CN111970321A

CN111970321A - Method, system, storage medium and electronic device for connection management in cluster

Info

Publication number: CN111970321A
Application number: CN202010626567.5A
Authority: CN
Inventors: 苏进文
Original assignee: Beijing Mininglamp Software System Co ltd
Current assignee: Beijing Mininglamp Software System Co ltd
Priority date: 2020-07-01
Filing date: 2020-07-01
Publication date: 2020-11-20

Abstract

The embodiment of the application discloses a method, a system, a storage medium and an electronic device for managing connection in a cluster. The method comprises the following steps: if the client and the server adopt a non-blocking input and output NIO mode for communication based on a Transmission Control Protocol (TCP), acquiring a binding relationship between the client and a server channel after receiving a communication request initiated by the client, wherein the binding relationship is established by the server; and establishing a channel for the client according to the binding relationship.

Description

Method, system, storage medium and electronic device for connection management in cluster

Technical Field

The present disclosure relates to the field of information processing, and in particular, to a method, a system, a storage medium, and an electronic device for managing connections in a cluster.

Background

In a TCP (Transmission Control Protocol) -based cluster communication scenario, an NIO (New I/O, non-blocking input/output) mode may be used for communication, and if a client does not bind a channel, when a plurality of requests are sent, the client cannot find a corresponding client when serving as a server in a cluster to monitor the corresponding channel, thereby causing communication failure.

Fig. 1 is a schematic diagram illustrating communication between a client and a server in the related art. As shown in fig. 1, a BIO (Blocking I/O) mode is used for communication, a server deploys different servers for a same module, then each server starts a different port to monitor a request of a client, and then the client remotely connects a corresponding server by locally configuring a server port to realize one-to-one communication. Fig. 2 is a schematic diagram of a communication flow in the communication diagram shown in fig. 1. As shown in fig. 2, as long as the port of the server is not changed, the communication between the two parties can always be kept smooth regardless of the change of the ip address of the client.

Fig. 3 is a communication flow chart of a client and a server in the related art. As shown in fig. 3, the client and the server use UDP (User Datagram Protocol) instead of TCP, and the execution steps of the server in this scheme mainly include processes of establishing a socket, binding the socket to an address structure, reading and writing data, and closing the socket. The execution steps of the client mainly comprise the processes of socket establishment, data reading and writing and socket closing. The main difference between the two flows of the server and the client is the binding (bind ()) function to the address, and the client can not bind the address and the port.

In practical application, both the above two schemes are not suitable for the communication scenario between the client and the server in the cluster communication system.

Disclosure of Invention

In order to solve any technical problem, embodiments of the present application provide a method, a system, a storage medium, and an electronic device for connection management in a cluster.

To achieve the purpose of the embodiment of the present application, an embodiment of the present application provides a method for managing connections in a cluster, including:

if the client and the server adopt a non-blocking input and output NIO mode for communication based on a Transmission Control Protocol (TCP), acquiring a binding relationship between the client and a server channel after receiving a communication request initiated by the client, wherein the binding relationship is established by the server;

and establishing a channel for the client according to the binding relationship.

A system for connection management in a cluster, comprising:

a connection management device for performing the method described above;

and the server is used for communicating with the client through the channel determined by the connection management device.

A storage medium having a computer program stored therein, wherein the computer program is arranged to perform the method as described above when executed.

An electronic device comprising a memory having a computer program stored therein and a processor arranged to execute the computer program to perform the method as described above.

One of the above technical solutions has the following advantages or beneficial effects:

if the client and the server communicate in the NIO mode based on the TCP protocol, the binding relationship between the client and the channel of the server is obtained after a communication request initiated by the client is received, wherein the binding relationship is established by the server, and the channel is established for the client according to the binding relationship, so that the maintenance cost of establishing the channel by the client is reduced.

Additional features and advantages of the embodiments of the application will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by the practice of the embodiments of the application. The objectives and other advantages of the embodiments of the application may be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.

Drawings

The accompanying drawings are included to provide a further understanding of the embodiments of the present application and are incorporated in and constitute a part of this specification, illustrate embodiments of the present application and together with the examples of the embodiments of the present application do not constitute a limitation of the embodiments of the present application.

FIG. 1 is a diagram illustrating a communication between a client and a server in the related art;

FIG. 2 is a schematic diagram of a communication flow in the communication diagram of FIG. 1;

FIG. 3 is a flowchart illustrating a communication between a client and a server in the related art;

fig. 4 is a flowchart of a method for managing connections in a cluster according to an embodiment of the present application;

fig. 5 is a structural diagram of a connection management system in a cluster according to an embodiment of the present application;

fig. 6 is a schematic diagram of a trunking communication system provided in the embodiment of the present application.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present application more apparent, the embodiments of the present application will be described in detail below with reference to the accompanying drawings. It should be noted that, in the embodiments of the present application, features in the embodiments and the examples may be arbitrarily combined with each other without conflict.

In the process of implementing the present application, the inventor conducts technical analysis on the related art, and finds that the related art has at least the following problems, including:

the communication mechanism shown in fig. 1 has the following problems:

1. the communication belongs to a BIO communication mode and does not belong to an NIO mode, so that the number of requests of the user quantity is greatly limited, and the concurrency is very low;

2. because the communication between the client and the server is completely carried out through the fixed port, once the server of the node crashes, the communication of the client is completely cut off, no disaster tolerance effect exists, and the cluster CAP principle is not followed, wherein the CAP principle comprises Consistency (Consistency), Availability (Availability) and Partition tolerance (Partition tolerance).

3. Since each client needs to select a fixed server and port from the local configuration file for binding when sending a request, a large number of request exceptions are caused once the server cluster is changed and not synchronized to the client, and the maintenance cost of this mode is high.

The communication mechanism shown in fig. 3 has the following problems:

1. because the UDP protocol does not provide mechanisms such as confirmation, sequence numbers, overtime retransmission and the like, the UDP datagram can be copied and reordered in the network, the datagram can not reach the final destination, the sequence of each datagram can not be ensured, each datagram can not reach once, and the communication transmission is lack of reliability;

2. since each UDP datagram has a length, if a datagram arrives at the destination correctly, the length of the datagram will be delivered to the recipient along with the data. Whereas TCP is a byte stream protocol without any (on-protocol) record boundaries.

3. The client and the server based on the UDP protocol do not need to have a long-term relationship before sending the datagram, and a connection establishing process through handshaking is not needed before sending the datagram by the UDP, so that no connection exists between the client and the server.

Based on the above analysis, the embodiments of the present application provide the following solutions, including:

fig. 4 is a flowchart of a method for managing connections in a cluster according to an embodiment of the present application. As shown in fig. 4, the method shown in fig. 4 includes:

step 401, if the client and the server communicate in the NIO mode based on the TCP protocol, after receiving a communication request initiated by the client, obtaining a binding relationship between the client and a server channel, wherein the binding relationship is established by the server;

in an exemplary embodiment, a TCP protocol is used for communication in a cluster, and in the related art, a client performs special processing on a channel to ensure normal connection of communication between two parties, but the binding relationship is maintained by a server, so that the maintenance cost of the client is reduced.

Step 402, establishing a channel for the client according to the binding relationship;

in an exemplary embodiment, a channel may be established with the client for communication according to the binding relationship established at the server.

According to the method provided by the embodiment of the application, if the client and the server communicate in the NIO mode based on the TCP protocol, after a communication request initiated by the client is received, the binding relationship between the client and the channel of the server is obtained, wherein the binding relationship is established by the server, the channel is established for the client according to the binding relationship, and the maintenance cost of the channel established by the client is reduced.

In an exemplary embodiment, the binding relationship is obtained by:

acquiring identification information of the client;

calculating a target server corresponding to the identification information according to a preset calculation strategy;

and establishing a binding relationship between the client and the target server.

The identification information of the client can be at least one of an IP address of the client, an MAC address of the client and identity information and an MAC address of a user logging in the client; the identification information can uniquely represent the client, so that the identity of the client can be more accurately distinguished, and a data basis is provided for a subsequent calculation target server.

In an exemplary embodiment, the calculating, according to a preset calculation policy, a target server corresponding to the identification information includes:

calculating the identification information to obtain a numerical value corresponding to the identification information;

taking the total number of the servers as a divisor and the numerical value as a dividend, and executing division operation to obtain a remainder corresponding to the identification information;

and determining the target server according to the remainder corresponding to the identification information.

Because the identification information can be a group of numerical values or a combination of a plurality of groups of data, the length of the identification information can be effectively shortened by calculating the identification information, the subsequent determination of the calculation amount required to be executed by the target server is facilitated, and the processing complexity is reduced.

The above calculation method may be to extract a numerical value of a preset number of bits in the identification information to form a new numerical value, taking a numerical value sequence of a certain identification information as 1234567890 as an example, and the preset positions are 5 th, 7 th and 9 th bits, then the numerical value sequence formed after extraction is 579, and as can be seen by comparison, the length of the calculated numerical value is shorter than the length of the numerical value of the identification information, so that the calculation is more convenient.

In an exemplary embodiment, the identification information is subjected to a hash calculation to obtain the value.

When the identification information includes a plurality of characters, the processing of the identification information can be completed by using hash calculation, and the versatility of the calculation processing is improved, wherein the plurality of characters include at least one of numbers, symbols, letters, and words.

In addition, when the hash calculation is used for processing the numerical value, the server can utilize the Current HashMap to maintain the binding relationship, and the management efficiency of the server is improved.

In an exemplary embodiment, after the establishing the binding relationship between the client and the target server, the method further includes:

and when detecting that the identification information of the client side is changed, reestablishing the binding relationship for the client side, and establishing a channel with the client side by using the reestablished binding relationship.

In the above exemplary embodiment, after receiving a communication request from a client, first determining whether identification information of the client changes, and if so, determining the changed identification information; and judging whether the changed identification information belongs to the information used for establishing the binding relationship, if the changed identification information belongs to the identification information used for establishing the binding relationship, reestablishing the binding relationship, and establishing a channel with the client by using the reestablished binding relationship so as to ensure the normal communication with the client, reduce the occurrence of data loss and improve the transmission efficiency.

Taking the identification information of the client as an IP address as an example, after the binding relationship is established by using the IP address of the client, if the IP address of the client changes, the binding relationship is established for the client again because the changed IP address may be different from the binding relationship determined by the IP address before the change, so as to ensure the standard uniformity of the establishment of the binding relationship and facilitate maintenance and management.

Fig. 5 is a structural diagram of a connection management system in a cluster according to an embodiment of the present application. As shown in fig. 5, the apparatus includes:

connection management means for performing any of the methods described above;

According to the system provided by the embodiment of the application, if the client and the server communicate in the NIO mode based on the TCP protocol, after a communication request initiated by the client is received, the binding relationship between the client and the channel of the server is obtained, wherein the binding relationship is established by the server, the channel is established for the client according to the binding relationship, and the maintenance cost of the channel established by the client is reduced.

In an exemplary embodiment, the connection management apparatus is deployed on a reverse proxy server on a server side.

The processing pressure of the server can be shared by the reverse proxy server to complete the operation.

In one exemplary embodiment, the number of the connection management apparatuses is at least two;

the system further comprises:

and the load balancing equipment is used for distributing the communication request to one of the at least two connection management devices according to a preset load balancing strategy after receiving the communication request of the client.

The load balancing equipment can distribute the communication request of the client to a plurality of connection management devices, realize the sharing of processing pressure and reduce the processing pressure of the connection management devices.

The following is a description of application examples provided in the embodiments of the present application:

when a server monitors a client request, the client sends the request to the server for the first time in a cluster, a server performs hash binding on a channel and a user ip, when the client sends the request to the cluster again, the request firstly passes through a cloud to reach an nginx cluster, then the nginx cluster performs hash processing on the client ip and finds out a corresponding server according to the number of the servers, then the server finds out the corresponding channel by using the ConcurrentHashMap of the server for communication, once the client ip is changed, the connection establishing process is repeated once, and data cannot be lost.

Fig. 6 is a schematic diagram of a trunking communication system provided in the embodiment of the present application. As shown in fig. 6, the SLB in the system implements load balancing, nginx performs iphhash calculation on the IP address of a client to determine that a client and a certain back-end web machine can establish a stable session, and the server records a binding relationship based on a convurrenthashmap manner to establish a session; wherein the execution steps include:

1. a client sends a request to an SLB cluster for the first time;

the SLB cluster is a server load balancing mechanism, and does not support a permanent and effective IPhash strategy, wherein the IPhash directs a certain ip request to the same back-end web machine, so that a client under the ip and the certain back-end web machine can establish a stable session.

2. After receiving a request sent by a client, the SLB cluster forwards the request in a polling/weighting mode to form a corresponding nginx cluster;

wherein nginx is a high-performance http and reverse proxy web server, and also provides mailbox service, and may be IMAP (Internet Mail Access Protocol), POP3(Post Office Protocol 3, version 3 of Post Office Protocol), or SMTP (Simple Mail Transfer Protocol) service

3. The nginx cluster receives the request forwarded by the SLB cluster, hash operation is carried out by using the ip of the client, and then the remainder is obtained according to the total number of the server clusters, so that a server with a corresponding number is found;

4. when a certain server in the server cluster receives a client request forwarded by nginx, performing first connection, binding a user and a channel by using ConcurrentHashMap, and establishing connection;

5. when the client sends the request or the heartbeat again, the request or the heartbeat passes through the SLB cluster and the NGINX cluster, but the corresponding server is found by performing hash operation according to the client ip after NGINX;

6. after a certain server is requested, retrieving corresponding user information from ConcurrentHashMap, and finding a corresponding channel because the user and the channel are bound previously, thereby ensuring smooth communication;

wherein, ConcurrentHashMap is a HashMap implementation with safe and high efficiency of one thread in Java;

7. if the client changes the ip under the condition of dial-up networking or network disconnection, due to the fact that corresponding heartbeat detection exists between the client and the server in the TCP protocol, once the ip cannot be connected, an automatic reconnection mechanism is set, so that the steps 1 to 4 are repeated, and communication is guaranteed to be smooth all the time.

And 8, if any one of the servers is hung, the dividend for remainder can be changed to solve the problem according to the IPhash of the client, and the client can always find the server corresponding to the channel.

According to the method provided by the embodiment of the application, when communication is carried out in a cluster based on a TCP protocol, the server end completes the binding relationship between the client and the channel so as to ensure the normal connection of the communication between the client and the server end, the problem that the maintenance cost of the client is increased due to the fact that the client carries out special processing on the channel in the related technology is solved, the maintenance cost is reduced, and meanwhile, the communication requirement in the cluster is met.

A storage medium having a computer program stored therein, wherein the computer program is arranged to perform the method of any of the above when executed.

An electronic device comprising a memory having a computer program stored therein and a processor arranged to run the computer program to perform the method of any of the above.

It will be understood by those of ordinary skill in the art that all or some of the steps of the methods, systems, functional modules/units in the devices disclosed above may be implemented as software, firmware, hardware, and suitable combinations thereof. In a hardware implementation, the division between functional modules/units mentioned in the above description does not necessarily correspond to the division of physical components; for example, one physical component may have multiple functions, or one function or step may be performed by several physical components in cooperation. Some or all of the components may be implemented as software executed by a processor, such as a digital signal processor or microprocessor, or as hardware, or as an integrated circuit, such as an application specific integrated circuit. Such software may be distributed on computer readable media, which may include computer storage media (or non-transitory media) and communication media (or transitory media). The term computer storage media includes volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data, as is well known to those of ordinary skill in the art. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, Digital Versatile Disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store the desired information and which can accessed by a computer. In addition, communication media typically embodies computer readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transport mechanism and includes any information delivery media as known to those skilled in the art.

Claims

1. A method for managing connections in a cluster, comprising:

2. The method of claim 1, wherein the binding relationship is obtained by:

acquiring identification information of the client;

3. The method according to claim 2, wherein the calculating the target server corresponding to the identification information according to a preset calculation policy includes:

4. The method of claim 3, wherein the identification information is hashed to obtain the value.

5. The method according to any one of claims 2 to 4, wherein after the establishing the binding relationship between the client and the target server, the method further comprises:

6. A system for connection management in a cluster, comprising:

connection management means for performing the method of any one of claims 1 to 5;

7. The system of claim 6, wherein the connection management device is deployed on a server-side reverse proxy server.

8. The system according to claim 6 or 7, characterized in that:

the number of the connection management devices is at least two;

the system further comprises:

9. A storage medium, in which a computer program is stored, wherein the computer program is arranged to perform the method of any of claims 1 to 5 when executed.

10. An electronic device comprising a memory and a processor, wherein the memory has stored therein a computer program, and wherein the processor is arranged to execute the computer program to perform the method of any of claims 1 to 5.