CN108924215B

CN108924215B - Service discovery processing method and device based on tree structure

Info

Publication number: CN108924215B
Application number: CN201810688306.9A
Authority: CN
Inventors: 朱旋; 段盼; 王苏文
Original assignee: Beijing SF Intra City Technology Co Ltd
Current assignee: Beijing SF Intra City Technology Co Ltd
Priority date: 2018-06-28
Filing date: 2018-06-28
Publication date: 2021-03-19
Anticipated expiration: 2038-06-28
Also published as: CN108924215A

Abstract

The embodiment of the invention discloses a service discovery processing method and a device based on a tree structure, wherein the method comprises the following steps: if the fact that the name data are transmitted into the local port is monitored, analyzing the name data to obtain a Key value Key corresponding to the name data, and reconstructing and storing a tree structure in the memory according to the Key corresponding to the name data; monitoring the operation of all keys in a preset path, and searching each instance of the tree structure according to the target Key if the target Key is judged to be changed according to the operation; and if the target instance corresponding to the target Key is found in the tree structure, updating the local subscription service corresponding to the target instance. By establishing a tree structure and adopting the tree structure to store the example, the data maintenance cost is greatly reduced, and the maintenance is easier when the levels are more; when the change of the target Key is monitored, the corresponding target Key is searched through the tree structure, the updating of the local subscription service is realized, and the problem of data consistency is solved with extremely low cost.

Description

Service discovery processing method and device based on tree structure

Technical Field

The embodiment of the invention relates to the technical field of communication, in particular to a service discovery processing method and device based on a tree structure.

Background

The service discovery is used for solving the decoupling of calling resources between services and maintaining the corresponding relation between the service name and the service instance list. When remote resource access exists, the IP ports of all the instances can be acquired in real time through the service names, and then access is carried out.

In the prior art, the data storage form is basically a mapping of a service name and an instance set. Such as defining a service name that is backed by a set of instance information. When the instance information needs to be acquired, acquiring a set of the instance information through the service name, and selecting a certain instance. In the existing service discovery implemented by the storage scheme based on the mapping between the service name and the instance set, if it is desired to implement the multi-level feature, it can only be implemented by establishing a service group and including several other service name corresponding instance sets, such as: all the instances of the machine room B under the service name A are obtained, only the service name and the corresponding instance set of the machine room B can be created, and then the service name A and the service name B are associated in a cascade relation, so that not only all the instances of the service can be obtained, but also the instances of the machine room B under the service can be obtained.

In the process of implementing the embodiment of the present invention, the inventor finds that the existing method needs to solve the problem of data consistency at a large cost, and when the hierarchy is more, the relationship between the groups becomes complicated, thereby greatly increasing the maintenance cost.

Disclosure of Invention

Because the existing method has the problems, the embodiment of the invention provides a service discovery processing method and device based on a tree structure.

In a first aspect, an embodiment of the present invention provides a service discovery processing method based on a tree structure, including:

if monitoring that name data are transmitted into a local port, analyzing the name data to obtain a Key value Key corresponding to the name data, and reconstructing and storing a tree structure in a memory according to the Key corresponding to the name data;

monitoring the operation of all keys in a preset path, and searching each instance of the tree structure according to a target Key if the target Key is judged to be changed according to the operation;

and if the target instance corresponding to the target Key is found in the tree structure, updating the local subscription service corresponding to the target instance.

Optionally, if the target instance corresponding to the target Key is found in the tree structure, after the local subscription service corresponding to the target instance is updated, the method further includes:

analyzing the intranet IP address of the current terminal, acquiring all instances of the current terminal from a named cluster server according to the intranet IP address, providing health state detection service for the IP port of each instance, and feeding back the detection result of the health state detection service to the named cluster server.

Optionally, the keys are named according to a hierarchy of the tree structure.

Optionally, the tree structure is divided according to an enterprise organization structure and a service type.

In a second aspect, an embodiment of the present invention further provides a service discovery processing apparatus based on a tree structure, including:

the data analysis module is used for analyzing the name data to obtain a Key value Key corresponding to the name data if the name data transmitted from the local port is monitored, and reconstructing and storing a tree structure in the memory according to the Key corresponding to the name data;

the instance searching module is used for monitoring the operation of all keys in a preset path, and searching each instance of the tree structure according to a target Key if the change of the target Key is judged and known according to the operation;

and the service updating module is used for updating the local subscription service corresponding to the target instance if the target instance corresponding to the target Key is found in the tree structure.

Optionally, the apparatus further comprises:

and the health detection module is used for analyzing the intranet IP address of the current terminal, acquiring all instances of the current terminal from the named cluster server according to the intranet IP address, providing health state detection service for the IP port of each instance, and feeding back the detection result of the health state detection service to the named cluster server.

Optionally, the keys are named according to a hierarchy of the tree structure.

In a third aspect, an embodiment of the present invention further provides an electronic device, including:

at least one processor; and

at least one memory communicatively coupled to the processor, wherein:

the memory stores program instructions executable by the processor, which when called by the processor are capable of performing the above-described methods.

In a fourth aspect, an embodiment of the present invention further provides a non-transitory computer-readable storage medium storing a computer program, which causes the computer to execute the above method.

According to the technical scheme, the tree structure is established, the tree structure is adopted to store the example, the data maintenance cost is greatly reduced, and the maintenance is easy when the number of levels is large; when the change of the target Key is monitored, the corresponding target Key is searched through the tree structure, the updating of the local subscription service is realized, and the problem of data consistency is solved with extremely low cost.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.

Fig. 1 is a schematic flowchart of a service discovery processing method based on a tree structure according to an embodiment of the present invention;

FIG. 2 is a schematic diagram of a tree structure according to an embodiment of the present invention;

fig. 3 is an interaction diagram of a service discovery processing system based on a tree structure according to an embodiment of the present invention;

fig. 4 is a schematic flowchart of a service discovery processing method based on a tree structure according to another embodiment of the present invention;

fig. 5 is a schematic structural diagram of a service discovery processing apparatus based on a tree structure according to an embodiment of the present invention;

fig. 6 is a logic block diagram of an electronic device according to an embodiment of the present invention.

Detailed Description

The following further describes embodiments of the present invention with reference to the accompanying drawings. The following examples are only for illustrating the technical solutions of the present invention more clearly, and the protection scope of the present invention is not limited thereby.

Fig. 1 is a flowchart illustrating a tree structure-based service discovery processing method provided in this embodiment, and includes:

s101, if the fact that the name data are transmitted into the local port is monitored, analyzing the name data to obtain a Key value Key corresponding to the name data, and reconstructing and storing a tree structure in a memory according to the Key corresponding to the name data.

Wherein the Key is named according to the hierarchy of the tree structure.

The tree structure is divided according to an enterprise organization structure and a service type.

Specifically, a local fixed port is monitored, and a service name resolution function is provided for other services deployed by the current terminal. Analyzing the incoming name data to judge which instances form the read instance set, reading a Key of a corresponding prefix of an ETCD (open source distributed Key value pair data storage system), and reconstructing and storing a tree structure in a memory according to the result.

It should be noted that the collection of instances in the stored tree structure is referred to as a local subscription service of the terminal.

S102, monitoring the operation of all keys in a preset path, and searching each instance of the tree structure according to a target Key if the target Key is judged to be changed according to the operation.

S103, if the target instance corresponding to the target Key is found in the tree structure, updating the local subscription service corresponding to the target instance.

Specifically, the present embodiment defines the service level, and takes fig. 2 as an example, the level to which the service instance belongs is divided from the highest level (subsidiary company) to the lowest level (machine room). Based on the tree structure, all the instance information below the node can be read through searching of one node, but the searching characteristic of the tree structure needs to be ensured on the data storage.

In a specific implementation process, as shown in fig. 1, an SFNS Naming Cluster is a set of distributed Cluster services, is deployed at a server side in a distributed manner, and stores and queries instance information according to a predetermined structure by providing an API for an external management platform and an application having service registration; the SFNS Naming Agent is a basic component which is deployed on each physical machine, is mainly connected with the SFNS Naming Cluster, monitors the change of service, updates local data in real time and monitors local instances. Each instance deployed on the physical machine (backend service a and backend service B in fig. 3) reads the list of instances corresponding to the service identifier that the SFNS Naming Agent wants to resolve.

The service identifier is not a simple service name, but is defined according to a requirement hierarchy and a certain rule. Such as:

service a. subsysa. company a. svc represents a list of gz machine room instances of service a under product line a under subsidiary a.

service a. subsysa. company a. svc represents a list of all machine room instances of service a under product line a under subsidiary a.

subsysa.company a.svc represents a list of all machine room instances of product line a under subsidiary a.

1, gz. service a. subsysa. company a. svc represents a list of examples of gz machine room example 1 for service a under product line a under subsidiary a only.

The tree structure can be adopted to define the height and the hierarchy according to different requirements without other limitations. In the aspect of data storage, Etcd is selected as the storage of service discovery data, and is the core in the SFNS Naming Cluster. And a Naming Admin back-end module is developed to control data in the Etcd to be written and modified according to rules, and the Naming Admin provides an internal API (application program interface) interface and receives node change requests from a management platform and an automatic registration module. The tree structure of the data stored in the Etcd is embodied on the Key. For example, there is a service named C under the product line B under the subsidiary company a, and the first instance is added to the GZ machine room of this service, then the storage in etcd is as follows:

key: /prefix/a (subsidiary name)/b (product line name)/c (service name)/gz (machine room name)/1 (example number)

VALUE：{“host”：10.1.1.1,“port”：8080,”status”：1,“enable”：true}

Based on such storage rules, it can be found that when each instance is added, a Key is correspondingly added in the Etcd. As in the example of fig. 2 under the same node, the Key prefixes stored in Etcd should be identical (/ prefix/a/b/c/gz /). After the Key contains the tree structure, it becomes simple to read instance information according to a hierarchy, for example: when all the aforementioned instances of the gz computer room are to be read, an instance list can be obtained as long as all keys with prefixes of/prefix/a/b/c/gz/are read; the method is used for reading the instance list of the whole product line, and only all keys with prefixes of prefix/a/b/are read, so that the complex problem of the multi-level reading service instance is simplified, and the method is more flexible.

Each initialized physical machine is connected with the SFNS Naming Cluster and has the capability of resolving the service identifier. In addition, the SFNS provides a standardized SDK for each back-end service, so that the back-end service can interact with the SFNS Naming Agent without cost, submit the subscribed service identification and analyze the corresponding real-time updated service instance.

For example, the name is: and reading all leaf node information under serviceA under subsysB under the servicec in the tree structure storage, namely, a Key with prefix of/prefix/compositc/subsysB/serviceA. A set of all instances under all rooms of serviceA may be obtained. The name is: and 1, gz, serviceA, subsysB, and company C, reading the information of the first leaf node under gz of serviceA under subsysB under company C in the tree structure storage, namely the Key of/prefix/serviceC/subsysB/serviceA/gz/1, and only obtaining a set containing one instance. All the read instance sets can be stored in the memory according to the tree structure relationship. When the next resolution request of a different service name gz.servicea.subsysb.componyc is requested, since the tree that once resolved servicea.subsysb.componyc exists in the memory and gz.servicea.subsysb.componyc is its subtree, it can be returned directly from the memory data without pulling the server.

Compared with the service discovery of set association, the method provided by the embodiment can acquire different dimensional sets of different services as required without defining a plurality of sets and then performing set association, and the acquired content of the service instance set is directly extracted semantically, so that research and development, operation and maintenance are facilitated.

In the embodiment, in the change of the physical machine monitoring the subscribed service, the implementation is simple, and the processing efficiency is improved. Such as: in the case that an instance is included in multiple service sets, only one change is monitored, each Naming-Agent only needs to update the information of one leaf node at most 1 time, and all subscribed services including the information of the instance are updated. However, the multi-set association method discovered by other services requires that each subscription service containing the update instance triggers a change. Firstly, this may require a complex design on the interaction protocol of the long connection, for example, the long connection informs the server in the process of establishing the connection to monitor which subscribed service sets, and secondly, how to ensure all the subscribed sets containing the change instance, all updates at one time, and ensure data consistency. And the variation of the set is not equal to that of the examples, so that the examples in the set need to be subjected to differential comparison or complete coverage. Consider a set containing all instances of the entire company, which triggers a change whenever there is a change in an instance, and each physical machine needs to do a differential comparison or full coverage.

In the embodiment, the tree structure is established, and the tree structure is adopted to store the example, so that the data maintenance cost is greatly reduced, and the maintenance is easy when the number of levels is large; when the change of the target Key is monitored, the corresponding target Key is searched through the tree structure, the updating of the local subscription service is realized, and the problem of data consistency is solved with extremely low cost.

Further, on the basis of the above embodiment of the method, after S103, the method further includes:

s104, analyzing the intranet IP address of the current terminal, acquiring all instances of the current terminal from the named cluster server according to the intranet IP address, providing health state detection service for the IP port of each instance, and feeding back the detection result of the health state detection service to the named cluster server.

In particular, based on the API function and data storage of the SFNS name Cluster, a non-centralized service discovery service can be implemented. As shown in FIG. 2, on each physical machine, a component named SFNS Naming-Agent is deployed.

The work of the Naming-Agent is divided into 3 major parts:

and starting a server by the thread, monitoring a local fixed port, and providing a service name resolution function for other services deployed by the current physical machine. And analyzing the incoming name to judge which instances form the read instance set, reading a Key of the ETCD corresponding to the prefix, and rebuilding and storing the result in a memory, wherein the stored tree instance set is called a service subscribed by the physical machine.

And establishing long connection with the Etcd, and monitoring the write-in, change and deletion operations of all keys with root paths/prefix. And if the changed instance belongs to the locally subscribed service, updating the instance data of the corresponding path in the memory. If the node does not exist in the number in the memory, the number is ignored.

Analyzing the intranet IP address of the current physical machine, acquiring all instance information of the current physical machine from the SFNS Naming Cluster based on the IP, sniffing the health state of the IP PORT of each instance, and reporting the state back to the SFNS Naming Cluster.

The convenience of service discovery for this tree-structured implementation is illustrated in an overall scenario, as shown in fig. 4:

when a flow accesses the access layer service through the domain name, the request reverse proxy is carried out through mapping of the domain name and the service identifier and analyzing an instance list corresponding to the service identifier.

The request reaches a back-end service A, and the back-end service A needs to remotely access a back-end service B besides completing self service logic; the self business logic executed by the back-end service B needs to access the back-end storage service to write or inquire data. In the whole process, whether the reverse proxy or remote access of other services, the instance address information, namely IP and port, of the downstream service is required to be known.

Assuming that when the management platform creates these several services, the back-end service a is named as backservice _ a, the back-end service B is backservice _ B, and the back-end storage service is dbservice _ d, if they belong to the product line subsys _ s of the company _ c:

when the access layer performs reverse proxy on the backend service a, the access layer only needs to configure the service identifier as backservice _ a.

When the backend service a remotely calls the backend service B, the same process is performed, but if the backend service a wants to access the backend service B with the computer room, the service identifier is only configured according to gz.

If the back-end service is a non-open service, only the machine where the back-end service in the product line is located can be accessed, and when IP white list authorization is carried out, only subsys _ s _ company _ c.svc needs to be configured, and all the machines where the service is located under the product can be opened.

The storage service shown in fig. 4 also needs to perform IP authorization, but only opens for the service B, and just needs to configure backservice _ B.

Therefore, although a plurality of hierarchies are used in various scenes to acquire different numbers of instance sets to meet the requirements, only one piece of so-called service group information needs to be maintained on the aspect of service discovery management, which is the convenience brought by realizing service discovery according to a tree structure. The realization is simple, and the structure is clear, reduces fortune dimension cost.

Fig. 5 is a schematic structural diagram illustrating a tree structure based service discovery processing apparatus according to this embodiment, where the apparatus includes: a data analysis module 501, an instance search module 502, and a service update module 503, wherein:

the data analysis module 501 is configured to, if it is monitored that name data is transmitted from a local port, analyze the name data to obtain a Key value Key corresponding to the name data, and reconstruct and store a tree structure in a memory according to the Key corresponding to the name data;

the instance searching module 502 is configured to monitor operations of all keys in a preset path, and if a target Key is determined to be changed according to the operations, search each instance of the tree structure according to the target Key;

the service updating module 503 is configured to update the local subscription service corresponding to the target instance if the target instance corresponding to the target Key is found in the tree structure.

Specifically, if monitoring that name data is transmitted into a local port, the data analysis module 501 analyzes the name data to obtain a Key value Key corresponding to the name data, and reconstructs and stores a tree structure in a memory according to the Key corresponding to the name data; the instance searching module 502 monitors the operations of all keys in a preset path, and if a target Key is judged to be changed according to the operations, searches for each instance of the tree structure according to the target Key; if the target instance corresponding to the target Key is found in the tree structure, the service updating module 503 updates the local subscription service corresponding to the target instance.

Further, on the basis of the above embodiment of the apparatus, the apparatus further comprises:

Further, on the basis of the above apparatus embodiment, the keys are named according to the hierarchy of the tree structure.

Further, on the basis of the above device embodiment, the tree structure is divided according to an enterprise organization structure and a service type.

The service processing apparatus based on the tree structure according to this embodiment may be configured to execute the foregoing method embodiments, and the principle and the technical effect are similar, which are not described herein again.

Referring to fig. 6, the electronic device includes: a processor (processor)601, a memory (memory)602, and a bus 603;

wherein the content of the first and second substances,

the processor 601 and the memory 602 communicate with each other through the bus 603;

the processor 601 is used for calling the program instructions in the memory 602 to execute the methods provided by the above-mentioned method embodiments.

The present embodiments disclose a computer program product comprising a computer program stored on a non-transitory computer readable storage medium, the computer program comprising program instructions which, when executed by a computer, enable the computer to perform the methods provided by the above-described method embodiments.

The present embodiments provide a non-transitory computer-readable storage medium storing computer instructions that cause the computer to perform the methods provided by the method embodiments described above.

The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.

Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. With this understanding in mind, the above-described technical solutions may be embodied in the form of a software product, which can be stored in a computer-readable storage medium such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in the embodiments or some parts of the embodiments.

It should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.

Claims

1. A service discovery processing method based on a tree structure is characterized by comprising the following steps:

if monitoring that name data are transmitted into a local port, analyzing the name data to obtain a Key value Key corresponding to the name data, and reconstructing and storing a tree structure in a memory according to the Key corresponding to the name data; the Key is named according to the hierarchy of the tree structure;

2. The method according to claim 1, wherein if the target instance corresponding to the target Key is found in the tree structure, after the local subscription service corresponding to the target instance is updated, the method further comprises:

3. The method according to claim 1 or 2, wherein the tree structure is partitioned according to enterprise organizational structure and service type.

4. A tree structure-based service discovery processing apparatus, comprising:

the data analysis module is used for analyzing the name data to obtain a Key value Key corresponding to the name data if the name data transmitted from the local port is monitored, and reconstructing and storing a tree structure in the memory according to the Key corresponding to the name data; the Key is named according to the hierarchy of the tree structure;

5. The apparatus of claim 4, further comprising:

6. The apparatus according to claim 4 or 5, wherein the Key is named according to the hierarchy of the tree structure.

7. The apparatus of claim 4 or 5, wherein the tree structure is partitioned according to enterprise organizational structure and service type.

8. An electronic device, comprising:

at least one processor; and

at least one memory communicatively coupled to the processor, wherein:

the memory stores program instructions executable by the processor, the processor invoking the program instructions to perform the method of any of claims 1 to 3.

9. A non-transitory computer-readable storage medium storing a computer program that causes a computer to perform the method according to any one of claims 1 to 3.