CN109698757A - Switch master/slave device, the method for restoring user data, server and the network equipment - Google Patents

Abstract

The invention discloses a kind of switching master/slave device, the method for restoring user data, server and the network equipments, this method comprises: instruction host apparatus starts reservation service, wherein reservation service includes at least: user data is written into customer data base；Whether the working condition for detecting host apparatus is normal；In the case where the working condition exception of host apparatus, determine a stand-by equipment as new host apparatus according to preset configuration strategy；Indicate new host apparatus to resume work from customer data base the user data that state is abnormal host apparatus.The present invention carries out the backup of user data in existing Internet resources by customer data base, make reliably to be selected and be switched between master/slave device, it solves host apparatus in the prior art and stand-by equipment corresponds, its corresponding stand-by equipment can only be used when host apparatus failure, other equipment resource in network, the low problem of network utilization cannot be made full use of.

Description

Switch master/slave device, the method for restoring user data, server and the network equipment

Technical field

The present invention relates to the communications fields, more particularly to a kind of switching master/slave device, the method for restoring user data, service Device and the network equipment.

Background technique

With the development of broadband metropolitan area network and broadband Internet protocol (IP, Internet Protocol) business, broadband access Server (BRAS, Broadband Remote Access Server) is used as a kind of broadband access equipment, can be intelligently It realizes the service such as certification, authorization, charging of user, easily can also provide a variety of IP value-added services according to the needs of users. It can complete the data access of broadband IP user, realize the broadband IP online of business premises and neighbours living.It is located at network Edge, provide broadband inserting service, realize multiple business convergence and forwarding, be able to satisfy different user to transmission capacity and band The requirement of wide utilization rate, therefore be the core equipment of broadband user's access.

Carried on present broadband IP network using more and more, including online management of personal money, 4K ultra high-definition Web TV (IPTV, Internet Protocol Television), Video chat, the service such as real-time game, to broadband inserting service More stringent requirements are proposed for reliability, when link or equipment failure cause broadband inserting service to be interrupted when, Neng Gou Alternate device or chain road rapidly self-healing restore, and prevent the long-time of business from interrupting.On the other hand, the increase of broad band network application Network size and number of devices is caused to be growing, operator needs to make full use of Internet resources, mitigates equipment and service management The pressure of maintenance.

Conventional wide band access device mainly solves the integrity problem of equipment by the way of two-node cluster hot backup, needs in Liang Tai Configuration backup link between equipment, the access user data on host apparatus are real-time transmitted to stand-by equipment by backup link On, pass through Virtual Router Redundacy Protocol (VRRP, Virtual Router Redundancy Protocol) between two equipment To control the active-standby switch of equipment.Due to there is fixed backup relation between master/slave device, when main equipment breaks down, by Disconnected business can only be restored on the stand-by equipment for having backup link connection relationship with main equipment, and cannot be in other free devices Upper recovery business, is not able to satisfy the requirement for making full use of Internet resources.

Host apparatus in the prior art is solved therefore, it is necessary to a kind of method and stand-by equipment corresponds, host apparatus event Its corresponding stand-by equipment can only be used when barrier, cannot make full use of other equipment resource in network, network utilization is low to ask Topic.

Summary of the invention

The present invention provides a kind of switching master/slave device, the method for restoring user data, server and the network equipment, to solve Certainly host apparatus and stand-by equipment correspond in the prior art, and when host apparatus failure can only use its corresponding spare to set It is standby, other equipment resource in network, the low problem of network utilization cannot be made full use of.

In order to solve the above technical problems, on the one hand, the present invention provides a kind of method for switching master/slave device, comprising: instruction Host apparatus starts reservation service, wherein the reservation service includes at least: user data is written into customer data base； Whether the working condition for detecting the host apparatus is normal；In the case where the working condition exception of the host apparatus, according to Preset configuration strategy determines a stand-by equipment as new host apparatus；Indicate the new host apparatus from the number of users According to the user data that state is abnormal host apparatus of resuming work in library.

Further, indicate that the new host apparatus resumes work state from the customer data base as the primary of exception After the user data of equipment, further includes: the instruction new host apparatus is that the primary of exception is set according to the working condition Standby user data starts the reservation service.

Further, instruction host apparatus starts reservation service, comprising: constructs N number of backup logic group, and standby in each logic Select a default equipment as host apparatus in M equipment in part group, other equipment conducts in addition to the host apparatus The stand-by equipment of the host apparatus, wherein N is the integer greater than 0, and M is the integer greater than 1；By N number of backup logic group In the information of each backup logic group be sent to the corresponding host apparatus of the backup logic group so that the host apparatus starts Reservation service.

Further, before constructing N number of backup logic group, further includes: obtain the address information of N number of equipment；Obtain institute State the address information of customer data base；Each equipment and the user in N number of equipment are established according to the address information of acquisition The communication channel of database.

Further, the information of N number of backup logic group is sent to after each equipment, further includes: according to default The backup logic group where pre-set user equipment is determined with rule, wherein the preset matching rule is N number of backup logic The matching rule of each backup logic group in group；User's request of the pre-set user equipment is forwarded to the backup logic group Host apparatus.

Further, the method also includes: whether the host apparatus of test job abnormal state restores to schedule Normal operating conditions；It, will be described new in the case where the host apparatus of the working condition exception restores normal operating conditions Host apparatus is changed to the stand-by equipment of the current logic backup group, and the working condition that will restore normal operating conditions Abnormal host apparatus is changed to host apparatus.

On the other hand, the present invention provides a kind of method of restoring user data characterized by comprising receives backup and adjusts Spend the switching command that device is sent；Restore the user data of the first equipment from customer data base according to the switching command；According to The user data starts reservation service, wherein the reservation service includes at least: user data is written to customer data base In.

On the other hand, the present invention provides a kind of server for switching master/slave device, includes at least memory, processor, institute It states and is stored with computer program on memory, which is characterized in that the processor is executing the computer journey on the memory Following steps are realized when sequence: instruction host apparatus starting reservation service, wherein the reservation service includes at least: by number of users According to write-in into customer data base；Whether the working condition for detecting the host apparatus is normal；In the work of the host apparatus In the case where abnormal state, determine a stand-by equipment as new host apparatus according to preset configuration strategy；It indicates described new Host apparatus resume work from the customer data base user data that state is abnormal host apparatus.

Further, the processor executes instruction the new host apparatus and resumes work from the customer data base shape After the step of state is all customer data of abnormal host apparatus, also execution following steps: instruction is described new primary to be set It is standby that the reservation service is started according to all customer data that working condition is abnormal host apparatus.

Further, when the processor executes instruction the step of host apparatus starting reservation service, following steps: structure are executed N number of backup logic group is built, and selects in M equipment in each backup logic group a default equipment as host apparatus, Stand-by equipment of other equipment as the host apparatus in addition to the host apparatus, wherein N is the integer greater than 0, and M is Integer greater than 1；It is corresponding that the information of each backup logic group in N number of backup logic group is sent to the backup logic group Host apparatus so that the host apparatus start reservation service

Further, before the processor executes the step of constructing N number of backup logic group, following steps are also executed: being obtained The address information of N number of equipment；Obtain the address information of the customer data base；According to the foundation of the address information of acquisition The communication channel of each equipment and the customer data base in N number of equipment.

Further, the step of information of N number of backup logic group is sent to each equipment by processor execution it Afterwards, following steps are also executed: determining the backup logic group where pre-set user equipment according to preset matching rule, wherein described Preset matching rule is the matching rule of each backup logic group in N number of backup logic group；By the pre-set user equipment User's request be forwarded to the host apparatus of the backup logic group.

Further, the processor also executes following steps: the primary of test job abnormal state is set to schedule It is standby whether to restore normal operating conditions；In the case where the host apparatus of the working condition exception restores normal operating conditions, The new host apparatus is changed to the stand-by equipment of the current logic backup group, and the institute that will restore normal operating conditions The host apparatus for stating working condition exception is changed to host apparatus.

On the other hand, the present invention provides a kind of network equipment, includes at least memory, processor, deposits on the memory Contain computer program, which is characterized in that the processor is realized as follows when executing the computer program on the memory Step: the switching command of backup scheduler dispatches is received；Restore the first equipment from customer data base according to the switching command User data；Start reservation service according to the user data, wherein the reservation service includes at least: by user data Write-in is into customer data base.

The present invention can be according to pre- establishing when making host apparatus operation irregularity by user data library backup user data The determining stand-by equipment of strategy is set as new host apparatus, and indicates that new host apparatus is extensive directly from customer data base It is multiplexed user data, to complete equipment switching, restores user access network.The present invention passes through number of users in existing Internet resources The backup that user data is carried out according to library, makes reliably to be selected and be switched between master/slave device, solves the prior art Middle host apparatus and stand-by equipment correspond, and when host apparatus failure can only use its corresponding stand-by equipment, cannot be abundant Utilize other equipment resource in network, the low problem of network utilization.

Detailed description of the invention

Fig. 1 is the method flow diagram for switching master/slave device in first embodiment of the invention；

Fig. 2 is the method flow diagram of restoring user data in second embodiment of the invention；

Fig. 3 is the broadband access system schematic diagram of fifth embodiment of the invention；

Fig. 4 is the broadband access system schematic diagram of sixth embodiment of the invention；

Fig. 5 is the broadband access system schematic diagram of seventh embodiment of the invention.

Specific embodiment

It is corresponded to solve host apparatus in the prior art and stand-by equipment, when host apparatus failure can only use it Corresponding stand-by equipment cannot make full use of other equipment resource in network, the low problem of network utilization.The present invention provides A kind of method, the method for restoring user data, server and network equipment switching master/slave device, below in conjunction with attached drawing and reality Example is applied, the present invention will be described in further detail.It should be appreciated that specific embodiment described herein is only used to explain this Invention does not limit the present invention.

The first embodiment of the present invention provide it is a kind of switch master/slave device method, method flow diagram as shown in Figure 1, Specifically include step S101 to S104:

S101, instruction host apparatus start reservation service, wherein reservation service includes at least: by user data be written to In customer data base；

Whether S102, the working condition for detecting host apparatus are normal；

S103 determines that one spare sets according to preset configuration strategy in the case where the working condition exception of host apparatus For as new host apparatus；

S104, indicate new host apparatus to resume work from customer data base the user that state is abnormal host apparatus Data.

In the present embodiment, all host apparatus and stand-by equipment all can be one or more of following equipment: wide Band remote access server, wideband network gateway (BNG, Broadband Network Gateway), broadband services gateway (BSG, Broadband Service Gateway), business router (SR, Service Router), BAS Broadband Access Server (BAS, Broadband Access Server), Broadband Remote Access Server (BRAS, Broadband Remote Access Server) etc., it is logical by being established in forwarding surface all comprising providing the application program of broadband inserting service in above-mentioned all devices Road, completion and the data interaction between user equipment, customer data base and server.All user equipmenies can in the present embodiment Think residential gateway (RG, Residential Gateway), user side equipment (CPE, Customer Premise Equipment), personal computer (PC, Personal Computer), the networking telephone (VoIP, Voice over Internet Protocol), Web TV (IPTV, Internet Protocol Television), set-top box (STB, Set Top Box) And integrated access equipment (IAD, Integrated Access Device) etc..

Server is used as backup scheduler, is used to indicate host apparatus starting reservation service, wherein reservation service can To include that user data is written into customer data base, it also may include receiving broadband user's access authentication, authorize, at charging Reason writes user's forwarding table, and user is allowed to access the service such as internet (Internet).Wherein, user data includes: user equipment Media Access Control address (MAC, Media Access Control), Internet protocol address (IP, Internet Protocol Address), gateway, subnet mask, access style, session id, on-line time, flowing of access statistics, certification letter Breath and authorization and charge information.

Backup scheduler is by the communication channel established between host apparatus, and timing is inquired or reception host apparatus active The host apparatus working condition reported, including current host apparatus whether normal operation, each interface of host apparatus be normal Or it disconnects, whether the connectivity of link between host apparatus interface and user equipment is normal etc..Backup scheduler is at one section If inquiry is less than host apparatus in, or does not receive the information of host apparatus active reporting, then it is assumed that host apparatus work Make it is abnormal, if backup scheduler can receive or inquire the working condition of host apparatus, but host apparatus reports it Some interfaces is abnormal, then it also hold that host apparatus operation irregularity.In the case where host apparatus operation irregularity, backup is adjusted Degree device determines a stand-by equipment as new host apparatus according to preset configuration strategy and indicates new host apparatus from user Resume work in database the user data of abnormal host apparatus.Also, backup scheduler also can indicate that new primary set It is standby that reservation service is started according to the user data of recovery, so that the user equipment being connected on former host apparatus is accessed net again Network.In order to select optimal stand-by equipment as host apparatus according to the actual situation, preset configuration strategy can be following methods One or more of: the least equipment of selection number of users, selection central processing unit (CPU, Central Processing Unit) the smallest equipment of utilization rate, the smallest equipment of selection memory usage or each parameter of selection multiplied by weighted value it is cumulative and The smallest equipment.

In the present embodiment, backup scheduler is by construction logic backup group, to indicate that host apparatus starts reservation service. All equipment is divided into N number of backup logic group by backup scheduler, and there may be M equipment in each backup logic group, upper It states and selects a default equipment as host apparatus in M equipment, other equipment in addition to host apparatus are as host apparatus Stand-by equipment, and N number of backup logic group information is generated after configuration, by backup logic group each in N number of backup logic group Information be sent to the corresponding host apparatus of backup logic group so that host apparatus start reservation service.It will be appreciated that N is the integer greater than 0, and M is the integer greater than 1, that is to say, that may exist the case where only one backup logic group, it can also There is the case where only there are two equipment.

It, can be by configuring ground between backup scheduler, equipment and customer data base before construction logic backup group The modes such as location, subscription obtain the address informations such as mutual physical address, and establish communication channel by existing communication protocol It carries out to complete mutual connection, existing communication protocol includes network configuration protocol (NETCONF, Network Configuration Protocol), transparent interprocess commications protocol (TIPC, Transparent Inter process Communication), transmission control protocol agreement (TCP, Transmission Control Protocol), user datagram Agreement (UDP, User Datagram Protocol), hypertext transfer protocol (HTTP, Hyper Text Transfer Protocol), Simple Network Management Protocol (SNMP, Simple Network Management Protocol) etc..

User equipment determines the backup logic where the user equipment when initiating request according to preset matching rule first Group is handled in the host apparatus that user's request of user equipment is sent to counterlogic backup group.Wherein, preset matching The rule that rule is arranged when being construction logic backup group, each backup logic group may correspond to the different knots of preset matching rule Fruit, the host apparatus that belonging to the request that the backup logic group is handled should be forwarded in the backup logic group are handled.Default Particular content with rule can be configured according to the actual conditions of network, can be, but not limited to configuration method below: root It matched according to the modular arithmetic result of the MAC Address of user equipment, matched according to packet receiving port numbers, class is requested according to user Type matches etc..

Further, new host apparatus of the stand-by equipment as current logic backup group is determined in backup scheduler Later, current logic backup group information also synchronizes update, and the information of updated backup logic group is sent to network In new host apparatus, new host apparatus is by the INFORMATION DISCOVERY oneself of updated backup logic group in current backup group In host apparatus switched to by stand-by equipment, it is standby to restore current logic from customer data base according to the instruction of backup scheduler The all customer data of part group, allows the original subscriber of current logic backup group to continue to access and access Internet, and will be new The user data upload of the current logic backup group of generation is to customer data base.

And further, in current logic backup group, backup scheduler can test job to schedule Whether the host apparatus of abnormal state restores normal operating conditions, if the host apparatus of working condition exception test when Restore to work normally, then can switch back into the host apparatus newly changed as the stand-by equipment of current logic backup group, and will be extensive The host apparatus of the working condition exception of multiple normal operating conditions is changed to current logic backup group host apparatus.For example, logic In backup group A, current host apparatus is X, and stand-by equipment is Y and Z, and for backup scheduler when being timed inquiry, discovery can not The information of host apparatus X active reporting is received, host apparatus X operation irregularity is determined, at this point, there are 20 use on stand-by equipment Y There are 200 users at family on stand-by equipment Z, then backs up scheduler and select stand-by equipment Y as the new primary of backup logic group A Equipment, and equipment Y is notified to download all customer data of backup logic group A from customer data base, to restore backup logic group A The internet function of middle user.After a period of time, backup scheduler tests host apparatus X and restores to work normally, then logic is standby The host apparatus of part group A switches back into X, all customer data of backup logic group A is downloaded from customer data base by X, by equipment Y Stand-by equipment as backup logic group A.

The present invention can be according to pre- establishing when making host apparatus operation irregularity by user data library backup user data The determining stand-by equipment of strategy is set as new host apparatus, and indicates that new host apparatus is extensive directly from customer data base It is multiplexed user data, to complete equipment switching, restores user access network.In existing Internet resources by customer data base into The backup of row user data makes reliably to be selected and be switched between master/slave device, solve primary in the prior art Equipment and stand-by equipment correspond, and when host apparatus failure can only use its corresponding stand-by equipment, cannot make full use of net Other equipment resource in network, the low problem of network utilization.

The second embodiment of the present invention provides a kind of method of restoring user data, and flow chart is as shown in Fig. 2, specific Including step S201 to S203:

S201 receives the switching command of backup scheduler dispatches；

S202 restores the user data of the first equipment according to switching command from customer data base；

S203, according to user data start booking service, wherein reservation service includes at least: by user data be written to In customer data base.

Of the invention second implements the method provided, is mainly used in equipment, standby when the first equipment working state exception It is confirmed as after new host apparatus with equipment, which can receive the switching command of backup scheduler dispatches, and Restore the user data of the first equipment from customer data base according to switching command, and then original first is restored according to user data and is set The reservation services such as the network insertion of standby lower user.

Preferably, the first equipment is the host apparatus of a certain backup logic group, and the second equipment is in the backup logic group The stand-by equipment of first equipment selects the second equipment as new master when backup scheduler determines the first equipment operation irregularity With equipment, and the information (being equivalent to switching command) of the updated backup logic group is sent to the second equipment, the second equipment connects After the information for receiving the updated backup logic group, restore all numbers of users of the backup logic group from customer data base According to, and restarting reservation service makes the user of the backup logic group restore network insertion.

The present embodiment is restored from customer data base after receiving switching command according to switching command by the second equipment The user data of first equipment, so that the second equipment only carries out data recovery when needed, without carrying out real-time data Backup ensure that flexibility when data are restored and practicability.

The third embodiment of the present invention provides a kind of server for switching master/slave device, which includes at least storage Device, processor, computer program is stored on memory, and processor realizes step when executing the computer program on memory S101 to S104, and the step of with first embodiment of the invention is identical, details are not described herein.In actual use, the present embodiment Provided server can be the backup scheduler of physical presence, or the Virtual Service run in generic server Device.

It is logical by being established in forwarding surface all comprising providing the application program of broadband inserting service in all devices in network Road, completion and the data interaction between user equipment, customer data base and server.Server is used as backup scheduler, It is used to indicate host apparatus starting reservation service, wherein reservation service may include user data being written to customer data base In, it also may include receiving broadband user's access authentication, authorization, charging processing writes user's forwarding table, user is allowed to access interconnection The service such as net.Wherein, reservation service may include that user data is written into customer data base, also may include receiving broadband User's forwarding table is write in access authentication of user, authorization, charging processing, and user is allowed to access the service such as Internet.Preferably, it uses User data includes: MAC Address, IP address, gateway, subnet mask, access style, the session id, on-line time, visit of user equipment Ask traffic statistics, authentication information and authorization and charge information.

Backup scheduler is by the communication channel established between host apparatus, and timing is inquired or reception host apparatus active The host apparatus working condition reported, including current host apparatus whether normal operation, each interface of host apparatus be normal Or it disconnects, whether the connectivity of link between host apparatus interface and user equipment is normal etc..Backup scheduler is at one section If inquiry is less than host apparatus in, or does not receive the information of host apparatus active reporting, then it is assumed that host apparatus work Make it is abnormal, if backup scheduler can receive or inquire the working condition of host apparatus, but host apparatus reports it Some interfaces is abnormal, then it also hold that host apparatus operation irregularity.In the case where host apparatus operation irregularity, backup is adjusted Degree device determines a stand-by equipment as new host apparatus according to preset configuration strategy and indicates new host apparatus from user Resume work in database the user data of abnormal host apparatus.Also, backup scheduler also can indicate that new primary set It is standby that reservation service is started according to the user data of recovery, so that the user equipment being connected on former host apparatus is accessed net again Network.In order to select optimal stand-by equipment as host apparatus according to the actual situation, preset configuration strategy can be following methods One or more of: the least equipment of selection number of users, the smallest equipment of selection cpu busy percentage, selection memory usage are most Small equipment or each parameter of selection adds up and the smallest equipment multiplied by weighted value.

Backup scheduler is by construction logic backup group, to indicate that host apparatus starts reservation service.Backup scheduler will All equipment is divided into N number of backup logic group, and there may be M equipment in each backup logic group, selects in above-mentioned M equipment A default equipment is selected as host apparatus, stand-by equipment of other equipment as host apparatus in addition to host apparatus, and N number of backup logic group information is generated after configuration, and the information of backup logic group each in N number of backup logic group is sent to this The corresponding host apparatus of backup logic group, so that host apparatus starts reservation service.It will be appreciated that N is whole greater than 0 Number, M are the integer greater than 1, that is to say, that may exist the case where only one backup logic group, there may also be only two The case where a equipment.

Before construction logic backup group, between server, equipment and customer data base can by configuration address, order The modes such as read and obtain the address informations such as mutual physical address, and by existing communication protocol establish communication channel carry out with Complete mutual connection.

User equipment determines the backup logic where the user equipment when initiating request according to preset matching rule first Group is handled in the host apparatus that user's request of user equipment is sent to counterlogic backup group.Wherein, preset matching The rule that rule is arranged when being server construction backup logic group, each backup logic group may correspond to preset matching rule not With as a result, the host apparatus that the request for belonging to backup logic group processing should be forwarded in the backup logic group is handled.In advance If the particular content of matching rule can be configured according to the actual conditions of network, configuration side below can be, but not limited to Method: it matched according to the modular arithmetic result of the MAC Address of user equipment, matched, according to packet receiving port numbers according to user Request type matches etc..

Further, new host apparatus of the stand-by equipment as current logic backup group is determined in backup scheduler Later, current logic backup group information also synchronizes update, and the information of updated backup logic group is sent to network In new host apparatus, new host apparatus is by the INFORMATION DISCOVERY oneself of updated backup logic group in current backup group In host apparatus switched to by stand-by equipment, it is standby to restore current logic from customer data base according to the instruction of backup scheduler The all customer data of part group, allows the original subscriber of current logic backup group to continue to access and access Internet, and will be new The user data upload of the current logic backup group of generation is to customer data base.

And further, in current logic backup group, backup scheduler can test job to schedule Whether the host apparatus of abnormal state restores normal operating conditions, if the host apparatus of working condition exception test when Restore to work normally, then can switch back into the host apparatus newly changed as the stand-by equipment of current logic backup group, and will be extensive The host apparatus of the working condition exception of multiple normal operating conditions is changed to current logic backup group host apparatus.

The present embodiment is by user data library backup user data, can be according to default when making host apparatus operation irregularity Configuration strategy determines that a stand-by equipment as new host apparatus, and indicates new host apparatus directly from customer data base Restoring user data restores user access network to complete equipment switching.Pass through customer data base in existing Internet resources The backup for carrying out user data, makes reliably to be selected and be switched between master/slave device, solves and lead in the prior art It is corresponded with equipment and stand-by equipment, when host apparatus failure can only use its corresponding stand-by equipment, cannot make full use of Other equipment resource in network, the low problem of network utilization.

The fourth embodiment of the present invention provides a kind of network equipment, which includes at least memory, processor, Computer program is stored on memory, processor realizes step S201 extremely when executing the computer program on memory S203, and the step of with second embodiment of the invention is identical, details are not described herein.In actual use, the present embodiment is provided Server can be the network equipment with entity, or the virtual network device run in generic server.

Of the invention the 4th implements the network equipment provided, and when the first equipment working state exception, stand-by equipment is determined After new host apparatus, which can receive the switching command of backup scheduler dispatches, and according to switching command Restore the user data of the first equipment from customer data base, and then restores the net of user under former first equipment according to user data The reservation services such as network access.

Preferably, the first equipment is the host apparatus of a certain backup logic group, and the second equipment is in the backup logic group The stand-by equipment of first equipment selects the second equipment as new master when backup scheduler determines the first equipment operation irregularity With equipment, and the information (being equivalent to switching command) of the updated backup logic group is sent to the second equipment, the second equipment connects After the information for receiving the updated backup logic group, restore all numbers of users of the backup logic group from customer data base According to, and restarting reservation service makes the user of the backup logic group restore network insertion.

Further, after a period of work, the discovery of backup scheduler can not receive the second equipment actively to the second equipment The information reported determines the second equipment operation irregularity, selects another stand-by equipment, i.e. third in current logic backup group at this time New host apparatus of the equipment as current logic backup group, and updated backup logic group information is sent to third and is set It is standby；After third equipment receives the information, restore all customer data of current logic backup group from customer data base, and Restarting reservation service makes the user of the backup logic group restore network insertion.It will be appreciated that third equipment was restored User data include the first equipment as host apparatus when the user data that generates and the second equipment as host apparatus when production Raw user data.

The present embodiment is restored from customer data base after receiving switching command according to switching command by the network equipment The user data of first equipment, so that the network equipment only carries out data recovery when needed, without carrying out real-time data Backup ensure that flexibility when data are restored and practicability.

The fifth embodiment of the present invention provides a kind of broadband access system, and schematic diagram is as shown in Figure 3.In the present embodiment In, hardware module part includes: generic server, BRAS and user equipment.Wherein, generic server is for installing virtualization system System, such as OpenStack, can be used for creating virtual machine and virtual network；BRAS can distinguish broadband access protocol massages and use The data message of family access Internet is above sent at the virtual machine LB in generic server broadband access protocol massages Reason, and for user data message, Internet is forwarded to according to information in user's table in device forwards face；User equipment is hair The equipment for playing broadband user's access request is supplied to the ability that user accesses Internet after request certification passes through.

The step of the present embodiment, is described in detail below with reference to Fig. 3.

S301 creates 6 virtual machines in generic server, wherein starting broadband inserting service conduct in 3 virtual machines Broadband inserting service entity (is equivalent to the equipment in first embodiment of the invention), for the virtual of each starting broadband inserting service Machine distribution and injection unique identification ID, i.e. broadband inserting service ID, device id can be conduit number, innings number etc., three broadbands connect Enter service entities and is respectively designated as PP1, PP2 and PP3.

S302 starts backup scheduler service on 1 virtual machine, is named as MP and (is equivalent to first embodiment of the invention In backup scheduler), starting customer data base business (is equivalent to the user in first embodiment of the invention on 1 virtual machine Database), it is named as CDN, starts load sharing business on last 1 virtual machine, is named as LB.Backup scheduler MP and user Database CDN, load sharing business LB, virtual 2 layer network is disposed between broadband inserting service entity PP, communicate between them and adopt With TIPC agreement.

S303, backup scheduler service starting, issues a global service, customer data base and broadband inserting service entity Initiation of services, and subscribe to the global service of backup scheduler publication.

S304, backup scheduler perceives database and broadband inserting service entity subscribes to the service of oneself publication, obtains Subscriber Information, including subscriber address, such as job identifier (JID, Job IDentifier) etc..

The JID information of S305, backup scheduler dispatches customer data base give broadband inserting service entity.

S306, configures 3 backup logic groups, respectively backup group 1, backup group 2, backup group 3, in each backup group plus Enter PP1, PP2, PP3, the matching condition (the preset matching rule for being equivalent to first embodiment of the invention) of backup group is to broadband Source MAC value modulus 3 in access request message may be 0,1,2 to the value obtained after any 3 operation of MAC Address value modulus, match After the completion of setting, the backup logic group information in backup scheduler is as shown in table 1:

Table 1

Backup group	PP1	PP2	PP3	Matching condition
					Backup group 1	It is primary	It is spare	It is spare	MAC mould 3 value in source is equal to 0
Backup group 2	It is spare	It is primary	It is spare	MAC mould 3 value in source is equal to 1
					Backup group 3	It is spare	It is spare	It is primary	MAC mould 3 value in source is equal to 2

S307, backup scheduler dispatches backup logic group information to PP1, PP2, PP3.

S308, backup scheduler issue matching condition to LB, and matching condition is: MAC mould 3 value in source is equal to 0 message, forwarding To PP1 processing；MAC mould 3 value in source is equal to 1 message, is forwarded to PP2 processing；MAC mould 3 value in source is equal to 2 message, is forwarded to PP3 Processing.

S309, user equipment sends broadband access request message first, such as based on the PPP(Point to Point Protocol) of Ethernet (PPPoE, Point to Point Protocol over Ethernet) message, after BRAS equipment receives message, according to message The feature field of middle carrying judges whether it is broadband access protocol massages, for example EtherType field is 0x8863, indicates PPPoE Discovery phase message etc., BRAS are the discovery that broadband access protocol massages, are then sent to LB.

After S310, LB receive broadband access request message, 3 operation of mould, such as source are carried out to the source MAC in request message MAC is the message of 00:00:00:00:00:18, and after carrying out 3 operation of mould, obtained remainder (i.e. modulus value) is 0, according to matching item Part, message are sent to PP1；Source MAC is the message of 00:00:00:00:00:19, and after carrying out 3 operation of mould, obtained remainder is 1, according to matching condition, message is sent to PP2；Source MAC is the message of 00:00:00:00:00:20, after carrying out 3 operation of mould, Obtained remainder is 2, and according to matching condition, message is sent to PP3.

After S311, PP1 receive broadband access request message, 3 operation of mould is carried out to the source MAC in request message, such as Source MAC is 00:00:00:00:00:18, and after carrying out 3 operation of mould, obtained remainder is 0, looks into backup logic group information, is matched to Backup group 1, because indicating that PP1 is primary in backup group 1, so to receive broadband user online by PP1.

S312 saves user data, belonging to user data after PP1 completes the Certificate Authority processing of broadband user's access Backup logic group ID be 1, PP1 write according to the address JID of the customer data base of backup scheduler dispatches by reliable news Backup logic group ID and user data are to customer data base.

S313 keeps communication between backup scheduler and PP1, PP2, PP3, constantly whether detection PP is online, if PP Online, backup scheduler inquires the load condition of each PP, that is, counts and be currently accessed number of users on each PP.

S314, when backup scheduler does not receive request or the response message of PP transmission, backup scheduling whithin a period of time Server power-off or damage where device judges PP operation irregularity, such as PP1, backup scheduler can't detect PP1 presence, beginning time All backup logic group informations are gone through, discovery PP1 is the primary broadband inserting service entity of backup logic group 1, and PP2 and PP3 are The spare PP of backup logic group 1.According to the default configuration strategy in backup scheduler, select the least PP of access customer number as New primary broadband inserting service entity.Assuming that it is 300 that number of users, which is the number of users in 200, PP3, in PP2 at this time, then select Primary access service entity of the PP2 as backup logic group 1, backup scheduler update backup logic group information, such as table 2 after update It is shown:

Table 2

S315, backup scheduler notify updated backup logic group information to find oneself in logic to PP2 and PP3, PP2 Become primary from spare in backup group 1, then customer data base is notified to obtain all customer data that backup logic group ID is equal to 1. After PP2 gets user data, restore the access environment of all users of the backup logic group ID equal to 1 in PP2, including establish MAC and Session is user's table of index, increases the traffic statistics of user in PP2, increases access customer number, start to process and patrol The charging message for collecting all users of the backup group ID equal to 1 updates.

S316, backup scheduler notify LB to update load sharing forward rule, i.e., source MAC carries out 3 operation of mould in message As a result it is 0 message, is forwarded to PP2 processing.Load sharing rule change are as follows: MAC mould 3 value in source is equal to 0 message, is forwarded to PP2 processing；MAC mould 3 value in source is equal to 1 message, is forwarded to PP2 processing；MAC mould 3 value in source is equal to 2 message, is forwarded at PP3 Reason.

S317, PP2 start to receive the broadband access request message for belonging to backup group 1, authenticate to user, authorize and count Take processing.When new user is online, if belonging to the user of backup group 1, processing is received, writes the ID of user data and backup group 1 Information is to customer data base；When new user offline, then user data is deleted by PP2 notice customer data base.

Preferably, it in order to guarantee the reliability of broadband access system, prevents when generic server breaks down, it should All virtual machines being arranged under generic server go offline the case where not being available, can be according to the actual situation by multiple virtual machines It is separately positioned in multiple and different property servers, and multiple property servers can be interconnected by network and be counted It can be by being arranged in when guaranteeing that the virtual machine caused by a property server failure goes offline as host apparatus according to communication Stand-by equipment on other servers restores broadband inserting service.It is not been shown in detail in Fig. 2 and places multiple virtual machines respectively Multiple and different property servers the case where.

The sixth embodiment of the present invention provides another broadband access system, and networking schematic diagram is as shown in Figure 4.At this In embodiment, hardware module part includes: generic server (being equivalent to the backup scheduler in first embodiment of the invention), BRAS (is equivalent to the equipment in first embodiment of the invention), user equipment, convergence switch, core router (CR, Core Router), customer data base.Wherein, generic server is for installing virtualization system, such as OpenStack, and creation is transported thereon The virtual machine of row backup dispatcher software；BRAS can distinguish broadband access protocol massages and user accesses the data of Internet Message, for broadband access protocol massages, the LB virtual machine processing being above sent in generic server, and for user datagram Text is forwarded to Internet according to information in user's table in device forwards face；User equipment is asked for initiating broadband user's access It asks, after certification passes through, provides user's ability for accessing Internet；CR is used for core/backbone network packet routing forwarding；It converges Poly- interchanger is the layer 2 ethernet switch that the service message that docking networks carries out convergence forwarding；Customer data base is to provide to deposit Storage, inquiry update the server of wide band access user data.

The step of the present embodiment, is described in detail below with reference to Fig. 4.

As shown in figure 4, increasing backup scheduler and user data on the basis of existing broadband inserting service networking application Library, and function upgrading is carried out to traditional BRAS equipment, so that the BRAS equipment after upgrading has connection backup scheduler and user Database function handles backup logic group information ability, from customer data base restoring user data ability.Specific processing step is such as Under:

S401 starts backup scheduler application software, the IP address of configuration backup scheduler and port in generic server Number.

S402, start BRAS equipment, configuration broadband user access, configuration based on extensible markup language (XML, Extensible Markup Language) NETCONF service listen to IP address and port numbers.

S403 starts customer data base service on customer data base, configures the IP address and port numbers of customer data base.

S404, backup scheduler is as NETCONF client, configuration every BRAS equipment of connection.

S405 configures the customer data base address that each BRAS equipment needs to connect, backup scheduling on backup scheduler Device is issued to the customer data base address information configured on backup scheduler in BRAS equipment, BRAS by the channel NETCONF After equipment receives customer data base address, the TCP begun setting up between BRAS equipment and customer data base is reliably connected.

S406, configures 3 backup logic groups, respectively backup group 1, backup group 2, backup group 3, in each backup group plus Enter BRAS1, BRAS2, BRAS3, the matching condition of backup group is to receive BRAS equipment interface of broadband access request message etc. In the interface of configuration.After the completion of configuration, the backup logic group information in backup scheduler is as shown in table 3:

Table 3

Backup group	BRAS1	BRAS2	BRAS3	Matching condition
					Backup group 1	It is primary	It is spare	It is spare	Packet receiving port fei-0/1/0/1
Backup group 2	It is spare	It is primary	It is spare	Packet receiving port fei-0/1/0/2
					Backup group 3	It is spare	It is spare	It is primary	Packet receiving port fei-0/1/0/3

S407, backup scheduler dispatches backup logic group information to BRAS1, BRAS2, BRAS3.

S408, user equipment send broadband access request message first, such as dynamic host configuration protocol (DHCP, Dynamic Host Configuration Protocol) message, after BRAS equipment receives message, according to the feature field carried in message Judge whether it is broadband access protocol massages, for example the feature field of DHCP protocol message includes: that (1) EtherType is equal to 0x0800, (2) IP protocol fields are equal to 17, and (3) UDP port number is 67 or 68.

S409, after BRAS equipment forwarding surface receives broadband access protocol massages, forwarding surface is protocol massages and packet receiving end Mouth together on be sent to the control surface treatment of BRAS equipment.

S410, after BRAS equipment control plane receives protocol massages and packet receiving port, control plane begins stepping through backup logic Group information, it is assumed that the packet receiving port for receiving protocol massages is fei-0/1/0/1, meets the matching condition of backup logic group 1, and And in backup logic group 1, mark BRAS1 be it is primary, therefore, BRAS1 control this protocol massages of surface treatment, with user equipment Interaction protocol message authenticates user's access, authorization and charging, saves the user data generated in access treatment process. Meanwhile BRAS1 control plane sends user data and affiliated backup group ID to customer data base, by customer data base preservation and more This new user data.When BRAS2 equipment receives protocol massages and packet receiving port information, control plane begins stepping through backup logic group Information, discovery meet the matching condition of backup logic group 1, still, in backup logic group 1, mark BRAS2 be it is spare, therefore, BRAS2 control plane abandons this protocol massages, is not further processed.

S411 keeps communication between backup scheduler and BRAS1, BRAS2, BRAS3, constantly whether detect BRAS equipment Online, if BRAS equipment is online, the working condition of each port in each BRAS equipment is obtained by NETCONF, such as Detection port is to be in off-state in connection normal condition.

S412, backup scheduler count the load condition of the BRAS equipment of each connection state, for example count each BRAS The number of users accessed in equipment.

S413, when backup scheduler detects that off-state, such as cable or optical cable damage occurs in BRAS equipment interface.Example Such as, the fei-0/1/0/2 interface in BRAS2 equipment disconnects, and backup scheduler begins stepping through all backup logic group informations, sends out Existing packet receiving port fei-0/1/0/2 meets the matching condition of backup logic group 2, and information shows that BRAS2 is main in backup logic group 2 With equipment, but since packet receiving port disconnects, BRAS2 equipment cannot receive user's access protocol message, need to carry out active and standby Switching.According to the default configuration strategy in backup scheduler, select the least BRAS equipment of access customer number as newly primary Broadband inserting service entity.Assuming that it is 100 that number of users, which is the number of users in 10000, BRAS3, in BRAS1 at this time, then select Primary access service entity of the BRAS3 as backup logic group 2, backup scheduler update backup logic group information, after update such as Shown in table 4:

Table 4

Backup group	BRAS1	BRAS2	BRAS3	Matching condition
					Backup group 1	It is primary	It is spare	It is spare	Packet receiving port fei-0/1/0/1
Backup group 2	It is spare	It is spare	It is primary	Packet receiving port fei-0/1/0/2
					Backup group 3	It is spare	It is spare	It is primary	Packet receiving port fei-0/1/0/3

S414, backup scheduler notify updated backup logic group information to BRAS1, BRAS2 and BRAS3, work as BRAS3 It was found that oneself becomes primary from spare in backup logic group 2, then customer data base is notified to obtain backup logic group ID and be equal to 2 All customer data after BRAS3 gets user data, restores all users of the backup logic group ID equal to 2 in BRAS3 Access environment increases the traffic statistics of user, increase connects including establishing user's table that MAC and Session is index in BRAS3 Access customer number starts to process charging message update of all users of the backup logic group ID equal to 2 etc..

S415, BRAS3 equipment send free address resolution protocol (ARP, Address Resolution Protocol) report Text, the source MAC of message are the virtual mac address of counterlogic backup group 2.Belong to the subsequent transmission of user of backup group 2 Data message is forwarded to BRAS3 by interchanger, and BRAS3 completes the forwarding that user accesses Internet message.

S416, BRAS3, which start to receive, belongs to the broadband access request message of backup group 2, authenticates to user, authorization and Charging processing.When new user is online, if belonging to the user of backup group 2, processing is received, writes user data and backup group 2 Id information is to customer data base；When new user offline, then user data is deleted by BRAS3 notice customer data base.

The seventh embodiment of the present invention provides another broadband access system, and networking schematic diagram is as shown in Figure 5.At this In embodiment, hardware module part includes: generic server, user equipment, optical line terminal (OLT, Optical Line Terminal) convergence switch, data center's gateway (DCGW, Data Center GateWay), customer data base, backup are adjusted Spend device, leaf node (leaf) interchanger, pin knot point (Spine) interchanger.Wherein, generic server is for installing virtualization system System, such as OpenStack, can create virtual machine and virtual network, mountable operation application software in each virtual machine, such as BRAS Software, BRAS software can emulated physics BRAS equipment functions；User equipment initiates broadband user's access request, when certification passes through Afterwards, user's ability for accessing Internet is provided；OLT convergence switch is that the service message that docking networks carries out convergence forwarding Layer 2 ethernet switch provides optical cable connectivity port；DCGW is that data center connects inside and outside key node, is used for The translation conversion of content between progress different communication protocol；Customer data base: providing storage, and inquiry updates wide band access user number According to server；Backup scheduler is the server for running backup dispatcher software, or runs the virtual of backup dispatcher software Physical server where machine；Leaf interchanger is defined in data center's distributed kernel network (leaf-spine) framework Leaf leaf node is responsible for connection server and the network equipment；Spine interchanger is spine defined in leaf-spine framework Pin knot point, for connecting interchanger, offer postpones low-down nonblocking performance between any two port in guarantee node, It can satisfy the surface speed forwarding of tens of thousands of servers by certain port convergence ratio/super proportion.

The step of the present embodiment, is described in detail below with reference to Fig. 5.

S501 creates 3 virtual machines in generic server, starts BRAS simulation software in each virtual machine, respectively BRAS1 (is equivalent to the equipment in first embodiment of the invention), BRAS2, BRAS3, distributes and injects for each BRAS virtual machine IP is listened in unique identification ID, i.e. broadband inserting service ID, such as IP address, the NETCONF service for configuring each BRAS virtual machine Address and port numbers.

S502 starts backup scheduler application software, the IP address of configuration backup scheduler and port in generic server Number.

S503 starts customer data base service on database server, configures IP address and the port of customer data base Number.

S504, backup scheduler is as NETCONF client, configuration every BRAS virtual machine of connection.

S505, configures the customer data base address that each BRAS virtual machine needs to connect on backup scheduler, and backup is adjusted Device is spent by the channel NETCONF, and the customer data base address information configured on backup scheduler is issued on BRAS virtual machine, After BRAS virtual machine receives customer data base address, the TCP begun setting up between BRAS virtual machine and customer data base is reliable Connection.

S506, configures 3 backup logic groups, respectively backup group 1, backup group 2, backup group 3, in each backup group plus Enter BRAS1, BRAS2, BRAS3, the matching condition of backup group is to the source MAC value modulus 3 in broadband access request message, to appointing The value obtained after what 3 operation of MAC Address value modulus may be 0,1,2.Backup logic group after the completion of configuration, in backup scheduler Information is as shown in table 5:

Table 5

Backup group	BRAS1	BRAS2	BRAS3	Matching condition
					Backup group 1	It is primary	It is spare	It is spare	MAC mould 3 value in source is equal to 0
Backup group 2	It is spare	It is primary	It is spare	MAC mould 3 value in source is equal to 1
					Backup group 3	It is spare	It is spare	It is primary	MAC mould 3 value in source is equal to 2

S507, backup scheduler dispatches backup logic group information to BRAS1, BRAS2, BRAS3.

S408, user equipment send broadband access request message (such as DHCP message) first, and BRAS virtual machine receives message Afterwards, broadband access protocol massages, such as the feature of DHCP protocol message are judged whether it is according to the feature field carried in message Field includes: that (1) EtherType is equal to 0x0800, and (2) IP protocol fields are equal to 17, and (3) UDP port number is 67 or 68.

After S509, BRAS virtual machine receive broadband access protocol massages, backup logic group information is begun stepping through, and to width 3 operation of mould is carried out with the source MAC in access protocol message, for example, source MAC is the message of 00:00:00:00:00:3C, carries out mould After 3 operations, obtained remainder is that 0, BRAS1 virtual machine searches backup logic group information, and discovery backup group 1 meets matching condition, And information in backup group 1 show BRAS1 be it is primary, therefore, BRAS1 virtual machine starts to process this protocol massages, with user Equipment interaction protocol message authenticates user's access, authorization and charging, saves the number of users generated in access treatment process According to.Meanwhile BRAS1 virtual machine sends user data and affiliated backup group ID to customer data base, is saved by customer data base With this user data of update.And BRAS2 virtual machine receives protocol massages, and after carrying out 3 operation of mould to source MAC, obtained remainder Backup logic group information is searched for 0, BRAS2 virtual machine, discovery backup group 1 meets matching condition, still, in backup logic group 1 In, mark BRAS2 be it is spare, therefore, BRAS2 virtual machine abandons this protocol massages, is not further processed.Source MAC is 00: The message of 00:00:00:00:3D, after carrying out 3 operation of mould, obtained remainder is 1, then BRAS2 virtual machine handles this protocol massages, BRAS1 and BRAS3 virtual machine abandons after receiving this message, is not further processed.Source MAC is 00:00:00:00:00:3E's Message, after carrying out 3 operation of mould, obtained remainder is 2, then BRAS3 virtual machine handles this protocol massages, and BRAS1 and BRAS2 are virtual Machine abandons after receiving this message, is not further processed.

S510 keeps communication between backup scheduler and BRAS1, BRAS2, BRAS3, constantly detecting BRAS virtual machine is It is no online.

S511, backup scheduler counts the load condition of the BRAS virtual machine of each connection state, for example statistics is each The cpu busy percentage of BRAS virtual machine.

S512, when backup scheduler detects that BRAS virtual machine is not present, such as some service of operation BRAS virtual machine Device power down or exception.Such as when backup scheduler detects that BRAS3 virtual machine is not present, backup scheduler begins stepping through all Backup logic group information, discovery BRAS3 are the primary broadband inserting service entities of backup logic group 3, and BRAS1 and BRAS2 are to patrol Collect the spare BRAS virtual machine of backup group 3.Backup scheduler selects the BRAS virtual machine work that CPU is most not busy according to the strategy of configuration For new primary broadband inserting service entity, it is assumed that cpu busy percentage is that cpu busy percentage is in 10%, BRAS2 in BRAS1 at this time 80%, then primary access service entity of the BRAS1 as backup logic group 3 is selected, backup scheduler updates backup logic group letter Breath, after update as shown in table 6:

Table 6

Backup group	BRAS1	BRAS2	BRAS3	Matching condition
					Backup group 1	It is primary	It is spare	It is spare	MAC mould 3 value in source is equal to 0
Backup group 2	It is spare	It is primary	It is spare	MAC mould 3 value in source is equal to 1
					Backup group 3	It is primary	It is spare	It is spare	MAC mould 3 value in source is equal to 2

S513, backup scheduler notify updated backup logic group information to find oneself to BRAS1 and BRAS2, BRAS1 Become primary from spare in backup logic group 3, then customer data base is notified to obtain all users that backup logic group ID is equal to 3 Data.After BRAS1 gets user data, restore the access ring of all users of the backup logic group ID equal to 3 in BRAS1 Border, including establishing user's table that MAC and Session is index, the traffic statistics of increase user and access customer number in BRAS1, The charging message for starting to process all users of the backup logic group ID equal to 3 updates.

S514, BRAS1 equipment send gratuitous ARP packet, and the source MAC of message is the virtual of counterlogic backup group 3 MAC Address, the data message for belonging to the subsequent transmission of user of backup group 3 are forwarded to BRAS1 by leaf interchanger, and BRAS1 is complete The forwarding of Internet message is accessed at user.

S515, BRAS1, which start to receive, belongs to the broadband access request message of backup group 3, authenticates to user, authorization and Charging processing.When new user is online, if belonging to the user of backup group 3, processing is received, writes user data and backup group 3 Id information is to customer data base；When new user offline, if BRAS1 notice customer data base deletes user data.

Although for illustrative purposes, the preferred embodiment of the present invention has been disclosed, those skilled in the art will recognize It is various improve, increase and replace be also it is possible, therefore, the scope of the present invention should be not limited to the above embodiments.

Claims (14)

1. a kind of method for switching master/slave device characterized by comprising

Indicate that host apparatus starts reservation service, wherein the reservation service includes at least: user data is written to number of users According in library；

Whether the working condition for detecting the host apparatus is normal；

In the case where the working condition exception of the host apparatus, a stand-by equipment conduct is determined according to preset configuration strategy New host apparatus；

Indicate the new host apparatus to resume work from the customer data base user that state is abnormal host apparatus Data.

2. the method as described in claim 1, which is characterized in that the instruction new host apparatus is from the customer data base After user data of the state of resuming work for abnormal host apparatus, further includes:

Indicate that the new host apparatus is described pre- according to the user data starting that the working condition is abnormal host apparatus Fixed service.

3. the method as described in claim 1, which is characterized in that instruction host apparatus starts reservation service, comprising:

N number of backup logic group is constructed, and selects a default equipment as primary in M equipment in each backup logic group Equipment, stand-by equipment of other equipment as the host apparatus in addition to the host apparatus, wherein N is whole greater than 0 Number, M are the integer greater than 1；

The information of each backup logic group in N number of backup logic group is sent to the backup logic group is corresponding primary to be set It is standby, so that the host apparatus starts reservation service.

4. method as claimed in claim 3, which is characterized in that before constructing N number of backup logic group, further includes:

Obtain the address information of N number of equipment；

Obtain the address information of the customer data base；

The communication channel of each equipment and the customer data base in N number of equipment is established according to the address information of acquisition.

5. method as claimed in claim 3, which is characterized in that the information of N number of backup logic group is sent to each set After standby, further includes:

The backup logic group where pre-set user equipment is determined according to preset matching rule, wherein the preset matching rule is The matching rule of each backup logic group in N number of backup logic group；

User's request of the pre-set user equipment is forwarded to the host apparatus of the backup logic group.

6. the method as described in any one of claim 3 to 5, which is characterized in that the method also includes:

Whether the host apparatus of test job abnormal state restores normal operating conditions to schedule；

In the case where the host apparatus of the working condition exception restores normal operating conditions, the new host apparatus is become The stand-by equipment of the more described current logic backup group, and abnormal primary of the working condition that normal operating conditions will be restored Equipment changing is host apparatus.

7. a kind of method of restoring user data characterized by comprising

Receive the switching command of backup scheduler dispatches；

Restore the user data of the first equipment from customer data base according to the switching command；

Start reservation service according to the user data, wherein the reservation service includes at least: user data is written to use In user data library.

8. a kind of server for switching master/slave device includes at least memory, processor, is stored with computer on the memory Program, which is characterized in that the processor realizes following steps when executing the computer program on the memory:

Indicate that host apparatus starts reservation service, wherein the reservation service includes at least: user data is written to number of users According in library；

Whether the working condition for detecting the host apparatus is normal；

In the case where the working condition exception of the host apparatus, a stand-by equipment conduct is determined according to preset configuration strategy New host apparatus；

Indicate the new host apparatus to resume work from the customer data base user that state is abnormal host apparatus Data.

9. server as claimed in claim 8, which is characterized in that the processor executes instruction the new host apparatus from institute After stating the step of resuming work all customer data that state is abnormal host apparatus in customer data base, also execute following Step: the instruction new host apparatus is described pre- according to all customer data starting that working condition is abnormal host apparatus Fixed service.

10. server as claimed in claim 8, which is characterized in that the processor executes instruction the predetermined clothes of host apparatus starting When the step of business, following steps are executed:

N number of backup logic group is constructed, and selects a default equipment as primary in M equipment in each backup logic group Equipment, stand-by equipment of other equipment as the host apparatus in addition to the host apparatus, wherein N is whole greater than 0 Number, M are the integer greater than 1；

The information of each backup logic group in N number of backup logic group is sent to the backup logic group is corresponding primary to be set It is standby, so that the host apparatus starts reservation service.

11. server as claimed in claim 10, which is characterized in that the processor executes the step for constructing N number of backup logic group Before rapid, following steps are also executed:

Obtain the address information of N number of equipment；

Obtain the address information of the customer data base；

The communication channel of each equipment and the customer data base in N number of equipment is established according to the address information of acquisition.

12. server as claimed in claim 10, which is characterized in that the processor is executed N number of backup logic group Information the step of being sent to each equipment after, also execution following steps:

The backup logic group where pre-set user equipment is determined according to preset matching rule, wherein the preset matching rule is The matching rule of each backup logic group in N number of backup logic group；

User's request of the pre-set user equipment is forwarded to the host apparatus of the backup logic group.

13. the server as described in any one of claim 10 to 12, which is characterized in that the processor also executes following step It is rapid:

Whether the host apparatus of test job abnormal state restores normal operating conditions to schedule；

In the case where the host apparatus of the working condition exception restores normal operating conditions, the new host apparatus is become The stand-by equipment of the more described current logic backup group, and abnormal primary of the working condition that normal operating conditions will be restored Equipment changing is host apparatus.

14. a kind of network equipment includes at least memory, processor, is stored with computer program, feature on the memory It is, the processor realizes following steps when executing the computer program on the memory:

Receive the switching command of backup scheduler dispatches；

Restore the user data of the first equipment from customer data base according to the switching command；

Start reservation service according to the user data, wherein the reservation service includes at least: user data is written to use In user data library.