and obtaining the network status information of the working node, the method further comprises: According to the network status information, a taint label is marked for the corresponding working node, wherein the taint label is used to mark the network where the container group on the working node fails to communicate with the outside world.

8. According to the method of claim 5, determining the container group to be scheduled for the network failure on the working node according to the network status information comprises: Obtaining taint tolerance information of the container group on the working node; The network status information and the stain tolerance information are compared to determine a container group to be scheduled for a network failure on the working node.

9. A container cluster, comprising: A control node and multiple working nodes, wherein the control node is provided with a scheduling component, and the working nodes are provided with a monitoring component and a container group; The monitoring component is used to determine the container group to be scheduled due to network failure on the working node according to the scheduling method according to any one of claims 5 to 8; The scheduling component is used to schedule the container group to be scheduled to a working node with a normal network according to the scheduling method according to any one of claims 2 to 4.

10. A computing device comprising: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the method described in any one of claims 2 to 4 or any one of claims 5 to 8 are implemented.

11. A computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the method described in any one of claims 2 to 4 or any one of claims 5 to 8.

12. A computer program, wherein: When the computer program is executed in a computer, the computer is caused to execute the steps of the method described in any one of claims 2 to 4 or any one of claims 5 to 8.

Citation Information

Patent Citations

  • Container deployment method and device

    CN111506391A

  • Container group POD reconstruction method based on container cluster service and related equipment

    CN113872997A

  • Container group migration method, electronic equipment and storage medium

    CN114816656A

  • Scheduling method, scheduling system and container cluster

    CN116684258A

  • Communications methods and apparatus for migrating a network interface and / or IP address from one POD to another POD in a kubernetes system

    US20210328858A1

Cited By

  • Cluster deployment method and device and storage medium

    CN119254776A

  • Load balancing method suitable for private deployment cluster and related product

    CN120256149A