CN117014338A - Method and device for actively detecting and processing multi-master situation in stacked scene - Google Patents

Abstract

The invention relates to a method and a device for actively detecting and processing multi-master situations in a stacking scene, which utilize a stacking link to carry out fault detection, judge and split, before a stacking system is split into a plurality of subsystems to reconstruct the stacking system, each stacking device determines a stacking device list comprising all stacking device information of the subsystem, each stacking device sends the stacking device list of the stacking device or the stacking device role information to proxy equipment, each stacking device combines the stacking device role information and a priority mechanism fed back from the proxy equipment, or receives instructions made by the proxy equipment combined with the stacking device list and the priority mechanism of all stacking devices, judges whether to close the service of the stacking device and execute the service, so that one subsystem of the stacking system is kept normal, the multi-master problem can be solved, compared with the existing passive detection method, the method can be independently constructed by a new stacking subsystem, actively and rapidly complete multi-master detection and processing, and effectively reduces network fault time.

Description

Method and device for actively detecting and processing multi-master situation in stacked scene

Technical Field

The invention belongs to the technical field of network equipment stacking, and particularly relates to a method and a device for actively detecting and processing multi-master situations in a stacking scene.

Background

In a network device stacking scene, a plurality of member devices are interconnected and stacked through stacking links to form a logic device to provide expansion capability, so that collaborative work, unified management and uninterrupted maintenance of the plurality of member devices are realized. The roles of stacking devices in the stacking system include a master stacking device (master), a standby stacking device (standby) and a slave stacking device (slave), and a plurality of stacking devices are communicated through stacking links, each stacking device is connected with a proxy device through a physical connection, if the stacking links or member devices in the stacking system fail, stacking split can be caused, a plurality of stacking systems with the same configuration can occur in a network, which is the multi-master situation of a stacking scene, and particularly, the same three-layer IP address and the same two-layer mac address can cause data forwarding failure in the network, so that the multi-master situation after stacking split needs to be detected and processed to avoid the multi-master problem.

There are several schemes to detect multi-master situations, such as: the Cisco equipment supports double-master detection, detects the occurrence of double-master conditions through a private protocol PAgP (PortAggregationProtocol), and closes a stacking system where the original master equipment is located to solve the problem; the Hua is that the equipment detects the occurrence of Multi-master through a private protocol MAD (Multi-Activity detection), and determines a closed stacking system by comparing the starting time of the stacking system and the priority of a main switch of the stacking system; the Hua is that the equipment detects the occurrence of Multi-master through a private protocol MAD (Multi-Activity detection), and determines a closed stacking system by comparing the starting time of the stacking system and the priority of a main switch of the stacking system; H3C detects multi-master conditions by standard protocols LACP (LinkAggregationControlProtocol), BFD (BidirectionalForwardingDetection), ARP (AddressResolutionProtocol), etc., and determines a closed stacking system by the number of stacking system member devices or stacking master priority of the stacking system.

Although the above detection method can solve the multi-master problem by closing the partial stack system, it has the following drawbacks: when the stack is split, if the original stack main or stack standby is not available in the new stack subsystem, the stack subsystem member device will restart, after the new stack main is restarted, the multi-main condition can be detected and processed through the existing multi-main detection mechanism, and the time of the process may be different from tens of seconds to minutes, so that the existing multi-main detection processing time is longer when the new stack main is selected for the split stack subsystem due to the passive detection mechanism built by the new stack subsystem.

Disclosure of Invention

The invention aims to solve at least one of the technical problems to a certain extent, and provides a method and a device for actively detecting and processing multi-master situations in a stacking scene, which are independent of new stacking subsystem construction, and can actively and rapidly complete multi-master detection and processing before restarting a stacking system after stacking splitting, thereby reducing network failure time.

The technical scheme adopted for solving the technical problems is as follows:

a method for actively detecting and processing multi-master situations in a stacked scene, the method comprising:

before the stacking system is split into a plurality of subsystems to reconstruct the stacking system, each stacking device determines a stacking device list comprising all stacking device information of the subsystem, and each stacking device sends the stacking device list of the stacking device or the stacking device role information to the proxy device;

setting a priority mechanism, wherein each stacking device combines the stacking device role information and the priority mechanism fed back from the proxy device, or receiving instructions which are combined by the proxy device with the stacking device lists and the priority mechanisms of all the stacking devices, judging whether to close the service of the stacking device and execute the service, so that one subsystem of the stacking system keeps the service normal.

Further, each stacking device of the stacking system periodically sends detection messages to other stacking devices, and each stacking device receives reply messages fed back by the other stacking devices and determines that the stacking device is communicated with the corresponding stacking device according to the reply messages;

each stacking device maintains a stacking device list comprising the stacking device information and all other stacking device information communicated with the stacking device, and each stacking device sends the stacking device list of the stacking device to the proxy device;

and the proxy equipment judges that the stacking system is split according to the fact that the stacking equipment list of any stacking equipment does not comprise all stacking equipment information of the stacking system.

Further, the communication detection period of each stacking device sending detection messages to other stacking devices is less than or equal to the stacking heartbeat detection period;

setting service fault convergence time, and determining that the stacking device is communicated with the corresponding stacking device according to the fact that the time from sending the detection message to receiving the reply message is less than or equal to the service fault convergence time by each stacking device.

Further, roles of all stacking devices of the stacking system include stacking master, stacking slave and stacking slave, and a priority mechanism is set according to the number of the stacking master, the stacking slave and the stacking devices included in the subsystem.

Further, if any subsystem includes a stacking master, the stacking device to which the stacking device belongs sends the stacking device list and stacking master information of the stacking device to the proxy device, and after each stacking device receives the stacking master information fed back from the proxy device, the subsystem including the stacking master keeps normal service, and other subsystems close service.

Further, a priority mechanism is set according to the priority that the stacking master is higher than the stacking slave, if any subsystem comprises the stacking slave, the stacking device belonging to the stacking device sends a stacking device list and stacking slave information of the stacking device to the proxy device, each stacking device receives the stacking slave information fed back from the proxy device and does not receive the stacking master information, the subsystem comprising the stacking slave keeps normal service, and other subsystems close the service.

Further, a priority mechanism is set according to the priority of stacking devices higher than stacking slaves and the fact that the number of stacking slaves is more than the secondary priority, each stacking device receives an instruction of the proxy device and executes the instruction, after the proxy device does not receive stacking main information and stacking standby information, any subsystem with the largest number of stacking slaves is selected according to a stacking device list of all stacking devices to keep normal service, and other subsystems close service.

Further, the stacking device information comprises a physical mac, a priority mechanism is set according to the number of stacking slaves and the address size of the stacking slave physical mac, each stacking device receives an instruction of the proxy device and executes the instruction, after the proxy device does not receive the stacking master information and the stacking slave information, the subsystem with the largest stacking slave number and the smallest stacking slave physical mac address is selected to keep normal service according to the stacking device list of all the stacking devices, and other subsystems close service.

Further, after the service of the stacking system is kept normal, stacking recovery detection processing is carried out to reconstruct the stacking system.

An apparatus for actively detecting and processing multi-master situations in a stacked scene, comprising:

the detection module is used for determining a stacking device list of the stacking device comprising all stacking device information of the subsystem after the stacking system is split into a plurality of subsystems;

the first transceiver module is used for each stacking device to send the stacking device list or the stacking device role information of the stacking device to the proxy device and receive the stacking device role information or instructions fed back from the proxy device;

the first processing module is used for setting a priority mechanism for each stacking device, and judging whether to close the service of the stacking device and execute the service by combining the role information of the stacking device of the first transceiver module and the priority mechanism or the instruction;

the second transceiver module is used for receiving a stacking device list or stacking device role information of each stacking device by the agent device and feeding back the stacking device role information or instructions to each stacking device;

the second processing module is used for judging the split of the stacking system by the agent equipment according to the stacking equipment list of the second transceiver module, setting a priority mechanism, and generating instructions by combining the stacking equipment list of all the stacking equipment and the priority mechanism after the stacking system is split into a plurality of subsystems.

Further, the detection module is used for detecting connectivity of the stacking device and other stacking devices by each stacking device, and determining a stacking device list according to the connectivity.

Compared with the prior art, the invention has the beneficial effects that:

(1) Before a stack system is rebuilt after the stack of the existing mechanism is split, once the multi-master scene is detected, fault processing is immediately carried out, whether the service of the stack device is closed or not is judged and executed by combining the stack device role information fed back by the proxy device with the priority mechanism or receiving instructions made by combining the proxy device with the stack device list and the priority mechanism of all the stack devices, multi-master detection and processing can be actively and rapidly completed, one subsystem of the stack system keeps normal service, the multi-master problem can be solved, and compared with the existing passive detection method, the construction of a new stack subsystem is not relied on, so that the time of flow forwarding problem caused by the same service configuration in a network can be greatly shortened.

(2) According to the method, a stacking device list is maintained by detecting connectivity of each stacking device and other stacking devices, fault detection judgment splitting is carried out by using stacking links, and stacking splitting problems caused by any faults including stacking device faults, link faults and the like are detected and processed.

(3) The method sets the priority mechanism for the priority order according to the stack main, stack standby, the greater number of stack devices and the smaller stack slave physical mac address included in the subsystem, so that the availability of the stack system before the stack system is rebuilt after stack splitting can be further optimized.

Drawings

The foregoing and/or additional aspects and advantages of the invention will become apparent and may be better understood from the following description of embodiments taken in conjunction with the accompanying drawings in which:

FIG. 1 is a flow chart of a method of an embodiment of the present invention.

Fig. 2 is a flow chart of the priority mechanism determination of the method of the present invention.

FIG. 3 is a schematic diagram of the real-time example 1 chain stacking scenario to link failure scenario of the present invention;

FIG. 3a shows a chain stacking scenario in a normal state; FIG. 3b illustrates a link failure scenario of a chain stacking scenario; FIG. 3c shows a multi-master detection processing scenario after a stack system split.

FIG. 4 is a schematic diagram of the present invention real-time example 2 annular stacking scenario to stacking device failure scenario;

FIG. 4a shows a loop stacking scenario in a normal state; FIG. 4b illustrates a stacking device failure scenario for a ring stack scenario; FIG. 4c shows a multi-master detection processing scenario after a stack system split.

Fig. 5 is a schematic view showing the structure of the apparatus according to the embodiment 3 of the present invention.

Detailed Description

Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to like or similar elements or elements having like or similar functions throughout. The embodiments described below by referring to the drawings are illustrative and intended to explain the present invention and should not be construed as limiting the invention.

The terms "first," "second," and the like, are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include one or more such feature. In the description of the present invention, the meaning of "a plurality" or "a number" means two or more, unless specifically defined otherwise. The terms "coupled," "connected," and the like are to be construed broadly and may be mechanically coupled or electrically coupled, for example; can be directly connected or indirectly connected through an intermediate medium, and can be communication between two elements. The specific meaning of the above terms in the present invention can be understood by those of ordinary skill in the art according to the specific circumstances.

Because the existing multi-main detection method can solve the multi-main problem by closing part of the stacking system, the processing time is longer because the passive detection is built by the new stacking subsystem after the new stacking main is selected for the split stacking subsystem; in order to reduce network failure time, the invention provides a method for actively detecting and processing multi-master situations in a stacked scene, which comprises the following steps:

before the processing time is when the stacking system is split into a plurality of subsystems to reconstruct the stacking system, each stacking device determines a stacking device list comprising all stacking device information of the subsystem to which each stacking device belongs, and each stacking device sends the stacking device list of the stacking device or stacking device role information to the proxy device;

setting a priority mechanism, wherein each stacking device combines stacking device role information and the priority mechanism fed back by the proxy device, or receiving instructions which are combined by the proxy device with stacking device lists and the priority mechanism of all stacking devices, judging whether to close the service of the stacking device and execute the service, so that one subsystem of the stacking system keeps normal service, multi-master detection is performed when stacking normally, and once multi-master scenes are detected, fault processing can be immediately performed, multi-master detection and processing can be actively and rapidly completed, so that the multi-master problem can be solved, compared with the existing passive detection method, the method can be constructed without depending on a new stacking subsystem, and the time of flow forwarding problem caused by the same service configuration in a network can be greatly shortened.

As shown in fig. 1-2, the method further comprises:

s1, performing stacking heartbeat detection and communication detection after a stacking system is normally established, and specifically comprising the following steps:

s101: the stacking system is established by n stacking devices through stacking link communication, roles of all the stacking devices comprise a stacking master, a stacking standby and a stacking slave, and each stacking device is connected with the proxy device through a physical connection line.

S102: each stacking device periodically sends detection messages to other stacking devices through the stacking ports of the stacking devices, the communication detection period of each stacking device sending the detection messages to other stacking devices is less than or equal to the stacking heartbeat detection period, if the stacking heartbeat detection period is 1s, the period of sending the detection messages can be less than or equal to 1s.

S103: setting service fault convergence time, wherein each stacking device receives reply messages fed back by other stacking devices, and each stacking device determines that the stacking device is communicated with the corresponding stacking device according to the fact that the time from sending the detection message to receiving the reply message is less than or equal to the service fault convergence time, and can be flexibly adjusted according to the service fault convergence time requirement as an accessibility mechanism, for example, if no reply message of the opposite stacking device is received in three stacking heartbeat detection periods, the stacking device can judge that the service is unreachable.

S104: thus, each of the stack devices maintains a stack device list including the present stack device information and all other stack device information communicated with the present stack device, and each of the stack devices is identified by using the stack ID and/or one physical mac as the stack device information, the stack device information of any one of the stack devices is denoted as Di, and if the stack system is normal, the stack device list of each of the stack devices may be denoted as { D1, D2, D3 … Dn }, and each of the stack devices periodically transmits the stack device list of the present stack device to the proxy device.

When some stacking devices in the stacking system are not accessed according to the above accessibility mechanism due to component failure or stacking link failure, the devices are removed from the stacking device list of the stacking devices which are not accessed by other stacking devices, and as each stacking device periodically sends the stacking device list of the stacking device to the proxy device, the proxy device can judge the stacking system split according to the fact that the stacking device list of any stacking device does not comprise all stacking device information of the stacking system, determine the stacking system fault, and perform fault detection judgment split by using the stacking link, so that the stacking split problem caused by any faults including stacking device faults, link faults and the like can be further detected and processed.

S2, performing multi-master detection processing before the stacking system is split into a plurality of subsystems to reconstruct the stacking system, wherein the multi-master detection processing specifically comprises the following steps: each stacking device determines a stacking device list comprising all stacking device information of the subsystem to which the stacking device belongs, the stacking device list of each stacking device in each subsystem is consistent, and a priority mechanism is set according to the greater number of stacking masters, stacking backups and stacking slave physical mac addresses included in the subsystem and the smaller priority order, so that each stacking device processes as follows:

if the stacking device list of the stacking device contains a stacking master, the stacking device sends the stacking device list of the stacking device and stacking master information to the proxy device, the proxy device sends message broadcast containing the stacking master information to all the stacking devices, and the stacking device keeps normal service after receiving the stacking master information fed back from the proxy device.

If the stacking equipment list of the stacking equipment does not contain the stacking main and only contains the stacking standby, the stacking equipment sends the stacking equipment list of the stacking equipment and the stacking standby information to the proxy equipment, the proxy equipment sends the message broadcast containing the stacking standby information to all the stacking equipment, if the stacking equipment receives the stacking main information fed back by the proxy equipment, the service of the stacking equipment is closed, and otherwise, the service is kept normal.

If the stacking device list of the stacking device does not contain stacking master or stacking standby, the stacking device sends the stacking device list of the stacking device to the proxy device, if the proxy device does not receive stacking master information and stacking standby information, the stacking device keeps normal service according to the stacking device list of all the stacking devices, the stacking devices with the largest stacking slave quantity in the stacking device list and the smallest stacking slave physical mac address are judged to belong to the stacking device, the rest stacking device service is closed, an instruction is generated, a message broadcast containing the instruction is sent to all the stacking devices, and after the stacking device receives the instruction fed back from the proxy device, the stacking device keeps normal service or is closed according to the instruction.

After each stacking device is processed, if any subsystem comprises a stacking main, selecting the subsystem comprising the stacking main to keep the service normal, and closing the service of other subsystems; when no subsystem comprises a stacking master, and any subsystem comprises a stacking slave, selecting the subsystem comprising the stacking slave to keep normal service, and closing the service by other subsystems; under the condition that no subsystem comprises a stacking master and no subsystem comprises a stacking standby, selecting the subsystem with the largest number of stacking slaves to keep normal service, and closing the service by other subsystems; in the case that no subsystem comprises a stacking master and no subsystem comprises a stacking slave, and a plurality of subsystems with the largest stacking slave are included, selecting the subsystem with the largest stacking slave and the smallest stacking slave physical mac address to keep normal service, and closing the service by other subsystems; the method has the advantages that one subsystem of the stacking system is kept normal, multi-master detection and processing can be actively and rapidly completed, the multi-master problem can be solved, the time of flow forwarding problem caused by the same service configuration in a network is greatly shortened, and the availability of the stacking system before the stacking system is rebuilt after stacking split can be further optimized through an optimization mechanism.

S3, performing stack recovery detection processing, and rebuilding the stack system, wherein the stack recovery detection processing is the same as the existing mechanism, so that the description is omitted.

Example 1:

as shown in fig. 3, a method for actively detecting and handling multiple primary situations in a stacked scenario according to the present invention includes:

and establishing a chain-shaped stacking system by n stacking devices through stacking link communication in sequence, wherein the stacking device with stacking device information Dn has the role of a stacking master, the stacking device with stacking device information Dn-1 has the role of a stacking standby, the other stacking devices have the role of stacking slave, if the chain-shaped stacking system is normal, the stacking device list of each stacking device can be represented as { D1, D2, D3 … Dn }, and each stacking device periodically transmits the stacking device list of the stacking device to the proxy device.

When a stacking link between stacking devices with stacking device information Di-1 and Di and between stacking devices with di+1 and di+2 fails, other stacking devices cannot access the stacking devices with stacking device information Di and di+1, the other stacking devices remove the two stacking devices which are inaccessible from the stacking device list, the proxy device can judge that the stacking system is split into a plurality of subsystems according to the stacking device list of any stacking device not including all stacking device information of the stacking system, each stacking device determines the stacking device list including all stacking device information of the subsystem, the stacking device list of each stacking device of each subsystem is consistent, and when the stacking system is split into three subsystems which are respectively marked as (1), (2) and (3), then:

the stacking device list of each stacking device belonging to the subsystem (1) is { D1, D2 … Di-1};

the stacking device list of each stacking device belonging to the subsystem (2) is { Di, di+1};

the list of stacked devices belonging to the subsystem (3) is { Di+2 …, dn-1, dn }.

The stacking device in the stacking system will detect the fault at about the same time after the fault occurs, and the processing before the stacking system is split into multiple subsystems to reconstruct the stacking system is as follows:

each stacked device belonging to the subsystem (1) transmits a stacked device list { D1, D2 … Di-1} to the proxy device;

each stacking device belonging to the subsystem (2) sends a stacking device list { Di, di+1} to the proxy device;

each stacking device belonging to the subsystem (3) transmits a stacking device list { di+ …, dn-1, dn }, stacking master information, and stacking slave information to the proxy device;

the agent equipment sends message broadcast containing stacking main information and stacking standby information to all stacking equipment, and each stacking equipment belonging to the subsystem (1) closes the service after receiving the stacking main information and the stacking standby information fed back by the agent equipment from the agent equipment; after each stacking device belonging to the subsystem (2) receives stacking main information and stacking standby information fed back by the slave agent device of the agent device, closing the service; after each stacking device belonging to the subsystem (3) receives the stacking main information and stacking standby information fed back by the proxy device, the stacking device keeps the service normal, and can select the subsystem comprising the stacking main and stacking standby to keep the service normal under the condition of link failure, and other subsystems close the service, so as to actively complete multi-main detection and processing and reduce the network failure time.

Example 2:

as shown in fig. 4, a method for actively detecting and handling multiple primary situations in a stacked scenario according to the present invention includes:

the n stacking devices sequentially communicate through stacking links to form a chain, the chain-shaped head and tail stacking devices communicate through the stacking links to establish an annular stacking system, the role of the stacking device with stacking device information Dn is a stacking master, the role of the stacking device with stacking device information Dn-1 is a stacking standby, the roles of other stacking devices are stacking slave, if the annular stacking system is normal, the stacking device list of each stacking device can be represented as { D1, D2, D3 … Dn }, and each stacking device periodically sends the stacking device list of the stacking device to the proxy device.

When the stacking device with stacking device information of D1 and the stacking device of Di fail due to power failure or the like, the stacking device with stacking device information of D1 and the stacking device of Di cannot be accessed by other stacking devices, the other stacking devices remove the two stacking devices which are not reachable from the stacking device list, the proxy device can judge that the stacking system is split into a plurality of subsystems according to the fact that the stacking device list of any stacking device does not comprise all stacking device information of the stacking system, each stacking device determines the stacking device list comprising all stacking device information of the subsystem, the stacking device list of each stacking device of each subsystem is consistent, and when the stacking system is split into three subsystems, namely (1) and (2), the proxy device can determine that the stacking system is split into a plurality of subsystems according to the stacking device list of any stacking device which does not comprise all stacking device information of the stacking system:

the stacking device list of each stacking device belonging to the subsystem (1) is { D2, D3 … Di-1};

the list of stacked devices belonging to the subsystem (2) is { Di+1, di+2 … Dn }.

The stacking device in the stacking system will detect the fault at about the same time after the fault occurs, and the processing before the stacking system is split into multiple subsystems to reconstruct the stacking system is as follows:

each stacked device belonging to the subsystem (1) transmits a stacked device list { D2, D3 … Di-1} to the proxy device;

each stacking device belonging to the subsystem (2) transmits a stacking device list { di+1, di+2 … Dn }, stacking master information, and stacking slave information to the proxy device;

the agent equipment sends message broadcast containing stacking main information and stacking standby information to all stacking equipment, and each stacking equipment belonging to the subsystem (1) closes the service after receiving the stacking main information and the stacking standby information fed back by the agent equipment from the agent equipment; after each stacking device belonging to the subsystem (2) receives the stacking main information and stacking standby information fed back by the proxy device, the stacking device keeps the service normal, and can select the subsystem comprising the stacking main and stacking standby to keep the service normal under the condition of stacking device faults, and other subsystems close the service, so as to actively complete multi-main detection and processing and reduce network fault time.

Example 3:

as shown in fig. 5, based on the same inventive concept, an embodiment of the present invention further provides a device for actively detecting and processing multiple main situations in a stacked scene, including:

the detection module is used for detecting the connectivity of the stacking device and other stacking devices by each stacking device, and accordingly determining a stacking device list of the stacking device comprising all stacking device information of the subsystem after the stacking system is split into a plurality of subsystems;

the first transceiver module is used for each stacking device to send the stacking device list or the stacking device role information of the stacking device to the proxy device and receive the stacking device role information or instructions fed back from the proxy device;

the first processing module is used for setting a priority mechanism for each stacking device, and judging whether to close the service of the stacking device and execute the service by combining the role information of the stacking device of the first transceiver module and the priority mechanism or the instruction;

the second transceiver module is used for receiving a stacking device list or stacking device role information of each stacking device by the agent device and feeding back the stacking device role information or instructions to each stacking device;

the second processing module is used for judging the split of the stacking system by the agent equipment according to the stacking equipment list of the second transceiver module, setting a priority mechanism, and generating instructions by combining the stacking equipment list of all the stacking equipment and the priority mechanism after the stacking system is split into a plurality of subsystems.

Further, the detection module, the first transceiver module and the first processing module are configured on each stacking device, by periodically sending detection messages to other stacking devices, receiving reply messages fed back by other stacking devices, setting service fault convergence time, determining that each stacking device is communicated with the corresponding stacking device according to the time from sending the detection messages to receiving the reply messages, and maintaining an reachable stacking device list according to connectivity conditions, wherein each stacking device comprises the stacking device information and all other stacking device information communicated with the stacking device, and the first transceiver module periodically sends the stacking device list of the stacking device to the proxy device.

Further, the stacking device role information includes stacking main information and stacking standby information, the second transceiver module and the second processing module are configured on the proxy device, the second processing module is configured to determine that the stacking system is split into a plurality of subsystems, and after the second transceiver module receives the stacking device role information, control the second transceiver module to feed back the stacking device role information to each stacking device, and further when the stacking system fails, select a subsystem capable of keeping normal service according to a priority mechanism set by stacking main, standby or stacking member number conditions and the like contained in each subsystem.

The above list of detailed descriptions is only specific to practical embodiments of the present invention, and they are not intended to limit the scope of the present invention, and all equivalent embodiments or modifications that do not depart from the spirit of the present invention should be included in the scope of the present invention.

Claims (10)

1. A method for actively detecting and processing multi-master situations in a stacked scene is characterized by comprising the following steps:

before the stacking system is split into a plurality of subsystems to reconstruct the stacking system, each stacking device determines a stacking device list comprising all stacking device information of the subsystem, and each stacking device sends the stacking device list of the stacking device or the stacking device role information to the proxy device;

setting a priority mechanism, wherein each stacking device combines the stacking device role information and the priority mechanism fed back from the proxy device, or receiving instructions which are combined by the proxy device with the stacking device lists and the priority mechanisms of all the stacking devices, judging whether to close the service of the stacking device and execute the service, so that one subsystem of the stacking system keeps the service normal.

2. The method for actively detecting and handling multiple primary situations in a stacking scenario according to claim 1, wherein each stacking device of the stacking system periodically sends detection messages to other stacking devices, and each stacking device receives reply messages fed back by the other stacking devices and determines that the stacking device is communicated with the corresponding stacking device according to the reply messages;

each stacking device maintains a stacking device list comprising the stacking device information and all other stacking device information communicated with the stacking device, and each stacking device sends the stacking device list of the stacking device to the proxy device;

and the proxy equipment judges that the stacking system is split according to the fact that the stacking equipment list of any stacking equipment does not comprise all stacking equipment information of the stacking system.

3. The method for actively detecting and processing multi-master situations in a stacked scenario according to claim 2, wherein a communication detection period of each stacked device sending a detection message to other stacked devices is less than or equal to a stacked heartbeat detection period;

setting service fault convergence time, and determining that the stacking device is communicated with the corresponding stacking device according to the fact that the time from sending the detection message to receiving the reply message is less than or equal to the service fault convergence time by each stacking device.

4. The method for actively detecting a stacked scenario for processing a multi-master situation according to claim 1, wherein roles of all stacked devices of the stacking system include a stacked master, a stacked slave, and wherein a priority mechanism is set according to the number of stacked masters, stacked slaves, and stacked devices included in the subsystem.

5. The method for actively detecting and handling multiple main situations in a stacking scenario according to claim 4, wherein if any subsystem includes a stacking main, the stacking device to which the stacking device belongs sends a stacking device list and stacking main information of the stacking device to the proxy device, and after each stacking device receives the stacking main information fed back from the proxy device, the subsystem including the stacking main keeps normal service, and other subsystems close service.

6. The method for actively detecting and handling multiple primary situations in a stacking scenario according to claim 5, wherein a priority mechanism is set according to a priority of a stacking primary being higher than a stacking standby, if any subsystem includes a stacking standby, the stacking device of the stacking device sends a stacking device list and stacking standby information of the stacking device to the proxy device, each stacking device receives the stacking standby information fed back from the proxy device and does not receive the stacking primary information, the subsystem including the stacking standby keeps normal service, and other subsystems close service.

7. The method for actively detecting and handling multiple master situations in a stacked scenario according to claim 6, wherein the priority mechanism is set according to the priority of stacking devices higher than stacking slaves and the number of stacking slaves is more than the secondary priority, each stacking device receives an instruction of a proxy device and executes the instruction, where after the proxy device does not receive stacking master information and stacking slave information, any subsystem with the largest number of stacking slaves is selected to keep normal service according to a stacking device list of all stacking devices, and other subsystems close service.

8. The method for actively detecting and handling multiple master situations in a stacking scenario according to claim 7, wherein the stacking device information includes a physical mac, a priority mechanism is set according to the number of stacking slaves and the address size of the stacking slave physical mac, each stacking device receives an instruction of a proxy device and executes the instruction, after the proxy device does not receive the stacking master information and the stacking slave information, the subsystem including the largest number of stacking slaves and the smallest stacking slave physical mac address is selected to keep normal service according to the stacking device list of all the stacking devices, and other subsystems close service.

9. The method for actively detecting and processing multiple main situations in a stacking scenario according to claim 1, wherein the stacking system is reconfigured by performing a stacking recovery detection process after one subsystem selected by the stacking system keeps normal service.

10. An apparatus for actively detecting and processing multiple master situations in a stacked scene, comprising:

the detection module is used for detecting the connectivity of the stacking device and other stacking devices by each stacking device, and accordingly determining a stacking device list of the stacking device comprising all stacking device information of the subsystem after the stacking system is split into a plurality of subsystems;

the first transceiver module is used for each stacking device to send the stacking device list or the stacking device role information of the stacking device to the proxy device and receive the stacking device role information or instructions fed back from the proxy device;

the first processing module is used for setting a priority mechanism for each stacking device, and judging whether to close the service of the stacking device and execute the service by combining the role information of the stacking device of the first transceiver module and the priority mechanism or the instruction;

the second transceiver module is used for receiving a stacking device list or stacking device role information of each stacking device by the agent device and feeding back the stacking device role information or instructions to each stacking device;

the second processing module is used for judging the split of the stacking system by the agent equipment according to the stacking equipment list of the second transceiver module, setting a priority mechanism, and generating instructions by combining the stacking equipment list of all the stacking equipment and the priority mechanism after the stacking system is split into a plurality of subsystems.