HK1128564B

HK1128564B - A distributed master and standby managing method and system based on the network element

Info

Publication number: HK1128564B
Application number: HK09106261.0A
Authority: HK
Inventors: 江霞
Original assignee: 中兴通讯股份有限公司
Filing date: 2006-07-28
Publication date: 2011-12-09

Description

Method and system for distributed main/standby management based on network element

Technical Field

The present invention relates to a method for master/slave management in a network management system in the field of optical communication, and particularly to a method and a system for distributed master/slave management based on network elements.

Background

At present, in a network management system in the communication field, when a centralized network management system operates, if the network management system is abnormal, the network management system is unavailable, and the reliability is low. In order to solve the problem, a scheme of master and standby network management exists in a communication industry system, that is, two sets of network management work simultaneously, one master and one standby or one master and multiple standby, the master network management collects historical data, and manages network elements, and the master and standby network management method is usually based on network system management.

The existing master/slave network management method generally has two use modes:

(1) one way is based on upper database mirroring, i.e. the data of the primary network manager is completely mirrored by the secondary network manager, thus ensuring that all data will be retained when the primary network manager fails to operate. Referring to fig. 1, a primary server serves as a primary network manager to manage network elements, and a secondary server serving as a secondary network manager to completely mirror data of the primary server, so that all data can be recovered from the secondary server when the primary server is abnormal and fails to operate.

(2) Another method is a thermal monitoring method, because there is a master Manager (Manager) when managing network elements, in order to enhance the reliability of network management, there needs to be one or more secondary managers, and the data of the master and secondary managers are synchronized, and in case of failure of the master Manager, the secondary Manager can be switched to the master Manager management network management system. Referring to fig. 2, the primary server serves as a primary manager to manage the network element of the system, and the secondary server serves as a secondary manager to also monitor and manage the network element, the primary and secondary servers perform data synchronization, and when the primary server fails or operates abnormally, the secondary manager is directly switched to serve as the primary manager of the network element to manage the network element.

However, the conventional master and slave network management systems as shown in fig. 1 and 2 have the following problems:

(1) the main pipe is overloaded. When the system runs, all network element data of the network management system are interacted with the main network, so that the load of the main network is overlarge, and potential safety hazards exist.

(2) In the system, because the primary and secondary servers only adopt one access network element for access and the network elements are in one-way communication, when the network element ECC (Embedded Control Channel) is interrupted, some network elements are in monitoring failure. For example, in fig. 1 or fig. 2, when ECC is disconnected between the network element 11 and the network element 12, both the primary and secondary network managers can only manage the network element 11, and cannot manage any network element behind the network element 11.

Therefore, a main/standby management method capable of solving the defects that the load of a main network manager is too large, part of network element monitoring fails when an ECC is interrupted, the system operation reliability is low, and the like is needed.

Disclosure of Invention

The technical problem to be solved by the present invention is to provide a method and a system for distributed master/slave management based on network elements, so as to share the load of master/slave network managers and improve the reliability of network management.

The invention provides a method for distributed main/standby management based on network elements, which comprises the following steps:

(a) connecting all network elements, configuring the routing of the network elements into bidirectional intercommunication, and configuring the management authority of a server A and a server B in the network elements, wherein one part of the network elements configures the server A as a main station and configures the server B as an auxiliary station, and the other part of the network elements configures the server B as the main station and configures the server A as the auxiliary station;

(b) configuring a return strategy of a main station in two servers, wherein the two servers are connected with each other and are respectively connected to an access network element which configures the two servers as the main station, and after the two servers work, the two servers are used as the main station to manage the respectively allocated partial network elements and are used as secondary stations to monitor the other partial network elements;

(c) if the communication between the network element and the server configured as the main station is interrupted, the network element is switched, and the other server configured as the auxiliary station is used as the main station to manage the network element;

(d) and when the communication between the network element and the server configured as the master station is recovered, recovering the management of the network elements by the server as the master station according to the return strategy.

Further, the method can also have the following characteristics:

the step of connecting all network elements and configuring the routes of the network elements as bidirectional intercommunication in the step (a) includes:

connecting all network elements in series, so that the network element embedded control channels are intercommunicated;

the routing of each network element is configured for bi-directional interworking.

The step of configuring the management authority of the two servers in the network element in the step (a) comprises:

according to the principle of load balance of two servers, dividing all network elements into two parts in a balanced manner;

a part of network elements configure a server A as a primary station and configure a server B as a secondary station;

the other part of the network elements configure the server B as a primary station and the server A as a secondary station.

When the server in the step (b) is used as a master station to manage and configure the server as a network element of the master station, acquiring historical data of the network element, and performing corresponding configuration query on the network element; and when the server is used as a secondary station to manage and configure the secondary station as the network element of the primary station, inquiring and monitoring the network element.

And (c) performing data synchronization between the servers configured as the primary station and the secondary station during operation in the step (b), and maintaining the data synchronization between the primary station and the secondary station by performing database synchronization, wherein the historical data in the primary station and the historical data in the secondary station are all full.

The step (c) comprises:

when the network element detects that the communication with a server configured as a main station is interrupted, the network element informs the server configured as a secondary station of the network element to serve as the main station to manage the network element;

or when the server configured as the secondary station by the network element detects that the server configured as the primary station by the network element is abnormal, the secondary station server actively requests the permission of the primary station from the network element, and the network element determines whether to switch the secondary station to manage the network element as the primary station according to the state of the current primary station server.

The return strategy in the step (b) is configured to be in a master return mode or a non-return mode,

if the network element is in the master station return mode, when the communication between the network element and the server configured as the master station is recovered in the step (d), recovering the server configured as the master station to manage the network element, switching the current master station as the secondary station, and synchronizing data during the interruption period by the recovered master station server;

if the master station is in a non-return mode, in the step (d), when the communication between the network element and the server configured as the master station is recovered, the server configured as the master station is recovered to be used as a secondary station to manage the network element, data during the interruption period is synchronized, and the network element is accessed in a master station mode when the server of the current master station is disconnected.

In the step (B), the two servers, i.e., the server a and the server B, are respectively connected to a head-end network element and a tail-end network element in the series-connected network elements.

The invention also provides a distributed main and standby management system based on network elements, which comprises a plurality of network elements, a server A and a server B for managing the network elements, and is characterized in that:

all network elements are connected with each other, the routing configuration of the network elements is bidirectional intercommunication, and the management authority of a server A and a server B is configured in each network element, wherein one part of the network elements configures the server A as a main station and the server B as an auxiliary station, and the other part of the network elements configures the server B as the main station and the server A as the auxiliary station;

the network element is switched when the communication with the server configured as the main station is interrupted, and another server configured as the auxiliary station is used as the main station; after the communication with the server configured as the master station is recovered, switching the server back according to the return strategy;

the two servers are connected with each other and are configured with a return strategy of the master station, each server is respectively connected with an access network element which configures the server as the master station, manages the respectively allocated partial network elements as the master station during working, and monitors the other partial network elements as the secondary station;

and when the two servers are interrupted in communication and receive the notification of network element switching, the two servers change the main and auxiliary states of the servers for the network elements, and after the servers recover the communication with the network elements which configure the servers as the main station, the servers recover the management of the network elements according to the return strategy.

Further, the system can also have the following characteristics:

the server is used as a master station to manage and configure the network element as the master station, and the server collects historical data of the network element and carries out corresponding configuration inquiry on the network element; and when the network element configured as the secondary station is managed as the secondary station, inquiring and monitoring the network element.

And the two servers perform data synchronization when the primary station and the secondary station which are configured as network elements work normally, and maintain the data synchronization between the primary station and the secondary station by performing database synchronization, wherein the historical data in the primary station and the secondary station are full.

The server is used as a return strategy of the main station and is configured into a main station return mode or a non-return mode,

if the network element is in a master station return mode, when the communication between the network element and the server which is configured as the master station is recovered, the server which is configured as the master station is recovered to be used as the master station to manage the network element, the current master station is switched to be the secondary station, and meanwhile, the recovered master station server synchronizes data during the interruption period;

if the network element is in a master station non-return mode, when the communication between the network element and the server which is configured as the master station is recovered, the server which is configured as the master station is recovered to be used as a secondary station to manage the network element, data in the interruption period is synchronized, and the network element is accessed in a master station mode when the server of the current master station is disconnected.

The network element informs a server configured as a secondary station of the network element to serve as a main station to manage the network element when detecting that the communication with the server configured as the main station is interrupted; or when the server of the network element configured as the secondary station detects that the server of the network element configured as the primary station is abnormal, the secondary station server actively requests the permission of the primary station to the network element, and the network element determines whether to switch the secondary station to manage the network element as the primary station according to the state of the current primary station server.

All the network elements are connected in series, and the weapon is connected with the two servers, namely the server A and the server B, through the head end network element and the tail end network element.

When the two servers are used as the master station to manage and manage the network elements respectively, the managed network elements which are configured as the master station are balanced in number, and the load is balanced to manage corresponding network elements respectively.

By adopting the technical scheme of the invention, compared with the prior art, the network element needs to be configured with a bidirectional route, and both network managers are accessed with the network element; due to the adoption of the technical measures of distributed main and standby management based on the network elements, the load sharing of the main and auxiliary network managers is realized, and the reliability of the network management is improved. Furthermore, load balance can be achieved, and effects such as whole network management can still be achieved when local ECC is disconnected.

Brief description of the drawings

Fig. 1 is a schematic diagram of a conventional active/standby management technique based on an upper-level database mirror image;

FIG. 2 is a schematic diagram of a conventional thermal monitoring-based active/standby management technique;

fig. 3 is a schematic diagram of a distributed active/standby management method based on network elements according to an embodiment of the present invention;

fig. 4 is an example of interruption of a server 1 in the method for main/standby management based on network element distribution according to the technical solution of the present invention;

fig. 5 is an example of a network element ECC interrupt in the network element distributed primary/secondary management method according to the technical solution of the present invention.

Preferred embodiments of the invention

The technical solution of the present invention will be described in further detail with reference to the accompanying drawings and preferred embodiments.

The embodiment provides a method for distributed master/slave management based on a dense wavelength division multiplexing system network element, but the method is also used for network management systems in other types of optical communication networks.

The method is a distributed main and standby management system based on network elements, and is a dynamic protection mode of network management, wherein two network management servers are equal to each other, the two network management servers are connected with different gateway network elements, and each different network element has corresponding main and auxiliary authorities. All network elements in the system are ECC intercommunicated and are configured with bidirectional routes, each network element can be independently configured with a master manager (namely, a master network management server) and the functions and roles of the managers are equal, namely, one manager is a master station for one network element but can be a slave station for another network element.

Under normal conditions, all network elements can be managed by the main network management server and the standby network management server, the number of the network elements managed by the two network management servers as the main station is reasonably distributed, and load balance is achieved; when the main network management server is in a normal working state, the standby network management server corresponding to the network element is only in a monitoring state and cannot be configured. When the network is in an abnormal state, or the main network management server is abnormal, or the ECC of the network element is interrupted, the auxiliary network management server of the corresponding network element replaces the main network management server of the network element, and the auxiliary network management server has all the management functions of the main network management server so as to continuously manage part of the network or the whole network.

Referring to fig. 3, a distributed active/standby management method based on a network element according to this embodiment may include the following steps:

the first step is as follows: and connecting all network elements, configuring the routes of the network elements into bidirectional intercommunication, and configuring the management authorities of a server A and a server B in the network elements, wherein one part of the network elements configure the server A as a main station and the server B as an auxiliary station, and the other part of the network elements configure the server B as the main station and the server A as the auxiliary station.

Firstly, all network elements can be connected in series, and simultaneously, the routes of the network elements are configured, so that the network element routes are ensured to be communicated in two directions, and when one direction is not communicated, the network element routes can be switched to the routes in the other direction.

According to the actual system operation situation, the number of network elements managed by two servers as the master station needs to be planned, for example, the network element 11-1n is planned to use the server 1 as the master station, and the network element 21-2n is planned to use the server 2 as the master station. As for the network element, only the main network management collects historical data, and the historical data volume in the wavelength division network management system is large, the load balance of the two network management systems during normal operation is achieved as much as possible.

Then, according to the corresponding authority of the relevant manager in the planned and configured network element, configuring the network element 11-1n by taking the server 1 as a main station and the server 2 as a secondary station; the network element 21-2n is configured with the server 2 as the primary station and the server 1 as the secondary station.

The second step is that: the two servers are connected with each other and are respectively connected to an access network element which configures the two servers as the main station, and after the two servers work, the two servers are used as the main station to manage the respectively allocated partial network elements and used as the secondary station to monitor the other partial network elements.

Firstly, a return policy when the master station takes effect needs to be configured on two servers respectively, which includes:

the master station returns the mode, namely the master station becomes the corresponding master station immediately after being recovered;

the master station is in a non-return mode, namely the master station is recovered and then becomes a secondary station, and the secondary station becomes the master station when the master station applies for the primary station.

Then, the server 1 is connected to the server 2, the server 1 is accessed through the network element 11 as a master station for the network element 11-1n, and the server 2 is accessed through the network element 2n as a master station for the network element 21-2n, as shown in fig. 3, it is needless to say that both servers may be accessed to any other network element configured as a master station.

In the network management system shown in fig. 3, the server 1 is a main network manager of the network element 11. the corresponding server 2 is a sub-network manager of these network elements when in normal operation; the server 2 is simultaneously the primary network manager of the network element 21.. the network element 2n, and the server 1 is the secondary network manager thereof. These network elements ECC (Embedded Control Channel) are all interworking.

When the system normally operates, the main station manages the corresponding network elements, collects historical data of the network elements, and can perform operations such as corresponding configuration inquiry and the like on the network elements; the standby network manager can only inquire and monitor the network element, cannot perform configuration operation and does not collect historical data; the auxiliary network manager and the main network manager synchronize databases to ensure that data between the main network manager and the auxiliary network manager are completely consistent, and the historical data in the main network manager and the auxiliary network manager are full.

In different communication systems, the master station and the secondary station may have different administrative authorities and administrative functions according to different requirements of a specific system, but the secondary station must be able to monitor the network element.

The third step: and if the communication between the network element and the server configured as the primary station is interrupted, the network element is switched to use another server configured as the secondary station as the primary station to manage the network element.

In normal operation, the communication state between the network element and its master station needs to be detected:

when the network element detects that the heartbeat packet of the network management does not respond any more, the network element selects the network management configured as the secondary station as the primary station according to the administrator authority configured in advance, and sends a notice to inform the corresponding secondary station that the network management is switched to the management network element of the primary station, and the network element has all the authorities of the primary station;

or when one server detects that the other server is abnormal, the network element actively goes to request the master station permission, and the network element judges whether the corresponding master station permission can be given to the secondary station or not according to the currently managed network manager.

The fourth step: and when the communication between the network element and the server configured as the master station is recovered, recovering the management of the network elements by the server as the master station according to the return strategy.

When the communication between the network element and the main station is recovered, the network element is managed according to a preset return strategy, wherein the data of the main station needs to be synchronized with another auxiliary station replaced by the main station.

If the configured return strategy is the master station return mode, when the server and the network element establish a link, the network element detects that the master server is normal, and informs the secondary server to switch and recover to be used as the secondary station again; meanwhile, the main server synchronizes data during the interruption period; or the secondary server switches the primary station and the secondary station and synchronizes data when detecting that the primary server is normal.

If the configured return strategy is in a master station non-return mode, when the master station is recovered, the master station is recovered to be used as a secondary station management network element firstly, data in the interruption period is synchronized with a secondary server which is currently used as the master station, and the data is accessed in a master station mode to be used as the master station management network element when the secondary server is disconnected.

This step is substantially similar in technical implementation to the thermal monitoring master method.

Application example 1: server 1 and network element interruption connection failure.

Fig. 4 shows a schematic diagram of a system when a server 1 and a network element are disconnected and fail, according to the technical scheme of the active/standby management method of the present invention, the active/standby management when the server 1 and the network element are disconnected mainly includes the following steps:

(1) and connecting the network element 11.. network element 1n.. network element 21.. network element 2n, configuring a bidirectional route of the network element on the network element, configuring the authorities of the server 1 and the server 2 in the network element, and configuring the return strategy of the master station to be a master station return type.

Wherein the network element 11-1n is configured with the server 1 as the primary station and the server 2 as the secondary station, and the network element 21-2n is configured with the server 2 as the primary station and the server 1 as the secondary station.

(2) The method comprises the steps that a server 1 and a server 2 are connected into network elements, when the network elements normally run, the server 1 serves as a master station to manage network elements 11.. network elements 1n, and the server 2 serves as a slave station to manage the network elements; the server 2 manages the network elements 21.. network elements 2n as a master station, and the server 1 manages the network elements as a slave station; data synchronization is configured between the server 1 and the server 2, and the data of the two servers are completely consistent and are full data.

(3) Referring to fig. 4, when communication between the server 1 and the network element 11 fails, and at this time, when the communication mechanism of the network element 11.. times the network element 1n detects disconnection with the server 1, at this time, the network element 11.. times the network element 1n notifies the server 2 as its secondary station to switch to be the primary station of these network elements, at this time, the server 2 as the primary station manages all network elements on the network, collects all historical data, and the server 1 does not manage any network element.

(4) After the communication between the server 1 and the network element 11 is recovered, because the return strategy configured in the network element is a master station return type, the server 1 is switched to a master station of the network element 11.. No. 1n, and the server 2 is still switched to and recovered to a slave station of the network element 11.. No. 1n, wherein in the switching process, the server 1 needs to perform data synchronization with the server 2 to recover data during the communication interruption.

It can be seen from this example that, during normal operation, the system load is balanced, and when the server is disconnected, the system can still manage normally, and all data is not lost.

Application example 2: ECC break between network elements.

Fig. 5 shows a schematic diagram of a system when ECC is disconnected between network elements 11 and 12, according to the technical solution of the active/standby management method of the present invention, the active/standby management when ECC is disconnected between network elements mainly includes the following steps:

(3) Referring to fig. 5, when ECC is disconnected between the network element 11 and the network element 12, and at this time, the network element 12.. when the communication mechanism of the network element 2n detects that the network element 2n is disconnected from the server 1, at this time, the network element 12.. the network element 1n notifies the server 2 to switch to be the master station of these network elements, and at this time, the server 2 manages a part of the series of network elements 1 and all of the series of network elements 2 as the master station, collects all historical data, and the server 1 can only manage a part of the series of network elements 1 as the master station.

(4) After the ECC communication between the network element 11 and the network element 12 is recovered, since the master station return policy configured in the network element is a master station return mode, the server 1 is switched to be the master station of the network element 12.. network element 1n, and the server 2 is still the secondary station of the network element 11.. network element 1n.

It can be seen from this example that, during normal operation, the system load is balanced, and when the ECC is disconnected, the system can still manage normally, and all data is not lost.

The master and standby management method based on the network elements ensures that the master and standby network managers respectively manage a part of network elements, and reasonably distributes the number of the network elements managed by the managers as the master station to realize load balance; meanwhile, as the network elements accessed by the master manager and the standby manager are different, when the ECC is interrupted, all the network elements can be managed, and the reliability of the system can be improved.

The embodiments enclosed in this description are some typical examples embodying the inventive concept and do not exclude the use of other embodiments embodying the inventive design.

For example, in another embodiment, the access gateway network element of the server may not be set to connect the first or the last of the network elements in series, for example, in fig. 3, the server 2 is connected to the network element 21, and when the network element ECC is interrupted between the network element 21 and the network element 22, the network element 22-2n is not connected to any of the server a and the server B, so that the network management server cannot manage these network elements, and thus, the reliability of this connection is poor.

For another example, it is not necessary to allocate all network elements to connect with two servers in a regular manner with load balancing, and when the network elements are divided, load sharing must be performed for the two servers, and only according to the load condition of the system in actual operation, the load sharing is required to enable the system to enter the optimal operation state.

Industrial applicability

The invention discloses a method and a system for distributed main and standby management based on network elements, which is a dynamic network management protection mode of the distributed main and standby management system based on the network elements, wherein two network management servers are equal to each other, the two network management servers are connected with different gateway network elements, and each different network element has corresponding main and auxiliary authorities. Compared with the prior art, the network element needs to be configured with a bidirectional route, and the two network management servers are accessed with the network element; the two servers share the load together by reasonably distributing the number of the network elements managed by the two network management servers as the master station; when the network is in an abnormal state, or the main network management is abnormal, or the ECC of the network element is interrupted, the auxiliary network management of the corresponding network element replaces the main network management of the network element, and part of the network or the whole network is continuously managed, so that the reliability of the management of the network management of the system is improved.

Claims

1. A method for distributed master/slave management based on network elements is characterized by comprising the following steps:

(a) connecting all network elements and configuring the routes of the network elements into bidirectional intercommunication, wherein all the network elements are connected in series so that the embedded control channels of the network elements are intercommunicated, the route of each network element is configured into bidirectional intercommunication, a server A and a server B are respectively accessed to a head-end network element and a tail-end network element in the series network elements, and the two servers are connected with each other; configuring the management authority of a server A and a server B in network elements, wherein one part of the network elements configure the server A as a master station and the server B as a secondary station, and the other part of the network elements configure the server B as the master station and the server A as the secondary station;

(b) the method comprises the steps that return strategies of a main station are configured in two servers and are respectively connected to an access network element which configures the two servers as the main station, after the two servers work, the two servers serve as the main station to manage part of network elements which are respectively allocated, serve as secondary stations to monitor the other part of network elements, the return strategies are main station return modes or main station non-return modes, the main station return modes are the main station return modes which become the corresponding main station immediately after the main station is recovered, the main station non-return modes are that the main station is recovered and then becomes the secondary stations firstly, and the main station becomes the main station after the main;

(c) if the communication between the network element and the server configured as the main station is interrupted, the network element is switched, and another server configured as the auxiliary station is used as the main station to manage the network element;

(d) and when the communication between the network element and the server configured as the master station is recovered, recovering the management of the server as the master station on the network element according to the return strategy.

2. The method of claim 1, wherein the step of configuring the administrative rights of the two servers in the network element in step (a) comprises:

3. The method of claim 1, wherein in the step (b), when the server is used as a master station to manage the network elements configured as the master station, historical data of the network elements is collected, and corresponding configuration query is performed on the network elements; and when the server is used as a secondary station to manage the network element configured as the secondary station, inquiring and monitoring the network element.

4. The method of claim 1, wherein the servers in step (b) configured as primary and secondary stations are operatively synchronized to maintain data synchronization between the primary and secondary stations by performing database synchronization, wherein the historical data in the primary and secondary stations is full.

5. The method of claim 1, wherein step (c) comprises:

when the network element detects that the communication with a server which is configured as a main station is interrupted, the network element informs the server which is configured as a secondary station to be used as the main station to manage the network element;

or when the server configured as the secondary station by the network element detects that the server configured as the primary station by the network element is abnormal, the secondary station server actively requests the permission of the primary station to the network element, and the network element determines whether to switch the secondary station to manage the network element as the primary station according to the state of the current primary station server.

6. The method of claim 1, wherein the return policy in step (b) is configured as a master return mode or a non-return mode,

7. A distributed active/standby management system based on network elements comprises a plurality of network elements, a server A and a server B for managing the network elements, and is characterized in that:

all network elements are connected with each other, the routes of the network elements are configured to be in bidirectional intercommunication, all the network elements are connected in series, the embedded control channels of the network elements are communicated, the route of each network element is configured to be in bidirectional intercommunication, a server A and a server B are respectively accessed to a head-end network element and a tail-end network element in the series-connected network elements, and the two servers are connected with each other; management authorities of a server A and a server B are configured in each network element, wherein one part of the network elements configure the server A as a primary station and the server B as a secondary station, and the other part of the network elements configure the server B as the primary station and the server A as the secondary station;

the network element is switched when the communication with the server configured as the main station is interrupted, and another server configured as the auxiliary station is used as the main station; after communication with a server configured as a master station is recovered, switching the server back according to a return strategy, wherein the return strategy is a master station return mode or a master station non-return mode, the master station return mode is that the server becomes the corresponding master station immediately after the master station is recovered, and the master station non-return mode is that the server becomes a secondary station after the master station is recovered and becomes the master station after the master station applies for the master station;

8. The system of claim 7, wherein when the server is used as a master station to manage the network elements configured as the master station, historical data of the network elements is collected, and corresponding configuration query is performed on the network elements; and when the server is used as a secondary station to manage the network element configured as the secondary station, inquiring and monitoring the network element.

9. The system of claim 7, wherein the two servers synchronize data during normal operation of the primary and secondary stations configured as network elements, and maintain data synchronization between the primary and secondary stations by performing database synchronization, wherein the historical data in the primary and secondary stations is full.

10. The system of claim 7, wherein the server is configured as a master return policy as a master return mode or a non-return mode,

11. The system of claim 7, wherein the network element notifies a server configured as a secondary station to manage the network element as the primary station upon detecting an interruption of communication with the server configured as the primary station; or when the server configured as the secondary station by the network element detects that the server configured as the primary station by the network element is abnormal, the secondary station server actively requests the permission of the primary station from the network element, and the network element determines whether to switch the secondary station to manage the network element as the primary station according to the state of the current primary station server.

12. The system of claim 7, wherein when the two servers manage the network elements separately as the master, the number of managed network elements configuring them as the master is balanced, and the two servers manage the corresponding network elements separately in a load balanced manner.