WO2020143297A1

WO2020143297A1 - Disaster recovery method, apparatus and device for call center, and storage medium

Info

Publication number: WO2020143297A1
Application number: PCT/CN2019/116744
Authority: WO
Inventors: 杨忠诚; 陈小建; 王志鹏
Original assignee: 平安科技（深圳）有限公司
Priority date: 2019-01-10
Filing date: 2019-11-08
Publication date: 2020-07-16
Also published as: CN109660405A; CN109660405B

Abstract

The present application belongs to the technical field of call centers. Disclosed are a disaster recovery method, apparatus and device for a call center, and a computer-readable storage medium. The method comprises: starting a main call center, and collecting log information generated during a running process of the main call center; analyzing the log information in real time, and determining whether a running state of the main call center is normal; if the running state of the main call center is normal, when a call service is received at a pre-allocated virtual address, transferring, according to a first correlation, the call service to the main call center by means of a physical address of the main call center; and if the running state of the main call center is abnormal, starting a backup call center, and when the call service is received at the virtual address, transferring, according to a second correlation, the call service to the backup call center by means of a physical address of the backup call center.

Description

Disaster recovery method, device, equipment and storage medium of call center The

This application requires the priority of the Chinese patent application submitted to the China Patent Office on January 10, 2019, with the application number 201910029724.1 and the invention titled "Call Center Disaster Recovery Methods, Devices, Equipment, and Storage Media", all of which are approved by The reference is incorporated in the application.

Technical field

The present application relates to the technical field of call centers, and in particular to a disaster recovery method, device, equipment and storage medium for call centers.

Background technique

The call center, also known as the customer service center, originated in the 1930s and initially transferred the user's call to an answering desk or an expert. Since then, with the increase of calls and answers to be transferred, an interactive voice answering system has been established, which can realize the answering and processing of some common questions of customers by the machine "automatic attendant". The call center in the traditional sense refers to a call response center mainly based on telephone reception, providing customers with various telephone response services.

At present, the queuing and dispatching operation of the incoming call of the call center system launched by major enterprises is handled by the callcenter (call center) module of Freeswitch (a soft switch solution for a telephone). However, the callcenter module can only start one instance in a call center system. Therefore, when the callcenter module itself or the server where the callcenter module is located is abnormal, it will cause the entire call center system to queuing and dispatch all incoming calls, which will cause enterprises Serious impact.

Therefore, there is an urgent need to provide a disaster recovery method for call centers to reduce business risks, thereby ensuring the stability and reliability of the call center system.

Summary of the invention

The main purpose of the present application is to provide a disaster recovery method, device, equipment and storage medium for a call center, aiming to solve the technical problems of poor stability and low reliability of the call center system in the prior art.

To achieve the above purpose, the present application provides a disaster recovery method for a call center. The method includes the following steps:

Start the main call center and collect the log information generated during the operation of the main call center;

Real-time analysis of the log information to determine whether the main call center's operating status is normal;

If the operation status of the main call center is normal, when the pre-assigned virtual address receives the call service, the call service is transferred to the main call center through the physical address of the main call center according to the first correspondence, the first correspondence is virtual The correspondence between the address and the physical address of the main call center;

If the operation status of the main call center is abnormal, the backup call center is started, and when the call service is received at the virtual address, the call service is transferred to the backup call center through the physical address of the backup call center according to the second correspondence. The relationship is the correspondence between the virtual address and the physical address of the backup call center.

In addition, to achieve the above purpose, the present application also proposes a disaster recovery device for a call center, the device includes:

The log information collection module is used to start the main call center and collect the log information generated during the operation of the main call center;

The log information analysis module is used to analyze the log information in real time and determine whether the operation status of the main call center is normal;

The first call service transfer module is used to transfer the call service to the main call through the physical address of the main call center according to the first correspondence when the pre-assigned virtual address receives the call service when the main call center is operating normally Center, the first correspondence is the correspondence between the virtual address and the physical address of the main call center;

The second call service transfer module is used to start the backup call center when the operation status of the main call center is abnormal, and when the virtual address receives the call service, according to the second correspondence, pass the call service through the physical address of the backup call center Transferred to the backup call center, the second correspondence is the correspondence between the virtual address and the physical address of the backup call center.

In addition, in order to achieve the above purpose, the present application also proposes a disaster recovery device for a call center, the device includes: a memory, a processor, and a computer-readable storage stored on the memory and operable on the processor Instructions, when the computer-readable instructions are executed by the processor to implement the steps of the disaster recovery method of the call center as described above.

In addition, in order to achieve the above object, the present application also proposes a computer-readable storage medium having computer-readable instructions stored on the computer-readable storage medium, the computer-readable instructions being executed by a processor are implemented as described above The steps of the disaster recovery method of the call center.

In the disaster recovery method, device, equipment and storage medium of the call center of this embodiment, when judging whether the operation state of the main call center is normal, it specifically adopts the method of software detection, that is, after the main call center is started, it is collected through real-time The log information generated during the operation of the main call center, and then based on the analysis of the log information, determine whether the operation status of the main call center is normal. When it is determined that the operation status of the main call center is normal, when there is a call service, the main call center directly Call center processing; when it is determined that the operation status of the main call center is abnormal, the backup call center is started, and the backup call center takes over the call service processed by the main call center. This way of announcement not only can detect whether the server deploying the main call center When an exception occurs, it can also detect whether the process and port of the main call center are abnormal, thereby ensuring that the call center system can handle call services for users stably and reliably.

In addition, the disaster recovery method, device, equipment and storage medium of the call center of this embodiment provide users with a fixed and unique virtual address from beginning to end, and the switching between the physical addresses of the main and standby call centers is based on The operation status of the main and standby call centers and the pre-built first correspondence relationship and the second correspondence relationship are realized, so the user cannot perceive during the handover process, thereby effectively ensuring the user experience effect.

BRIEF DESCRIPTION

FIG. 1 is a schematic structural diagram of a disaster recovery device of a call center in a hardware operating environment according to an embodiment of the present application;

FIG. 2 is a schematic flowchart of a first embodiment of a disaster recovery method for a call center;

3 is a schematic flowchart of a second embodiment of a disaster recovery method for a call center;

4 is a structural block diagram of a first embodiment of a disaster recovery device of a call center of the present application.

The implementation, functional characteristics and advantages of the present application will be further described in conjunction with the embodiments and with reference to the drawings.

detailed description

It should be understood that the specific embodiments described herein are only used to explain the present application, and are not used to limit the present application.

Referring to FIG. 1, FIG. 1 is a schematic structural diagram of a disaster recovery device of a call center in a hardware operating environment according to an embodiment of the present application.

As shown in FIG. 1, the disaster recovery equipment of the call center may include: a processor 1001, such as a central processor (Central Processing Unit, CPU), communication bus 1002, user interface 1003, network interface 1004, memory 1005. Among them, the communication bus 1002 is used to implement connection communication between these components. The user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the optional user interface 1003 may further include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a wireless fidelity (WIreless-FIdelity, WI-FI) interface). The memory 1005 may be a high-speed random access memory (Random Access Memory (RAM) memory can also be a stable non-volatile memory (Non-Volatile Memory, NVM), such as disk storage. The memory 1005 may optionally be a storage device independent of the foregoing processor 1001.

Those skilled in the art may understand that the structure shown in FIG. 1 does not constitute a limitation on the disaster recovery equipment of the call center, and may include more or fewer components than those illustrated, or combine certain components, or different components Layout.

As shown in FIG. 1, the memory 1005 as a computer-readable storage medium may include an operating system, a network communication module, a user interface module, and a disaster recovery program of a call center.

In the disaster recovery equipment of the call center shown in FIG. 1, the network interface 1004 is mainly used for data communication with the network server; the user interface 1003 is mainly used for data interaction with the user; the processing in the disaster recovery equipment of the call center of this application The device 1001 and the memory 1005 may be provided in the disaster recovery device of the call center. The disaster recovery device of the call center calls the disaster recovery program of the call center stored in the memory 1005 through the processor 1001 and executes the call center provided in the embodiment of the present application. Disaster recovery methods.

An embodiment of the present application provides a disaster recovery method for a call center. Referring to FIG. 2, FIG. 2 is a schematic flowchart of a first embodiment of a disaster recovery method for a call center according to the application.

In this embodiment, the disaster recovery method of the call center includes the following steps:

In step S10, the main call center is started to collect log information generated during the operation of the main call center.

Specifically, the disaster recovery method of the call center provided in this embodiment is mainly applied to the pre-built call center system. In practical applications, the call center system generally includes a server with a main call center and a backup call center. The server and the switch used to control the main call center and the switch used to control the backup call center respectively.

Correspondingly, the operation of starting the call center mentioned in the above step S10 is specifically executed by the exchange for controlling the main call center connected to the access communication where the main call center is deployed.

In addition, the switch mentioned in this embodiment is specifically a switch based on a virtual routing redundancy protocol. In practical applications, Keepalived software may be used instead.

The so-called Keepalived is similar to layer3, 4 & The software of the 5 exchange mechanism, which is what we usually say about layer 3, layer 4 and layer 5 switching. Its role is mainly to detect the status of the server, and the entire detection process is automatically completed without human intervention.

For ease of understanding, the following examples illustrate:

For example, when using Keepalived to detect the status of a server, if a server goes down or the work fails, Keepalived will detect and remove the faulty server from the system, and use other servers to replace the work of the server. When the removed server works normally, Keepalived will automatically add the server to the server group, and all these tasks are automatically completed without manual intervention. All that is required is to repair the failed server.

In addition, it is worth mentioning that in actual applications, before performing the steps of starting the main call center, the main call center and the backup call center need to be determined. Regarding, the operations for determining the main call center and the backup call center can be roughly as follows: First, according to the service requirements of the call center, build a first call center and a second call center. Then, any one of the first call center and the second call center is designated as the main call center, and the other is designated as the backup call center.

For ease of understanding, this embodiment provides a specific implementation manner for determining the main call center and the backup call center, which are roughly as follows:

(1) According to the preset monitoring indicators, collect the first monitoring indicator data provided by the first call center and the second monitoring indicator data provided by the second call center in time intervals.

Specifically, the monitoring indicators mentioned above may be specified data traffic and address resolution protocol (Address Resolution Protocol (ARP), media access control (Media Access Control or Medium Access Control, MAC) address and other common network indicators, so as to determine whether the network status of the first call center and the second call center is stable and whether the concurrent processing capability is good according to these network indexes. In practical applications, the selection of monitoring indicators can be set by those skilled in the art as needed, and this case does not limit this.

Correspondingly, the above-mentioned first monitoring index data is the data corresponding to the monitoring index generated during the operation of the first call center; the second monitoring index data is the data generated during the operation of the second call center Data corresponding to monitoring indicators.

In order to facilitate understanding of the meaning of the network index data in this embodiment, the following examples illustrate: If the monitoring index is ARP, the first monitoring index data is the number of ARP provided by the first call center during operation. The second monitoring indicator data is the number of ARPs provided by the second call center during operation; if the monitoring indicator is a MAC address, the first monitoring indicator data is the MAC address drift of the first call center during operation, the second The monitoring index data is the MAC address drift during the operation of the second call center.

It is not difficult to find from the above description that, in this example, when the first monitoring indicator data and the second monitoring indicator data corresponding to the preset monitoring indicators are collected from the first call center and the second call center, they are collected in specific time periods, thus It can make the subsequent analysis analyze the change of the monitoring index data corresponding to the monitoring index within a period, rather than just analyzing the situation at a certain point in time. Therefore, it is possible to effectively eliminate the transient anomalies and ensure the rationality of the determined main call center and backup call center as much as possible.

(2) Analyze the first monitoring indicator data in each period and the second monitoring indicator data in each period respectively to obtain the monitoring indicator change rate corresponding to the first call center and the monitoring indicator change rate corresponding to the second call center.

Specifically, in practical applications, in order to facilitate the analysis of the first monitoring indicator data in each period and the second monitoring indicator data in each period, in order to quickly and accurately obtain the monitoring indicator change rate of the first call center and the second call The change rate of the monitoring index of the center can be analyzed based on the preset analysis model to analyze the first monitoring index data of each period and the second monitoring index data of each period respectively to obtain the monitoring index change rate and the corresponding number of the first call center 2. The rate of change of monitoring indicators corresponding to the call center.

Regarding, the method of constructing the analysis model can be constructed based on the deep machine learning method in practical applications. Specifically, the deep learning method described in this embodiment specifically adopts an unsupervised learning method (such as Deep Trust Network (Deep Belief Nets (DBNs)) and supervised learning methods (such as Convolutional neural network networks, CNNs)) to build analytical models.

For easy understanding, the following is a brief description of the operation of building an analysis model based on deep machine learning methods. The specific steps are as follows:

(2-1) Construct the first training model based on the sample data.

Specifically, the constructed first training model specifically includes an input layer, an output layer, and multiple hidden layers. In addition, multiple hidden layers are located between the input layer and the output layer, and all layers are fully connected.

In addition, in order to ensure the accuracy of the training results, a filter can be added before each layer to filter out the interference information in the sample data.

In addition, it should be understood that the sample data used when constructing the first training model in this embodiment may specifically be the massive data stored in the major data platforms, thus ensuring that the constructed first training model can be better The rich intrinsic information and features of the data are carved out so that the analysis model trained based on the first training model can better predict the rate of change corresponding to the monitoring index.

(2-2) According to the preset layering standard, split the initial network layer of each hidden layer in the first training model into at least two sub-network layers.

Specifically, the above-mentioned layering standards are specifically used to specify how to split the initial network layer into at least two sub-network layers of large size. For example, when the size of the initial network layer is a convolution kernel of size, the layering criterion may be that the convolution kernel of size is divided into two convolution kernels. In this way, when the subsequent steps (2-3) and (2-4) are performed, the first training model after splitting the initial network layer is used for training, the network depth of the training model can be increased, so that the subsequent The trained analysis model can accurately predict the rate of change corresponding to the monitoring index.

(2-3) Use the unsupervised training method that rises from the bottom, starting with the hidden layer connected to the input layer in the first training model, and ending with the hidden layer connected to the output layer in the first training model. The sub-network layer in each hidden layer in the training model is trained to obtain the second training model.

Specifically, in practical applications, the data for constructing the first training model may be calibrated data or uncalibrated data. The training method of the first training model constructed for different training data will also be different. For ease of understanding, the following uses the training data as uncalibrated data as an example for specific description. Specifically, when using the unsupervised training method that rises from the bottom to train the sub-network layer in each hidden layer in the first training model, you need to train the first layer (the hidden layer connected to the input layer) first, and learn the first The parameters of the first layer. Then, after learning the parameters of the first layer, the output of the first layer is used as the input of the second layer, and so on. After learning the n-1 layer, the output of the n-1 layer is taken as the n-th layer. Input, train the nth layer, and get the parameters of each layer. Due to the limitation of the capacity of the first training model and the sparseness constraint, the structure of the data itself can be learned during the training process, thereby obtaining a second training model that has more representative characteristics than the input.

(2-4) A top-down supervised training method is used, starting with the hidden layer connected to the output layer in the second training model, and the hidden layer connected to the input layer in the second training model as the end point. The sub-network layer in each hidden layer in the training model is trained to obtain an analysis model.

Specifically, the top-down supervised training method is used to train each layer in the second training model obtained in step (2-3), so that the error is transmitted from top to bottom, thereby reaching the entire network. Fine-tune, and then get an analysis model that can get better results.

(3) According to the preset main call center selection rules, the first call center whose monitoring index change rate meets the main call center selection rules is designated as the main call center, and the second call center is designated as the backup call center; or, the monitoring indicators The second call center whose change rate meets the selection rules of the main call center is designated as the main call center, and the first call center is designated as the backup call center.

Specifically, in practical applications, the main call center selection rule may stipulate that a call center with stable network status and good concurrent processing capability is selected as the main call center.

Correspondingly, when selecting the main call center, by comparing the change rate of the monitoring index of the first call center and the change rate of the monitoring index of the second call center, a call center that meets the selection rules of the main call center is selected as the main call Center, the other as a backup call center.

For example, if the monitoring index change rate in the first call center meets the selection rules of the main call center, the first call center is designated as the main call center, and the second call center is designated as the backup call center; for example, the second call center The rate of change of the monitoring index meets the selection rules of the main call center, and the second call center is designated as the main call center, and the first call center is designated as the backup call center.

In addition, it should be understood that the "first" in the above-mentioned "first call center" is only used to distinguish the call center from other call centers, and does not limit the call center itself, and the first call center It can be any call center in the call center system. Correspondingly, the "second" in the "second call center" is only used to distinguish the call center from other call centers, and does not limit the call center itself, and the second call center can be used in the call center system. Any call center other than a call center. In addition, both the first call center and the second call center are deployed on the server in practical applications. Therefore, when one of them is designated as the main call center and the other is the backup call center, their respective The server can be called the main server and the backup server.

Step S20: Perform real-time analysis on the log information to determine whether the operation status of the main call center is normal.

Specifically, the log information mentioned in this embodiment mainly includes data information generated by each port and process of the main call center, and various data information generated during the operation of the server deploying the main call center.

Therefore, by analyzing the collected log information, not only can the running status of each port and process of the main call center be detected, but also the server where the main call center is deployed can be detected.

Correspondingly, through analysis and judgment, if it is detected that the main call center is operating normally (both the main call center and the server where the main call center is deployed), then the operation in step 30 is performed; if the main call center is detected to be abnormal (main If the call center and/or the server deploying the main call center is abnormal), the operation in step S40 is performed.

Step S30: When the pre-allocated virtual address receives the call service, the call service is transferred to the main call center through the physical address of the main call center according to the first correspondence.

Specifically, the virtual address mentioned above specifically refers to Virtual IP Address, commonly known as virtual IP, or VIP for short. It is an IP address that is not connected to a specific computer or a network interface card (NIC) in a computer.

Correspondingly, the physical address, the so-called interconnection protocol between the networks (Internet Protocol, IP). In this case, the IP used to indicate the main call center is its real physical IP.

It should be noted that in practical applications, the VIP is allocated to the user for access. After the data packet generated by the user during the access process is sent to this VIP address, the VIP is transferred to the real physical IP, which is described above According to the first correspondence, the call service is transferred to the main call center through the physical address of the main call center, so that the main call center processes the call service.

In addition, it is worth mentioning that the first correspondence mentioned above is specifically the correspondence between the virtual address and the physical address of the main call center.

In step S40, the backup call center is started, and when the call service is received at the virtual address, the call service is transferred to the backup call center through the physical address of the backup call center according to the second correspondence.

Specifically, in practical applications, when it is determined that the operation status of the main call center is abnormal, Keepalived, which is used to control the main call center, will notify Keepalived, which is used to control the backup call center, and the execution subject is switched to control the backup call center. Keepalived, the backup call center is started by Keepalived, which controls the backup call center, that is, the callcenter module in the backup call center is started, so that the backup call center enters the working state, so that when the call service is received at the virtual address, the backup call center can be controlled Keepalived transfers the call service to the backup call center through the physical address of the backup call center according to the second correspondence, so that the backup call center can process the call service.

In addition, in order to facilitate the understanding of the above switching method, the following is an example:

For example, the physical IP of the server deploying the main call center is: 10.37.255.1, the physical IP of the backup server deploying the backup call center is: 10.37.255.2, the pre-assigned VIP is: 10.37.255.3, and the VIP under normal circumstances: 10.37 .255.3 is bound to the physical IP of the main call center: 10.37.255.1, so when users visit 10.37.255.3, they are actually accessing 10.37.255.1. When the main call center or the server where the main call center is deployed is abnormal , VIP will automatically switch to the physical IP corresponding to the backup call center: 10.37.255.2, when the user accesses 10.37.255.3, actually accesses 10.37.255.2. However, corresponding to users, the switch of physical IP is insensitive, and users only need to access 10.37.255.3 during the entire access process.

It is not difficult to find from the above description that the disaster recovery method of the call center provided in this embodiment adopts the software detection method when judging whether the operation state of the main call center is normal, that is, after the main call center is started, it is collected in real time The log information generated during the operation of the main call center, and then based on the analysis of the log information, determine whether the operation status of the main call center is normal. When it is determined that the operation status of the main call center is normal, when there is a call service, the main call center directly Call center processing; when it is determined that the operation status of the main call center is abnormal, the backup call center is started, and the backup call center takes over the call service processed by the main call center. This way of announcement not only can detect whether the server deploying the main call center When an exception occurs, it can also detect whether the process and port of the main call center are abnormal, thereby ensuring that the call center system can handle call services for users stably and reliably.

In addition, in the disaster recovery method of the call center provided in this embodiment, a fixed and unique virtual address is provided to the user from beginning to end, and the physical address switching between the main and backup call centers is based on the main and backup call centers. Is implemented in the running state and the pre-built first correspondence relationship and the second correspondence relationship, so the user cannot perceive during the handover process, thereby effectively ensuring the user experience effect.

In addition, it is worth mentioning that in practical applications, in order to ensure the security of the information generated by the call center system, when the first call center and the second call center are deployed, the server and the first call center can also be deployed. Deploy a second blockchain-based data storage area in the server where the second call center is deployed, or deploy several servers separately to store the information generated by the call center system.

The creation of the data storage area based on the blockchain is roughly as follows: according to the call service supported by the call center system, several blocks are established, and several blocks are connected in sequence to form a blockchain. Correspondingly, the operation of adding the information to be stored to the blockchain is roughly as follows:

(1) After transferring the received call service to the corresponding call center, obtain the service data generated in the process of processing the call service in real time; (2) Perform data conversion processing on the service data according to the preset conversion rules to obtain the pending Encrypted character string; (3) According to the call service, determine the block that stores the business data, and obtain the block number of the block; (4) According to the pre-established mapping relationship table, determine the original key of the block, mapping relationship table Correspondence between the block number and the original key; (5) Obtain the original key vector of the original key, use the homomorphic encryption algorithm to encrypt the original key vector to obtain the homomorphic encryption vector; (6) Use Ha The Greek key algorithm encrypts the random number in the homomorphic key vector to obtain the hash result; (7) uses the symmetric encryption algorithm to encrypt the ciphertext data in the homomorphic encryption vector to obtain the symmetric encryption result; (8) Kazakhstan The XOR result and the symmetric encryption result are XORed to generate the target key; (9) The target key is used to encrypt the encrypted character string to obtain the ciphertext data corresponding to the business data, and the ciphertext data is stored in the block.

Regarding the above-mentioned preset conversion rule, in practical applications, it may be a binary conversion rule.

Correspondingly, the character string to be encrypted after the data conversion processing operation on the business data is a binary character string composed of only 0 and 1.

It should be noted that the reason why the business data is converted into a binary format in this embodiment is to facilitate subsequent processing, and to avoid abnormality in the encrypted information due to inconsistent data formats.

In addition, the block number mentioned above is specifically an identification number assigned to the block by the technician when the block is created according to the call service supported by the call center system. In addition, in practical applications, in order to ensure the uniqueness and security of the original key, the original key can be generated according to the unique digital certificate applied for from the digital center. For example, when generating the original key of any block, you can first obtain the digital certificate corresponding to the block, and then arrange and combine the characters in the digital certificate according to the preset rules to obtain the original key of the block. In addition, in practical applications, in order to further improve the security of business data, before performing data conversion operations on the business data according to the preset conversion rules, the characters in the business data can also be arranged and combined according to the preset rules, and then In accordance with preset conversion rules, perform data conversion operations on the arranged and combined business data.

Referring to FIG. 3, FIG. 3 is a schematic flowchart of a second embodiment of a disaster recovery method for a call center of the present application. The disaster recovery method of the call center provided in this embodiment is further improved on the basis of the first embodiment. The specific improvement is that after determining that the operation status of the main call center is abnormal and starting the backup call center, the monitoring of the main The operation status of the call center, and when the operation status of the main call center is restored to normal, the virtual address drift will stop the operation of the backup call center before the physical address of the main call center. For ease of understanding, the following is a detailed description with reference to FIG. 3:

Step S50, start the backup call center, and monitor whether the running status of the main call center returns to normal.

Specifically, since the backup call center is started, the executive body has been replaced by Keepalived for controlling the backup call center, so the monitoring of the operation status of the main call center may be specifically controlled by the Keepalived slave controlling the main call of the backup call center. The center's Keepalived obtains the log information of the main call center at the current moment, and then analyzes the obtained log information to determine whether the operation status of the main call center has returned to normal.

Accordingly, the obtained log information of the main call center may also include log information of each port and thread of the main call center, and log information of the server where the main call center is deployed.

In step S60, if the running state of the main call center returns to normal, a stop operation instruction is sent to the backup call center to stop the backup call center, and after the backup call center stops running, the main call center is restarted.

Specifically, in order to prevent the virtual address from switching back and forth between the main call center and the backup call center to form a "brain split", after detecting that the operation state of the main call center returns to normal, the disaster recovery of the call center given in this embodiment The method is to first send the stop call instruction to the backup call center by Keepalived, which controls the backup call center, so that the custom stop script on the backup call center closes the call center module and ESL application of the backup call center according to the stop operation instruction. At the same time, after the above modules and applications are closed, Keepalived, which controls the backup call center, will restart and enter the standby state. At this time, the executive body switches to Keepalived, which controls the main call center, and Keepalived, which controls the main call center, restarts the main call center, so that Subsequent call services can be transferred to the main call center for processing.

It should be noted that the "brain split" mentioned in this embodiment refers specifically to the switch between the main call center and the backup call center. As a result of incomplete switching or other reasons, the client and the slave ( Slave) mistakenly thought that two active The master (host) eventually leaves the entire cluster in a chaotic state. And ESL, specifically an interface provided by Freeswitch, the full name is event-sodket-library, in practical applications is usually understood as a telephone event processing component. Through the ESL interface, Freeswitch realizes external control, filters the message events sent by FreeSwitch, and encapsulates it into DTO (data transmission object). The specific use is to obtain the soft phone status of the agent through ESL. When the agent calls out or incoming The phone bar is flipped to identify the current state of the agent, such as busy, busy, idle, etc.

It is not difficult to find from the above description that after the disaster recovery method of the call center provided in this embodiment detects that the operation status of the main call center returns to normal, the virtual address does not directly drift back to the main call center, but first to the backup call center Send a stop operation command, and restart the main call center after the backup call center stops running according to the stop operation command, so that the virtual address drifts back to the main call center. This switching method effectively prevents the brain split caused by the virtual address switching back and forth. Ensure the normal processing of subsequent call services.

In addition, in practical applications, in order to facilitate the management personnel of the call center system to be able to know the situation of sending abnormal call centers in time, and quickly and accurately locate the fault problem, when the operating state of the main call center is abnormal, start backup After the call center, you can also generate warning information based on the collected log information of the main call center, and then notify the administrator of the main call center according to the preset warning notification method.

In addition, in order to allow the management personnel to view the warning information in time, and respond to the possible abnormalities of the main call center in time, such as remotely adjusting the parameters of the main call center, or directly to the site to check the main call center or deploy the server of the main call center Status, after sending the warning information to the management personnel, you can also control the device that receives the warning information to make warnings such as ringing, vibration, and screen flashing.

It should be noted that those of ordinary skill in the art may understand that all or part of the steps to implement the above embodiments may be completed by hardware, or by computer-readable instructions to instruct related hardware. The computer-readable instructions may be stored in In a computer-readable storage medium, the aforementioned computer-readable storage medium may be a read-only memory, a magnetic disk, or an optical disk. In addition, an embodiment of the present invention also proposes a computer-readable storage medium having computer-readable instructions stored on the computer-readable storage medium. When the computer-readable instructions are executed by a processor, a call center as described above is implemented Steps of the disaster recovery method. The computer-readable storage medium may be a non-volatile computer-readable storage medium.

Referring to FIG. 4, FIG. 4 is a structural block diagram of a first embodiment of a disaster recovery apparatus of a call center of the present application. The disaster recovery apparatus of the call center provided in the embodiment of the present application includes: a log information collection module 4001, a log information analysis module 4002, a first call service transfer module 4003, and a second call service transfer module 4004. Among them, the log information collection module 4001 is used to start the main call center to collect the log information generated during the operation of the main call center; the log information analysis module 4002 is used to perform real-time analysis on the log information to determine whether the main call center's operating status is Normal; the first call service transfer module 4003 is used to transfer the call service through the physical address of the main call center according to the first correspondence when the pre-assigned virtual address receives the call service in the normal operation state of the main call center To the main call center; the second call service transfer module 4004 is used to start the backup call center when the main call center is abnormal, and when the virtual address receives the call service, according to the second correspondence, pass the call service The physical address of the backup call center is transferred to the backup call center.

It should be noted that the first correspondence in this embodiment is specifically the correspondence between the virtual address and the physical address of the primary call center; accordingly, the second correspondence is specifically the physical relationship between the virtual address and the backup call center The correspondence between addresses.

In addition, in actual applications, before performing the steps of starting the main call center, the main call center and the backup call center need to be determined. Regarding, the operations of determining the main call center and the backup call center can be roughly as follows: First, build the first call center and the second call center according to the service requirements of the call center; then, set any of the first call center and the second call center One is designated as the primary call center, and the other is designated as the backup call center.

For ease of understanding, this embodiment provides a specific implementation method for determining the main call center and the backup call center, which is roughly as follows: First, according to the preset monitoring indicators, the first monitoring indicator data provided by the first call center is collected in time intervals And the second monitoring indicator data provided by the second call center; then, the first monitoring indicator data and the second monitoring indicator data in each period are analyzed separately to obtain the monitoring indicator change rate and the first monitoring indicator corresponding to the first call center The change rate of the monitoring index corresponding to the second call center; finally, according to the preset main call center selection rule, the first call center whose monitoring index change rate meets the main call center selection rule is designated as the main call center, and the second call center is designated as The backup call center; or, the second call center whose monitoring index change rate meets the selection rules of the main call center is designated as the main call center, and the first call center is designated as the backup call center.

In addition, it is worth mentioning that, in practical applications, in order to facilitate the analysis of the first monitoring indicator data and the second monitoring indicator data in each time period, in order to quickly and accurately obtain the monitoring indicator change rate of the first call center And the change rate of the monitoring indicators of the second call center, based on the preset analysis model, the first monitoring indicator data of each period and the second monitoring indicator data of each period can be analyzed respectively to obtain the monitoring indicators corresponding to the first call center The rate of change and the rate of change of monitoring indicators corresponding to the second call center.

Regarding, the method of constructing the analysis model can be constructed based on the deep machine learning method in practical applications. Specifically, the deep learning method described in this embodiment specifically adopts a combination of unsupervised learning and supervised learning to build an analysis model.

In order to facilitate understanding, the following is a brief description of the operation of building an analysis model based on deep machine learning methods. The specific steps are as follows: first, build a first training model based on sample data; then, according to preset layering standards, the first training The initial network layer of each hidden layer in the model is split into at least two sub-network layers; then, the bottom-up unsupervised training method is used, starting with the hidden layer connected to the input layer in the first training model, and the first training The hidden layer connected to the output layer in the model is the end point, and the sub-network layers in each hidden layer in the first training model are sequentially trained to obtain the second training model; finally, a top-down supervised training method is used to The hidden layer connected to the output layer in the second training model is the starting point, and the hidden layer connected to the input layer in the second training model is the end point. The sub-network layers in each hidden layer in the second training model are sequentially trained to obtain the analysis model.

It is not difficult to find from the above description that the disaster recovery device of the call center provided in this embodiment adopts the software detection method when judging whether the operation status of the main call center is normal, that is, after the main call center is started, it is collected through real-time The log information generated during the operation of the main call center, and then based on the analysis of the log information, determine whether the operation status of the main call center is normal. When it is determined that the operation status of the main call center is normal, when there is a call service, the main call center directly Call center processing; when it is determined that the operation status of the main call center is abnormal, the backup call center is started, and the backup call center takes over the call service processed by the main call center. This way of announcement not only can detect whether the server deploying the main call center When an exception occurs, it can also detect whether the process and port of the main call center are abnormal, thereby ensuring that the call center system can handle call services for users stably and reliably.

In addition, the disaster recovery device of the call center provided in this embodiment provides users with a fixed and unique virtual address from beginning to end, and the physical addresses of the primary and secondary call centers are switched according to the primary and secondary call centers. The running state and the pre-built first correspondence relationship and the second correspondence relationship are realized, so the user cannot perceive during the handover process, thereby effectively ensuring the user experience effect.

Based on the first embodiment of the disaster recovery apparatus of the call center described above, a second embodiment of the disaster recovery apparatus of the call center of the present application is proposed.

In this embodiment, the disaster recovery device of the call center further includes: a main call center monitoring module and a switching module. Among them, the main call center monitoring module is used to monitor whether the operation status of the main call center returns to normal after starting the backup call center; the switching module is used to send a stop to the backup call center when the operation status of the main call center returns to normal Run the instruction to stop the backup call center and restart the main call center after the backup call center stops running.

It is not difficult to find from the above description that the disaster recovery device of the call center provided in this embodiment, after detecting that the operating state of the main call center returns to normal, the virtual address does not directly drift back to the main call center, but first to the backup call center Send a stop operation command, and restart the main call center after the backup call center stops running according to the stop operation command, so that the virtual address drifts back to the main call center. This switching method effectively prevents the brain split caused by the virtual address switching back and forth. Ensure the normal processing of subsequent call services.

In addition, in practical applications, in order to facilitate the management personnel of the call center system to be able to know the situation of the call center sending abnormalities in time, and quickly and accurately locate the fault problem, when the operating state of the main call center is detected abnormal, start backup After the call center, you can also generate warning information based on the collected log information of the main call center, and then notify the administrator of the main call center according to the preset warning notification method.

In addition, for technical details that are not described in detail in the device embodiments, reference may be made to the disaster recovery method of the call center provided in any embodiment of the present application, and details are not described herein again.

Claims

A disaster recovery method for a call center, characterized in that the method includes the following steps:

Start the main call center and collect the log information generated during the operation of the main call center;

Perform real-time analysis on the log information to determine whether the operation status of the main call center is normal;

If the operation status of the main call center is normal, when the pre-assigned virtual address receives the call service, the call service is transferred to the main call through the physical address of the main call center according to the first correspondence Center, the first correspondence is the correspondence between the virtual address and the physical address of the main call center;

If the operation status of the main call center is abnormal, a backup call center is started, and when the call service is received at the virtual address, the call service is passed through the physical address of the backup call center according to the second correspondence relationship When transferred to the backup call center, the second correspondence is the correspondence between the virtual address and the physical address of the backup call center.
The method according to claim 1, wherein before the step of activating the main call center, the method further comprises the following steps:

Build the first call center and the second call center according to the call center business requirements;

Any one of the first call center and the second call center is designated as the main call center, and the other is designated as the backup call center.
The method according to claim 2, characterized in that, any one of the first call center and the second call center is designated as the primary call center, and the other is designated as the backup call center. The steps include:

Collecting, according to preset monitoring indicators, the first monitoring indicator data provided by the first call center and the second monitoring indicator data provided by the second call center in time intervals;

Separately analyzing the first monitoring indicator data in each period and the second monitoring indicator data in each period to obtain the monitoring indicator change rate corresponding to the first call center and the monitoring indicator change rate corresponding to the second call center;

According to a preset main call center selection rule, the first call center whose monitoring index change rate meets the main call center selection rule is designated as the main call center, and the second call center is designated as the backup call Center; or, the second call center whose change rate of the monitoring index meets the selection rule of the main call center is designated as the main call center, and the first call center is designated as the backup call center.
The method according to claim 3, wherein the analysis of the first monitoring indicator data in each period and the second monitoring indicator data in each period respectively analyzes the rate of change of the monitoring indicator corresponding to the first call center The step of monitoring the change rate of the indicator corresponding to the second call center includes:

Based on the preset analysis model, analyze the first monitoring indicator data in each period and the second monitoring indicator data in each period respectively to obtain the monitoring indicator change rate corresponding to the first call center and the second call center Rate of change of monitoring indicators.
The method according to claim 4, characterized in that, before the step of analyzing the first monitoring indicator data in each period and the second monitoring indicator data in each period based on a preset analysis model, the method It also includes the following steps:

Based on the deep machine learning method, construct the analysis model;

Wherein, the step of constructing the analysis model based on deep machine learning method includes:

Construct the first training model according to the sample data;

Split the initial network layer of each hidden layer in the first training model into at least two sub-network layers according to preset layering standards;

Using the unsupervised training method that rises from the bottom, the hidden layer connected to the input layer in the first training model is used as a starting point, and the hidden layer connected to the output layer in the first training model is used as an end point. A sub-network layer in each hidden layer in the training model is trained to obtain a second training model;

Using a top-down supervised training method, the hidden layer connected to the output layer in the second training model is used as the starting point, and the hidden layer connected to the input layer in the second training model is used as the end point. The sub-network layer in each hidden layer in the second training model is trained to obtain the analysis model.
The method according to any one of claims 1 to 5, wherein after the step of starting a backup call center, the method further comprises the following steps:

Monitoring whether the running state of the main call center returns to normal;

If the operation state of the main call center returns to normal, send a stop operation instruction to the backup call center to stop the backup call center, and restart the main call after the backup call center stops running center.
The method according to any one of claims 1 to 5, wherein after the step of starting a backup call center, the method further comprises the following steps:

Generate warning information according to the log information;

According to a preset warning notification method, the warning information is notified to the manager of the main call center.
A disaster recovery device for a call center, characterized in that the device includes:

The log information collection module is used to start the main call center and collect the log information generated during the operation of the main call center;

The log information analysis module is used for real-time analysis of the log information to determine whether the operation status of the main call center is normal;

The first call service transfer module is used to pass the call service through the physical of the main call center according to the first correspondence when the call service is received in a pre-assigned virtual address in the normal operation state of the main call center The address is transferred to the primary call center, and the first correspondence is the correspondence between the virtual address and the physical address of the primary call center;

The second call service transfer module is used to start a backup call center when the operation state of the main call center is abnormal, and when the virtual address receives the call service, according to the second correspondence, the call The service is transferred to the backup call center through the physical address of the backup call center, and the second correspondence is the correspondence between the virtual address and the physical address of the backup call center.
A disaster recovery device for a call center, characterized in that the device includes: a memory, a processor, and computer-readable storage instructions stored on the memory and operable on the processor, the computer-readable When the instruction is executed by the processor, the processor is caused to perform the following steps:

Start the main call center and collect the log information generated during the operation of the main call center;

Perform real-time analysis on the log information to determine whether the operation status of the main call center is normal;

If the operation status of the main call center is normal, when the pre-assigned virtual address receives the call service, the call service is transferred to the main call through the physical address of the main call center according to the first correspondence Center, the first correspondence is the correspondence between the virtual address and the physical address of the main call center;

If the operation status of the main call center is abnormal, a backup call center is started, and when the call service is received at the virtual address, the call service is passed through the physical address of the backup call center according to the second correspondence relationship When transferred to the backup call center, the second correspondence is the correspondence between the virtual address and the physical address of the backup call center.
The disaster recovery equipment of the call center according to claim 9, wherein before the step of starting the main call center, the processor is further configured to perform the following steps:

Build the first call center and the second call center according to the call center business requirements;

Any one of the first call center and the second call center is designated as the main call center, and the other is designated as the backup call center.
The disaster recovery equipment of the call center according to claim 10, wherein any one of the first call center and the second call center is designated as the main call center, and the other is designated as the The steps to back up the call center include:

Collecting, according to preset monitoring indicators, the first monitoring indicator data provided by the first call center and the second monitoring indicator data provided by the second call center in time intervals;

Separately analyzing the first monitoring indicator data in each period and the second monitoring indicator data in each period to obtain the monitoring indicator change rate corresponding to the first call center and the monitoring indicator change rate corresponding to the second call center;

According to a preset main call center selection rule, the first call center whose monitoring index change rate meets the main call center selection rule is designated as the main call center, and the second call center is designated as the backup call Center; or, the second call center whose change rate of the monitoring index meets the selection rule of the main call center is designated as the main call center, and the first call center is designated as the backup call center.
The disaster recovery equipment of the call center according to claim 11, wherein the first monitoring indicator data in each period and the second monitoring indicator data in each period are analyzed respectively to obtain the correspondence of the first call center The steps of the monitoring index change rate and the monitoring index change rate corresponding to the second call center include:

Based on the preset analysis model, analyze the first monitoring indicator data in each period and the second monitoring indicator data in each period respectively to obtain the monitoring indicator change rate corresponding to the first call center and the second call center Rate of change of monitoring indicators.
The disaster recovery equipment of the call center according to claim 12, wherein the step of analyzing the first monitoring indicator data in each period and the second monitoring indicator data in each period based on a preset analysis model Previously, the processor was also used to perform the following steps:

Based on the deep machine learning method, construct the analysis model;

Wherein, the step of constructing the analysis model based on deep machine learning method includes:

Construct the first training model according to the sample data;

Split the initial network layer of each hidden layer in the first training model into at least two sub-network layers according to preset layering standards;

Using the unsupervised training method that rises from the bottom, the hidden layer connected to the input layer in the first training model is used as a starting point, and the hidden layer connected to the output layer in the first training model is used as an end point. A sub-network layer in each hidden layer in the training model is trained to obtain a second training model;

Using a top-down supervised training method, the hidden layer connected to the output layer in the second training model is used as the starting point, and the hidden layer connected to the input layer in the second training model is used as the end point. The sub-network layer in each hidden layer in the second training model is trained to obtain the analysis model.
The disaster recovery equipment of a call center according to any one of claims 9 to 13, wherein after the step of starting a backup call center, the processor is further configured to perform the following steps:

Monitoring whether the running state of the main call center returns to normal;

If the operation state of the main call center returns to normal, send a stop operation instruction to the backup call center to stop the backup call center, and restart the main call after the backup call center stops running center.
The disaster recovery equipment of a call center according to any one of claims 9 to 13, wherein after the step of starting a backup call center, the processor is further configured to perform the following steps:

Generate warning information according to the log information;

According to a preset warning notification method, the warning information is notified to the manager of the main call center.
A computer-readable storage medium, characterized in that computer-readable instructions are stored on the computer-readable storage medium, and when the computer-readable instructions are executed by a processor, the following steps are implemented:

Start the main call center and collect the log information generated during the operation of the main call center;

Perform real-time analysis on the log information to determine whether the operation status of the main call center is normal;

If the operation status of the main call center is normal, when the pre-assigned virtual address receives the call service, the call service is transferred to the main call through the physical address of the main call center according to the first correspondence Center, the first correspondence is the correspondence between the virtual address and the physical address of the main call center;

If the operation status of the main call center is abnormal, a backup call center is started, and when the call service is received at the virtual address, the call service is passed through the physical address of the backup call center according to the second correspondence relationship When transferred to the backup call center, the second correspondence is the correspondence between the virtual address and the physical address of the backup call center.
The computer-readable storage medium of claim 16, wherein before the step of activating the main call center, the processor is further configured to perform the following steps:

Build the first call center and the second call center according to the call center business requirements;

Any one of the first call center and the second call center is designated as the main call center, and the other is designated as the backup call center.
The computer-readable storage medium of claim 17, wherein any one of the first call center and the second call center is designated as the main call center, and the other is designated as the The steps to backup the call center include:

Collecting, according to preset monitoring indicators, the first monitoring indicator data provided by the first call center and the second monitoring indicator data provided by the second call center in time intervals;

Separately analyzing the first monitoring indicator data in each period and the second monitoring indicator data in each period to obtain the monitoring indicator change rate corresponding to the first call center and the monitoring indicator change rate corresponding to the second call center;

According to a preset main call center selection rule, the first call center whose monitoring index change rate meets the main call center selection rule is designated as the main call center, and the second call center is designated as the backup call Center; or, the second call center whose change rate of the monitoring index meets the selection rule of the main call center is designated as the main call center, and the first call center is designated as the backup call center.
The computer-readable storage medium according to claim 18, wherein the first monitoring indicator data in each period and the second monitoring indicator data in each period are analyzed to obtain the corresponding The steps of monitoring the index change rate and the monitoring index change rate corresponding to the second call center include:

Based on the preset analysis model, analyze the first monitoring indicator data in each period and the second monitoring indicator data in each period respectively to obtain the monitoring indicator change rate corresponding to the first call center and the second call center Rate of change of monitoring indicators.
The computer readable storage medium according to claim 19, wherein the step of analyzing the first monitoring indicator data and the second monitoring indicator data in each period based on a preset analysis model before the step , The processor is also used to perform the following steps:

Based on the deep machine learning method, construct the analysis model;

Wherein, the step of constructing the analysis model based on deep machine learning method includes:

Construct the first training model according to the sample data;

Split the initial network layer of each hidden layer in the first training model into at least two sub-network layers according to preset layering standards;

Using the unsupervised training method that rises from the bottom, the hidden layer connected to the input layer in the first training model is used as the starting point, and the hidden layer connected to the output layer in the first training model is used as the end point. A sub-network layer in each hidden layer in the training model is trained to obtain the second training model;

Using a top-down supervised training method, the hidden layer connected to the output layer in the second training model is used as the starting point, and the hidden layer connected to the input layer in the second training model is used as the end point. The sub-network layer in each hidden layer in the second training model is trained to obtain the analysis model. The