CN106657355B - Cluster management method and device - Google Patents

Abstract

A cluster management method and device, the method is applied to a cluster, the cluster includes a master device and a slave device, the method includes: when a protocol channel between the master device and the slave device fails, the The slave device changes the role of the master device; the slave device sends the cluster detection packet to the master device through the data channel; after receiving the cluster response packet sent by the master device, the slave device responds according to the cluster response message to exit the cluster; the slave device closes the slave device's port. By adopting this solution, the phenomenon that the BGP of the master device and the slave device repeatedly oscillates can be avoided.

Description

Cluster management method and device

technical field

The present invention relates to the technical field of virtual clusters, in particular to a cluster management method and device.

Background technique

Virtual machine cluster technology can reduce operating costs, expand the number of ports on a single node, etc., and is currently used more and more in router products; such as enterprise network scenarios (such as radio and television, China Merchants Bank, Thailand PEA, UK UK, etc.). When the virtual cluster technology is used, after the protocol channel between the master device and the slave device in a cluster fails, there will be a situation of dual master devices.

Due to the presence of dual-master devices, the two master devices will establish Border Gateway Protocol (English full name: Border Gateway Protocol, English abbreviation: BGP) neighbors with the upstream and downstream devices in the current cluster, so that the BGP protocol repeatedly oscillates, and will The route flapping of the network is affected, and the traffic of the node cannot be forwarded normally.

SUMMARY OF THE INVENTION

The invention provides a cluster management method and device, which can solve the problem of repeated oscillation of the BGP protocol of the master device and the slave device in the prior art.

The first aspect provides a cluster management method, which can be applied to a cluster. A cluster refers to a group of mutually independent computers interconnected through a network through a protocol. They form a network and are managed in a single system mode. . The devices in the cluster can share resources, share costs, share channel devices and services, etc. The cluster includes a master device and at least one slave device, and the master device refers to the cluster in which it is used to control. A slave device is used to collect data. Both the master device and the slave device can perform operations such as traffic forwarding between the devices connected to the upstream and downstream communication. When the protocol channel is normal, the device role of the master device is the role of the master device, and the device role of the slave device is the role of the slave device. The method includes:

When the protocol channel between the master device and the slave device fails, the slave device switches to the role of the master device, and then the slave device sends the cluster detection packet to the master device through the data channel. After the master device sends the cluster response message, it can be determined that the cluster to which it belongs has been collected and distributed, and then the slave device can exit the cluster according to the cluster response message.

After exiting the cluster where it is located, the slave device closes the port of the slave device. The closed ports include ge0/0/0, a loopback interface (loopback), a virtual local area network interface (vlanif), a service port, and the like.

Compared with the existing mechanism, in the present invention, after the protocol channel fails, the slave device that is transformed into the role of the master device sends the cluster detection message to the master device through the data channel, and then receives the cluster response sent by the master device. After the message is sent, the port of the slave device is forcibly closed, thereby isolating the original slave device. No processing is performed on the original master device, and the protocol between the original master device and the upstream and downstream devices of the original master device can still be established, so that the original service data can continue to be forwarded, so that the traffic forwarding of the original master device will not be affected. The BGP of the device flaps repeatedly.

In some possible designs, the cluster detection packet includes a first cluster detection field, and the cluster response packet includes a second cluster detection field.

Wherein, the first cluster detection field includes a first indicator bit, a second indicator bit, and a third indicator bit, and the first indicator bit in the first cluster detection field is used to indicate the sender of the cluster detection detection packet. The device role of the slave device before the protocol channel failure, the second indicator bit in the first cluster detection field indicates the device role of the slave device that sent the cluster detection detection packet after the protocol channel failure, so The third indication bit in the first cluster detection field indicates that the slave device applies for closing the port of the slave device.

Each of the second cluster detection fields includes a first indicator bit, a second indicator bit, and a third indicator bit, and the first indicator bit in the second cluster detection field is used to indicate the master device that sends the cluster response message. The device role before the protocol channel failure, the second indication bit in the second cluster detection field indicates the device role of the master device that sends the cluster response message after the protocol channel failure, the second The third indicator bit in the cluster detection field indicates that the slave device is permitted to close the slave device's port.

In some possible designs, when the cluster includes more than two slave devices and the protocol channel between the master device and at least one slave device fails, the method further includes:

Among the two or more slave devices, the slave device whose protocol channel is faulty with the master device receives the cluster detection packet sent by the master device, and returns a cluster response packet to the master device, so that all The master device closes the port of the master device.

For example, when the protocol channel between the master device and at least one slave device fails, if the current number of master devices is less than the number of slave devices in the cluster system, the original master device can send a cluster detection packet to each slave device. The slave device converted to the master device in the middle, the original master device is down. After receiving the cluster detection message sent by the master device, each slave device that is converted into the master device will return a cluster response message to the master device, so that the original master device thinks that the cluster has been dissolved and shuts down the master device. of each port.

If the number of master devices is equal to the number of slave devices in the cluster system, and the protocol channel between the master device and the slave device fails, in general, the slave device sends the cluster detection message first, and the original master device returns A cluster response message, the slave device closes its own port after receiving the cluster response message returned by the master device. In other embodiments, the master device can also send a cluster detection packet to the slave device, and then close its own port after receiving the cluster response packet returned by the slave device. The present invention is not limited, as long as repeated BGP oscillation can be avoided. It is enough to ensure the normal forwarding of traffic between nodes.

In some possible designs, when the protocol channel between the slave device and the master device recovers from a fault, the current device role is restored from the slave device in the master role to the slave role;

The slave device opens the port of the slave device, and the slave device can not only automatically release the down state of each port, but also manually release the down state of each port through a command line.

In some possible designs, after the slave device closes the port of the slave device, the method further includes:

The slave device configures the port state of each closed port of the slave device to the down state, so that after the protocol channel between the slave device and the master device recovers from a fault, the current device role is the same as the master device role. The slave device is restored to the role of the slave device, and then the slave device opens the port of the slave device. In the present invention, the slave device can not only automatically release the down state of each port, but also manually release the down state of each port through a command line.

A second aspect of the present invention provides a cluster device, which has the function of implementing the cluster management method corresponding to the above-mentioned first aspect. The functions can be implemented by hardware, or can be implemented by hardware executing corresponding software. The hardware or software includes one or more modules corresponding to the above functions, and the modules may be software and/or hardware.

In a possible design, a cluster device is applied to a cluster, the cluster includes a master device and a slave device, and the cluster device includes:

a processing module, configured to convert the slave device into the role of the master device when the protocol channel between the master device and the slave device fails;

a transceiver module for sending the cluster detection message to the master device through a data channel;

The processing module is further configured to, after receiving the cluster response message sent by the master device through the transceiver module, exit the cluster according to the cluster response message; and close the port of the slave device.

Optionally, the cluster detection packet includes a first cluster detection field, and the cluster response packet includes a second cluster detection field.

In some possible designs, the first cluster detection field includes a first indicator bit, a second indicator bit, and a third indicator bit, and the first indicator bit in the first cluster detection field is used to indicate that the cluster is to be sent The device role of the slave device that detects the detection packet before the protocol channel fails, and the second indication bit in the first cluster detection field indicates that the slave device that sent the cluster detection packet is after the protocol channel fails. the device role, the third indication bit in the first cluster detection field indicates that the slave device applies for closing the port of the slave device.

In some possible designs, the second cluster detection field includes a first indicator bit, a second indicator bit and a third indicator bit, and the first indicator bit in the second cluster detection field is used to indicate that the cluster is to be sent The device role of the master device of the response packet before the protocol channel failure, the second indication bit in the second cluster detection field indicates the device of the master device sending the cluster response packet after the protocol channel failure role, the third indication bit in the second cluster detection field indicates that the slave device is allowed to close the port of the slave device.

In some possible designs, when the cluster includes more than two slave devices, the transceiver module is also used for:

Receive a cluster detection packet sent by the master device, and return a cluster response packet to the master device, so that the master device closes the port of the master device.

In some possible designs, the processing module is also used to:

After the fault of the protocol channel between the slave device and the master device is restored, restore the current master device role of the slave device to the slave device role;

Open the port of the slave device.

In some possible designs, after the slave device closes the port of the slave device, the processing module is further configured to:

The port state of each closed port of the slave device is configured to be in the down state.

In a possible design, a cluster device is applied to a cluster, the cluster includes a master device and a slave device, and the cluster device includes:

at least one processor, memory and transceiver;

Wherein, the memory is used to store program code, and the processor is used to call the program code in the memory to perform the following operations:

When the protocol channel between the master device and the slave device fails, the slave device is converted into the role of the master device;

Send the cluster detection message to the master device through the data channel through the transceiver;

After receiving the cluster response message sent by the master device through the transceiver, exit the cluster according to the cluster response message; and close the port of the slave device.

Compared with the prior art, in the solution provided by the present invention, after the failure of the protocol channel, the slave device that is transformed into the role of the master device sends the cluster detection message to the master device through the data channel, and then sends the cluster detection message to the master device after receiving it. After receiving the cluster reply message, the port of the slave device is forcibly closed, thereby isolating the original slave device. No processing is performed on the original master device, and the protocol between the original master device and the upstream and downstream devices of the original master device can still be established, so that the original service data can continue to be forwarded, so that the traffic forwarding of the original master device will not be affected. The BGP protocol of the device fluctuates repeatedly. It can be seen that this ensures that the node has a device available and has minimal impact on the current network.

Description of drawings

1 is a schematic diagram of a network topology structure of a cluster in this embodiment;

FIG. 2 is a schematic flowchart of a cluster management method in this embodiment;

FIG. 3 is another schematic flowchart of the cluster management method in this embodiment;

FIG. 4 is a schematic structural diagram of a cluster device in this embodiment;

FIG. 5 is a schematic structural diagram of an entity device for executing the above cluster management method in this embodiment.

Detailed ways

The terms "first", "second" and the like in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It is to be understood that data so used may be interchanged under appropriate circumstances so that the embodiments described herein can be practiced in sequences other than those illustrated or described herein. Furthermore, the terms "comprising" and "having", and any variations thereof, are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device comprising a series of steps or modules is not necessarily limited to those expressly listed Those steps or modules, but may include other steps or modules not explicitly listed or inherent to these processes, methods, products or devices, the division of modules presented herein is only a logical division, In practical applications, there may be other divisions in implementation, for example, multiple modules may be combined or integrated in another system, or some features may be ignored, or not implemented, in addition, the coupling shown or discussed with each other Or direct coupling or communication connection may be through some interfaces, and indirect coupling or communication connection between modules may be electrical or other similar forms, which are not limited herein. In addition, the modules or sub-modules described as separate components may or may not be physically separated, may or may not be physical modules, or may be distributed into multiple circuit modules, and some or all of them may be selected according to actual needs. modules to achieve the purpose of the solutions of the embodiments of the present invention.

Embodiments of the present invention provide a cluster management method and device, which are mainly used in the field of virtual cluster technology, for example, in a cluster system, which can avoid repeated oscillation of the BGP protocol of a master device and a slave device, and ensure normal forwarding of node traffic. A detailed description will be given below.

The cluster in the present invention refers to a group of mutually independent computers interconnected through a network through a protocol, which constitute a network and are managed in a single system mode. Each device in the cluster can share resources, share costs, share channel devices and services, etc. It can be a virtual machine cluster. A virtual cluster is a single logical core router composed of multiple single chassis. The cluster may include one master device and at least one slave device, and both the master device and the slave device may be arranged in the chassis. The master device is used to control the cluster in which it resides. A slave device is used to collect data. Both the master device and the slave device can perform operations such as traffic forwarding between the devices connected to the upstream and downstream communication. As shown in Figure 1, Figure 1 is a schematic diagram of a network topology of the cluster. The master device (device A) and the slave device (device B) are deployed in the same network through protocols, and their physical locations are not limited. In Figure 1, device C and device D are uplink and downlink devices of the cluster, that is, BGP neighbors.

A protocol channel and a data channel exist between the master device and the slave device. The protocol channel refers to the communication link between the cluster chassis through the interface of the main control board. The protocol channel can be used to transmit protocol packets in the cluster. The data channel refers to the link between the cluster chassis, which is connected through the interface of the service control board. The data channel can be used to transmit data packets between the virtual cluster chassis.

After the protocol channel fails, the slave device may switch to the role of the master device, resulting in the phenomenon of more than two master devices in a cluster. At the same time, the device with the protocol channel failure will exit the cluster where it was before the failure. If there are more than two master devices, the BGP of the BGP neighbor device of the master device will be repeatedly flapped, and the traffic will be interrupted repeatedly, and the normal traffic forwarding cannot be guaranteed. In order to solve the above-mentioned technical problems, the embodiments of the present invention mainly provide the following technical solutions:

After the slave device is converted into the role of the master device, the slave device sends the cluster detection packet to the original master device through the data channel, and then forcibly closes the port of the slave device after receiving the cluster response packet sent by the original master device, thereby isolating the slave device. .

Through the above technical solutions, the problem of repeated oscillation of the BGP protocol when there are more than two master devices can be avoided, and the normal forwarding of traffic of each node can also be ensured.

Referring to FIG. 2 , a cluster management method provided by the present invention is exemplified below. The method is applied to a cluster, and the cluster includes a master device and a slave device. A protocol channel and a data channel exist between the master device and the slave device. In a cluster, the number of slave devices can include more than one. Under normal conditions of the protocol channel, there is only one master device in a cluster. The method includes:

101. When the protocol channel between the master device and the slave device fails, the slave device switches to the role of the master device.

102. The slave device sends a cluster detection packet to the master device through a data channel.

Wherein, the cluster detection packet includes a first cluster detection field. The first cluster detection field includes a first indicator bit, a second indicator bit and a third indicator bit, and the first indicator bit in the first cluster detection field is used to indicate a slave device that sends the cluster detection detection packet The device role before the protocol channel failure, the second indicator bit in the first cluster detection field indicates the device role of the slave device that sends the cluster detection detection packet after the protocol channel failure, the first The third indicator bit in a cluster detection field indicates that the slave device applies to close the port of the slave device.

In some application scenarios, the above-mentioned cluster detection packets can adopt the unified link layer management (English full name: Consolidated LinkLayer Management, English abbreviation: CLM)/fiber link layer management (English full name: FiberLinkLayer Management, English abbreviation: FLM) chain The expansion of the road layer protocol adds the first cluster detection field. The first cluster detection field occupies 2 bytes, and the definition format is as follows: the first 1-2 bits are the first indicator bits, which are used to indicate the cluster device before the protocol channel failure. Device role, where 01 represents the slave role (slave), and 10 represents the master device role (master). Bits 3-4 are the second indication bits, which are used to represent the current device role of the cluster device, wherein 01 represents the slave device role (slave), and 10 represents the master device role (master). Bits 5-6 indicate whether to apply for down, of which 01 indicates application for down. For example, in a cluster detection packet, the first cluster detection field may be displayed as 01100100 00000000.

103. After receiving the cluster detection packet, the master device sends a cluster response packet to the slave device.

Wherein, the cluster response message includes a second cluster detection field. Each of the second cluster detection fields includes a first indicator bit, a second indicator bit, and a third indicator bit, and the first indicator bit in the second cluster detection field is used to indicate the master device that sends the cluster response message. The device role before the protocol channel failure, the second indication bit in the second cluster detection field indicates the device role of the master device that sends the cluster response message after the protocol channel failure, the second The third indicator bit in the cluster detection field indicates that the slave device is permitted to close the slave device's port.

Correspondingly, the above cluster response message can also be extended by the CLM/FLM link layer protocol, adding a second cluster detection field. The second cluster detection field occupies 2 bytes, and the definition format is as follows: The first 1-2 bits represent the first An indication bit is used to indicate the device role of the cluster device before the protocol channel failure, wherein 01 represents the slave device role (slave), and 10 represents the master device role (master). Bits 3-4 represent the second indication bits, which are used to represent the current device role of the cluster device, wherein 01 represents the slave device role (slave), and 10 represents the master device role (master). 5-6 bits indicate whether to apply for down, and 10 indicates agree to down. For example, in the cluster reply message, the second cluster detection field may be displayed as 10101100 00000000.

For the above-mentioned cluster detection message and the above-mentioned cluster response message, the communication format of the two can be one of the following Table 1, or can be obtained by deformation on the basis of Table 1, which is not limited in the present invention.

Table 1

For the specific meaning of each element in Table 1 above, please refer to Table 2 below.

Table 2

In Table 2, CMD refers to cluster multi-frame detection (full name in English: cluster-chassis multi-user detection, English abbreviation: CMD), which can be used to detect and de-cluster the protocol, CMD is a kind of cluster detection message or cluster response message The specific embodiment may also have other names, which are not specifically limited by the present invention.

104. After receiving the cluster response message sent by the master device, the slave device exits the cluster according to the cluster response message.

105. The slave device closes the port of the slave device.

The closed ports include ge0/0/0, a loopback interface (loopback), a virtual local area network interface (full English name: virtual local area network interface, English abbreviation: vlanif), service ports, and the like. After closing each port of the slave device, the slave device may also configure the port state of each closed port of the slave device to the down state, so that after the protocol channel between the slave device and the master device recovers from failure , the current device role is restored from the slave device in the master device role to the slave device role, and then the slave device opens the port of the slave device. In the present invention, the slave device can not only automatically release the down state of each port, but also manually release the down state of each port through a command line.

Compared with the existing mechanism, in this embodiment of the present invention, after the protocol channel fails, the slave device that is transformed into the role of the master device sends the cluster detection packet to the master device through the data channel, and then receives the message sent by the master device. After the cluster responds to the message, the port of the slave device is forcibly closed, thereby isolating the original slave device. No processing is performed on the original master device, and the protocol between the original master device and the upstream and downstream devices of the original master device can still be established, so that the original service data can continue to be forwarded, so that the traffic forwarding of the original master device will not be affected. The BGP of the device flaps repeatedly.

In other words, when more than two master devices are detected, the original and slave devices can be isolated by forcibly closing the ports of the original and slave devices. In the above steps 104 and 105, the original master device does not perform any processing, and the original master device can continue to forward the original service data, so that the traffic forwarding of the original master device will not be affected. It can be seen that this ensures that the node has a device available and has minimal impact on the current network. At the same time, by linking the BFD feature, the impact on the network caused by the presence of more than two master devices can be further reduced to the millisecond level.

Optionally, in some inventive embodiments, when the cluster includes more than two slave devices, the protocol channel between the master device and at least one slave device is faulty. Since the number of original slave devices is more than the original master device, in this case, the following rules can also be set:

The master device may send a cluster detection message to each slave device in the cluster (including the slave device that has been transformed into the role of the master device, and/or the slave device that is still in the role of the slave device). Each slave device receives the cluster detection message sent by the master device, and returns a cluster response message to the master device, so that the original master device thinks that the cluster has been disbanded, and closes each port of the master device (the same as the previous part. the port of the slave device is the same or similar).

It should be noted that when the number of master devices and slave devices in the cluster is the same, if the protocol channel between the master device and the slave device fails, the slave device will send cluster detection packets to the master device first, and then , the slave device closes its own port after receiving the cluster response packet returned by the master device. In other embodiments, the master device can also send a cluster detection packet to the slave device, and then close its own port after receiving the cluster response packet returned by the slave device. The present invention is not limited, as long as repeated BGP oscillation can be avoided. It is enough to ensure the normal forwarding of traffic between nodes.

For ease of understanding, a specific application scenario is given below to illustrate the cluster management method of the present invention: As shown in FIG. 3 , the cluster includes a master frame (Master) and a slave frame (Slave), and a Protocol channels (connected via ports #17 and #18) and data channels (connected via ports #11 and #12, for example). After the protocol channel fails, the slave sends a cluster detection packet to the master through the data channel, and then the master returns a cluster response packet to the slave. After the Slave receives the cluster response message and thinks that the cluster has been dissolved, it will close its port and exit the cluster.

A cluster management method in the present invention is described above, and a cluster device that executes the above cluster management method is described below.

1. Referring to FIG. 4 , the cluster device 40 is described. The cluster device is applied to a cluster, and the cluster includes a master device and a slave device. The cluster device 40 includes:

A processing module 401, configured to convert the slave device into the role of the master device when the protocol channel between the master device and the slave device fails;

a transceiver module 402, configured to send a cluster detection message to the master device through a data channel;

The processing module 401 is further configured to, after receiving the cluster response message sent by the master device through the transceiver module 402, exit the cluster according to the cluster response message; and close the port of the slave device.

In this embodiment of the present invention, after the protocol channel fails, the processing module 401 transforms the slave device into the role of the master device, and the transceiver module 402 sends the cluster detection message to the master device through the data channel, and then receives the message sent by the master device. After the cluster reply message, the processing module 401 forcibly closes the port of the slave device, thereby isolating the original slave device. Without performing any processing on the original master device, the original master device can continue to forward the original service data, so that the traffic forwarding of the original master device will not be affected, and the phenomenon of repeated BGP oscillation between the master device and the slave device will be avoided. It can be seen that this ensures that the node has a device available and has minimal impact on the current network.

Optionally, the cluster detection packet includes a first cluster detection field, and the cluster response packet includes a second cluster detection field.

Optionally, in some embodiments of the invention, the first cluster detection field includes a first indicator bit, a second indicator bit, and a third indicator bit, and the first indicator bit in the first cluster detection field is used to indicate The device role of the slave device that sends the cluster detection detection packet before the protocol channel fails, and the second indication bit in the first cluster detection field indicates that the slave device that sent the cluster detection detection packet is in the The device role after a protocol channel failure, the third indication bit in the first cluster detection field indicates that the slave device applies for closing the port of the slave device.

Optionally, in some embodiments of the invention, the second cluster detection field includes a first indicator bit, a second indicator bit, and a third indicator bit, and the first indicator bit in the second cluster detection field is used to indicate the device role of the master device sending the cluster response message before the protocol channel fails, and the second indication bit in the second cluster detection field indicates that the master device sending the cluster response message is in the protocol channel The role of the device after the failure, the third indication bit in the second cluster detection field indicates that the slave device is allowed to close the port of the slave device.

Optionally, in some inventive embodiments, when the cluster includes more than two slave devices, the transceiver module 402 is further configured to:

Receive a cluster detection packet sent by the master device, and return a cluster response packet to the master device, so that the master device closes the port of the master device.

Optionally, in some inventive embodiments, the processing module 401 is further configured to:

After the fault of the protocol channel between the slave device and the master device is restored, restore the current master device role of the slave device to the slave device role;

Open the port of the slave device.

Optionally, in some embodiments of the invention, after the slave device closes the port of the slave device, the processing module 401 is further configured to:

The port state of each closed port of the slave device is configured to be in the down state.

It should be noted that, in the embodiment corresponding to FIG. 4 of the present invention, the entity device corresponding to the transceiver module may be a transceiver, and the entity device corresponding to the processing module may be a processor. The apparatus shown in FIG. 4 may have the structure shown in FIG. 5 , when one of the apparatuses has the structure shown in FIG. 5 , the processor and the transceiver in FIG. 5 implement the foregoing apparatus embodiments corresponding to the apparatus provided The processing module and the transceiver module have the same or similar functions, and the memory in FIG. 5 stores the program code that needs to be called when the processor executes the above cluster management method.

In the above-mentioned embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference may be made to the relevant descriptions of other embodiments.

Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the system, device and module described above can refer to the corresponding process in the foregoing method embodiments, which is not repeated here.

In the several embodiments provided in this application, it should be understood that the disclosed system, apparatus and method may be implemented in other manners. For example, the apparatus embodiments described above are only illustrative. For example, the division of the modules is only a logical function division. In actual implementation, there may be other division methods. For example, multiple modules or components may be combined or Can be integrated into another system, or some features can be ignored, or not implemented. On the other hand, the shown or discussed mutual coupling or direct coupling or communication connection may be through some interfaces, indirect coupling or communication connection of devices or modules, and may be in electrical, mechanical or other forms.

The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, that is, may be located in one place, or may be distributed to multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution in this embodiment.

In addition, each functional module in each embodiment of the present invention may be integrated into one processing module, or each module may exist physically alone, or two or more modules may be integrated into one module. The above-mentioned integrated modules can be implemented in the form of hardware, and can also be implemented in the form of software function modules.

If the integrated modules are implemented in the form of software functional modules and sold or used as independent products, they may be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium , including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage media include: U disk, mobile hard disk, read-only memory (full English name: Read-Only Memory, English abbreviation: ROM), random access memory (English full name: Random Access Memory, English abbreviation: RAM), magnetic disk Or various media such as optical discs that can store program codes.

The technical solutions provided by the present invention are described in detail above, and specific examples are used in this paper to illustrate the principles and implementations of the present invention. The descriptions of the above embodiments are only used to help understand the method of the present invention and its core idea; Meanwhile, for those of ordinary skill in the art, according to the idea of the present invention, there will be changes in the specific embodiments and application scope. In summary, the contents of this specification should not be construed as limiting the present invention.

Claims (6)

1. A cluster management method is applied to a cluster, wherein the cluster comprises a master device and a slave device, and the method comprises the following steps:

when a protocol channel between the master device and the slave device fails, the slave device is switched to a master device role;

the slave device sends the cluster detection message to the master device through a data channel;

after receiving a cluster response message sent by the master device, the slave device exits the cluster according to the cluster response message;

the slave device closes a port of the slave device;

the cluster detection message comprises a first cluster detection field, and the cluster response message comprises a second cluster detection field;

the first cluster detection field includes a first indication bit, a second indication bit and a third indication bit, the first indication bit in the first cluster detection field is used to indicate a device role of a slave device sending the cluster detection packet before the protocol channel failure, the second indication bit in the first cluster detection field indicates a device role of the slave device sending the cluster detection packet after the protocol channel failure, and the third indication bit in the first cluster detection field indicates that the slave device applies for closing a port of the slave device;

the second cluster detection field includes a first indication bit, a second indication bit and a third indication bit, the first indication bit in the second cluster detection field is used to indicate a device role of the master device sending the cluster response packet before the protocol channel failure, the second indication bit in the second cluster detection field indicates a device role of the master device sending the cluster response packet after the protocol channel failure, and the third indication bit in the second cluster detection field indicates that the slave device agrees to close a port of the slave device;

when more than two slave devices are included in the cluster, the method further comprises:

and the slave equipment with the fault protocol channel between the two or more slave equipment and the master equipment receives the cluster detection message sent by the master equipment and returns a cluster response message to the master equipment, so that the master equipment closes the port of the master equipment.

2. The method of claim 1, further comprising:

when the protocol channel between the slave device and the master device is recovered from the fault, the slave device with the current device role as the master device role is recovered to the slave device role;

the slave device opens a port of the slave device.

3. The method of claim 1, wherein after the slave device closes the port of the slave device, the method further comprises:

the slave device configures a port state of each closed port of the slave device to a down state.

4. A cluster device, wherein the cluster device is applied to a cluster, wherein the cluster comprises a master device and a slave device, and wherein the cluster device comprises:

the processing module is used for converting the slave equipment into a master equipment role after a protocol channel between the master equipment and the slave equipment fails;

the receiving and sending module is used for sending the cluster detection message to the main equipment through a data channel;

the processing module is further configured to quit the cluster according to the cluster response message after receiving the cluster response message sent by the master device through the transceiver module; and closing the port of the slave device;

the cluster detection message comprises a first cluster detection field, and the cluster response message comprises a second cluster detection field;

the first cluster detection field includes a first indication bit, a second indication bit and a third indication bit, the first indication bit in the first cluster detection field is used to indicate a device role of a slave device sending the cluster detection packet before the protocol channel failure, the second indication bit in the first cluster detection field indicates a device role of the slave device sending the cluster detection packet after the protocol channel failure, and the third indication bit in the first cluster detection field indicates that the slave device applies for closing a port of the slave device;

the second cluster detection field includes a first indication bit, a second indication bit and a third indication bit, the first indication bit in the second cluster detection field is used to indicate a device role of the master device sending the cluster response packet before the protocol channel failure, the second indication bit in the second cluster detection field indicates a device role of the master device sending the cluster response packet after the protocol channel failure, and the third indication bit in the second cluster detection field indicates that the slave device agrees to close a port of the slave device;

when more than two slave devices are included in the cluster, the transceiver module is further configured to:

and receiving a cluster detection message sent by the main equipment, and returning a cluster response message to the main equipment so that the main equipment closes the port of the main equipment.

5. The cluster device of claim 4, wherein the processing module is further configured to:

when the protocol channel between the slave equipment and the master equipment is recovered from the fault, recovering the current master equipment role of the slave equipment to the slave equipment role;

opening a port of the slave device.

6. The cluster device of claim 4, wherein the processing module, after the slave device closes the port of the slave device, is further configured to:

configuring a port state of each closed port of the slave device to a down state.

Images