CN109005111B

CN109005111B - IRF splitting and merging method and device

Info

Publication number: CN109005111B
Application number: CN201810973328.XA
Authority: CN
Inventors: 王阳; 廖以顺; 顾彩云
Original assignee: Hangzhou H3C Technologies Co Ltd
Current assignee: Hangzhou H3C Technologies Co Ltd
Priority date: 2018-08-24
Filing date: 2018-08-24
Publication date: 2021-07-23
Anticipated expiration: 2038-08-24
Also published as: CN109005111A

Abstract

The present disclosure provides a method and an apparatus for merging after IRF splitting, which can determine whether a second IRF group is a first IRF group in which the routing device is located before splitting when the routing device applies for adding the second IRF group; and if so, updating the state identifier of the routing equipment into a slave election identifier, wherein the slave election identifier is used for participating in the election of the slave equipment in the second IRF group. By applying the method, the split equipment does not need to be restarted when being added into the IRF group again after the fault is recovered, and can enter a stable state quickly after election and election, so that the network fault recovery period is shortened.

Description

IRF splitting and merging method and device

Technical Field

The present disclosure relates to the field of communications technologies, and in particular, to a method and an apparatus for combining after IRF splitting.

Background

An IRF (Intelligent Resilient Framework) virtualization stacking model can virtualize a plurality of devices into one device for management, so that a user can manage the devices logically by managing one device, thereby greatly reducing the maintenance workload of the user and simplifying the network deployment.

A plurality of devices in the IRF group can select one master device in an election mode, and the other devices are slave devices. When a new device joins the existing IRF group, re-election is triggered, but because the device in the existing IRF group is already a master device and is in operation, the newly joined device inevitably fails in election, and then joins the IRF group as a slave device after restart.

Disclosure of Invention

In view of this, the present disclosure provides a method and an apparatus for combining after IRF splitting to solve the problem in the prior art that the device is restarted due to failed election when the originally split device is re-added to the IRF group.

Specifically, the present disclosure is realized by the following technical solutions:

the present disclosure provides a method for combining after IRF splitting, which is applied to a routing device in an IRF network, and the method includes:

if the routing equipment applies for adding a second IRF group, judging whether the second IRF group is a first IRF group in which the routing equipment is positioned before splitting;

and if so, updating the state identifier of the routing equipment into a slave election identifier, wherein the slave election identifier is used for participating in the election of the slave equipment in the second IRF group.

Based on the same concept, the present disclosure provides an apparatus for combining after IRF splitting, which is applied to a routing device in an IRF network, and the apparatus includes:

a merging judgment unit, configured to judge whether a second IRF group is a first IRF group in which the routing device is located before splitting, if the routing device applies for joining the second IRF group;

a first changing unit, configured to update the status identifier of the routing device to a slave election identifier if the second IRF group is the first IRF group in which the routing device is located before splitting, where the slave election identifier is used to participate in the slave election of the second IRF group.

Based on the same concept, the present disclosure further provides a computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, and when the computer program is executed by a processor, the computer program implements any step of the above IRF split and merge method.

Therefore, the method and the device can judge whether the second IRF group is the first IRF group in which the routing device is located before splitting when the routing device applies for joining the second IRF group; and if so, updating the state identifier of the routing equipment into a slave election identifier, wherein the slave election identifier is used for participating in the election of the slave equipment in the second IRF group. By applying the method, the split equipment does not need to be restarted when being added into the IRF group again after the fault is recovered, and can enter a stable state quickly after election and election, so that the network fault recovery period is shortened.

Drawings

Fig. 1 is a schematic diagram of IRF networking in an exemplary embodiment of the present disclosure;

FIG. 2 is a process flow diagram of a method of IRF post-split consolidation in an exemplary embodiment of the present disclosure;

FIG. 3 is a schematic illustration of an election process in an exemplary embodiment of the present disclosure;

FIG. 4a is a hardware structure diagram of a routing device where an IRF split and merge device in an exemplary embodiment of the present disclosure is located;

fig. 4b is a logical block diagram of an apparatus for IRF split and merge in an exemplary embodiment of the present disclosure.

Detailed Description

Please refer to fig. 1, which is a schematic diagram of an IRF stacking networking, wherein when a device 1, a device 2, a device 3, and a device 4 are stacked in an IRF, one master device may be selected by election, and the others are slave devices. When a link between the device 2 and the device 3 fails, stack splitting is caused, the device 1 and the device 2 are split into an IRF group 1, the device 3 and the device 4 are an IRF group 2, and one master device is reselected from each IRF group, wherein the master device of the IRF group 1 is the device 1, and the master device of the IRF group 2 is the device 3. In order to prevent the device address conflict between the devices operating in the IRF group 1 and IRF group 2 networks, after detecting the split, the MAD (Multi-Active Detection) will close the group of IRFs according to a certain rule, and assume that the MAD closes the interfaces of the devices 1 and 2 in the IRF group 1, indicating that the devices 1 and 2 do not continue to operate, so as to prevent the device address conflict with the device addresses after the stack of the devices 3 and 4 is split.

After the IRF link detection is unobstructed, the device 1 and the device 2 in the IRF group 1 want to join the IRF group 2, and re-join the election, because the combined device 1 and the device 2 are devices that are originally stacked and split away, and when re-join the election, because the device 3 in the existing IRF group 2 is already a master device and is in operation, the device 1 and the device 2 will inevitably fail in the election, thereby causing the device 1 and the device 2 to join the IRF group 2 as a slave device after being restarted. But since the devices 1 and 2 do not have faults and do not need to be restarted, the election failure causes the devices 1 and 2 to be restarted, which can cause the time period for the network to reach the steady state to be prolonged, and therefore, the recovery time of the network is also longer. A reboot must therefore result whenever there is a newly added device. Therefore, even if the device is split out originally due to the network reason, when the device rejoins the IRF group, the device is restarted due to election failure, and the network recovery period is prolonged.

In order to solve the problems in the prior art, the present disclosure may enable the routing device to determine whether the second IRF group is the first IRF group in which the routing device is located before splitting when applying for adding the second IRF group; and if so, updating the state identifier of the routing equipment into a slave election identifier, wherein the slave election identifier is used for participating in the election of the slave equipment in the second IRF group. By applying the method, the split equipment does not need to be restarted when being rejoined into the IRF group after the fault is recovered, and can quickly enter a stable state after election and election, so that the current state identifier of the equipment is changed into a temporary identifier when the equipment determines to be split from the first IRF group in the network fault recovery period is shortened; when the equipment applies to join a second IRF group, judging whether the second IRF group is a first IRF group in which the equipment is located before splitting; if so, changing the temporary identifier of the equipment into a slave election identifier, wherein the slave election identifier is used for participating in the slave equipment election in the second IRF group; if not, restarting the equipment and initializing the state identifier of the equipment. By applying the method, the split equipment does not need to be restarted when being added into the IRF group again after the fault is recovered, and can enter a stable state quickly after election and election, so that the network fault recovery period is shortened.

Referring to fig. 2, it is a flowchart of a method for combining after IRF splitting in an exemplary embodiment of the present disclosure, where the method is applied to a routing device in an IRF network, and the method includes:

step 201, if the routing device applies for joining a second IRF group, determining whether the second IRF group is a first IRF group in which the routing device is located before splitting;

as an embodiment, the routing device may be split from the first IRF group for a plurality of reasons, for example, the reason may be that the routing device fails, or a link between the routing device and a connected device fails, or the like. When the above problem occurs, the routing device may receive the MAD message, and the MAD message may be used to notify the IRF of the split detection, the collision handling, and the failure recovery, thereby improving the availability of the system. When the routing device receives the MAD message, and determines that the routing device is interrupted in communication with other devices in the first IRF group according to the content of the MAD message, the routing device may determine that the routing device is split from the first IRF group, and close the routing device according to actual conditions to perform fault repair.

As an embodiment, when the routing device is split from the first IRF group where the routing device is located because of a self failure or a link failure, the routing device may be subjected to failure repair, and during this period, the state of the routing device itself may be unstable, so that in order not to affect normal service forwarding, the current state identifier of the routing device may be changed to a temporary identifier, specifically, when the current state identifier of the routing device is a main device identifier, the main device identifier is changed to a temporary identifier; and when the current state identifier of the routing equipment is the slave equipment identifier, changing the slave equipment identifier into a temporary identifier. After the temporary identifier is changed, the routing equipment temporarily does not participate in the election.

It should be noted that the status identifier of the routing device is used to indicate the status of the routing device, and includes at least a master device identifier, a slave device identifier, a master election identifier, a slave election identifier, a loading identifier, and a temporary identifier, where the master device identifier and the slave device identifier are both stable statuses of the routing device, and the other status identifiers are all unstable statuses. The routing device may correspond to different election rules through different state identifications.

In this embodiment, if the present routing device currently applies to join the second IRF group, it may be further determined whether the second IRF group is the same as the first IRF group in which the present routing device was located before splitting. Specifically, the device may obtain topology information of member devices of a second IRF group according to interaction with the member devices of the second IRF group; comparing the topology information of the device before splitting reserved by the device with the acquired topology information of the member devices, and if the topology information of the device before splitting is the same as the acquired topology information of the member devices, determining that the second IRF group is the first IRF group in which the device before splitting is located; and if not, determining that the second IRF group is not the first IRF group where the device is located before the splitting.

It should be noted that, when determining whether the second IRF group is the first IRF group in which the routing device is located before splitting, a single feature, such as a device MAC, may be used, but in a complex topology structure, the single feature may not completely and accurately identify one routing device, so the topology information compared in the present disclosure mainly includes: when the topology information comparison is completely the same, the routing device can be considered to be split from the IRF group which is currently requested to join. The compared topology information is relatively comprehensive, so that the topology information comparison method disclosed by the invention can be suitable for any topology structure, and the comparison accuracy can be ensured. When the topology is simple, such as the topology shown in fig. 1, it may be determined whether the routing device is a routing device split from the IRF group only by comparing one or more pieces of the topology information (e.g., comparing only the device MAC).

Step 202, if the second IRF group is the first IRF group where the routing device is located before splitting, updating the state identifier of the routing device to a slave election identifier, where the slave election identifier is used for participating in the election of the slave device in the second IRF group.

When the routing device determines that the current second IRF group is the first IRF group before splitting, it indicates that the routing device is split from the second IRF group, so that the routing device can be directly merged into the second IRF group without restarting the routing device, and the state identifier of the routing device is updated to the slave election identifier, so that the routing device can use the slave device election in the second IRF group by using the slave election identifier, and can become the slave device if the election is completed, so that the state of the routing device is stable.

When the routing device determines that the current second IRF group is not the first IRF group before splitting, it indicates that the routing device is not split from the second IRF group, so that the routing device can be restarted, initialize the state identifier of the routing device, and participate in the election again.

Compared with the situation that equipment needs to be restarted when the equipment is merged into an IRF group in the prior art, the method and the device can compare the topology information before the splitting with the topology information of the member equipment of the current IRF group, and if the topology information is the same, the routing equipment is considered to be the equipment of the IRF group, so that the restarting of the routing equipment can not be triggered, and the routing equipment enters a stable state through election and election by changing the current state identification into the election identification, thereby shortening the network fault recovery period.

In order to make the objects, technical solutions and advantages of the present disclosure more apparent, the following embodiments are further described in detail.

Please refer to fig. 3, which is a schematic diagram of an election process in an embodiment of the present disclosure, wherein each device individually executes the election process, and the process specifically includes:

step 301, initializing an IRF enabling device, and calculating topology information of the routing device;

step 302, if an interface of the routing device in the initialization state enables an IRF virtualization function, the routing device enters a competition state from the initialization state;

step 303, entering a master device state from an initialization state if no interface enables an IRF virtualization function in the initialization state;

step 304, if the election timer is overtime and a device with higher priority exists in the competition state, entering a slave competition state;

step 305, entering a main election state if the election timer is overtime and no higher priority device exists in the competition state;

step 306, if the slave election state receives the message expected to become the EXPM Down (expected to be called as the closing of the master device) and the priority of the slave election state is highest, the slave election state enters the master election state;

and 307, in the master competition state, if a higher-priority device is found and no other device is added into the IRF group, entering a slave competition state.

Step 308, entering a master device state if the timer is overtime or all devices have replied to join in the master election state;

step 309, in the slave device state, if the master device in the IRF group fails and the master device itself has the highest priority, entering a master race selection state;

step 310, in the slave device state, if the master device in the IRF group fails and is not the highest priority, entering a slave election state;

311, in the slave device state, if the notification message is received and is definitely added, and the version of the native software is judged to be consistent with the version of the EXPM, entering the slave device state;

step 312, after the notification message is received from the device state and explicitly added, if the version of the native software is determined to be inconsistent with the version of the EXPM, entering a loading state;

step 313, if receiving the MAD detection message and needing to close the device, the main device enters a temporary state;

step 314, if receiving the MAD detection message in the slave device state and needing to close the device, entering a temporary state;

step 315, entering a secondary election state if the link is restored and added into the original IRF in the temporary state;

and step 316, restarting the link if the link is recovered and added into the non-original IRF in the temporary state, re-entering the initialization, executing the steps and re-participating in the election.

The following illustrates a specific operation flow for steps 313 to 316.

Based on the situation of fig. 1, in the IRF networking, the device 1, the device 2, the device 3, and the device 4 select a master device as the device 1 through election, and the others are slave devices. Assuming that an IRF split is triggered by a link failure between the intermediate device 2 and the device 3, the split device 1 and the device 2 form an IRF group 1, and the device 3 and the device 4 form an IRF group 2. Meanwhile, two sides separately conduct IRF re-election again, and assume that the main device of the IRF group 1 is the device 1 and the main device of the IRF group 2 is the device 3. After the IRF group 1 and the IRF group 2 are elected, it is assumed that the MAD detects that the IRF group 1 needs to be closed based on the original rule, and the IRF group 2 continues to operate in the network, at which time the devices 1 and 2 of the IRF group 1 are closed. When receiving the MAD message, the device 1 and the device 2 may migrate their own state identifiers from the master device state and the slave device state to the temporary state, respectively, after confirming that the device is split from the original IRF group. During splitting, each device senses that the IRF topology changes, so that each device independently reserves original topology information, the topology information can be reserved by using an original topology structure, and the main fields of the original topology structure comprise member numbers, device MAC, roles, priorities, starting time, hop counts and the like. Device 1 and device 2 retain topology information prior to splitting, as shown in the following table.

Frame number	1	2	3	4
					MAC	MAC1	MAC2	MAC3	MAC4
Character	Master device	Slave device	Slave device	Slave device
					Priority level	P1	P2	P3	P4
Starting time	Time1	Time2	Time3	Time3
					Hop count	H1	H2	H3	H4

When the link failure between the device 2 and the device 3 is recovered, the IRF group 1 joins the IRF group 2 again, and the IRF election needs to be performed again, at this time, after the member device of the IRF group 1 interacts with the device of the IRF group 2, the topology information of the member devices of the IRF group 2, that is, the device 3 and the device 4, is obtained, the device 1 and the device 2 can compare the topology information retained locally with the topology information of the device 3 and the device 4, and as a result, it can be known that other information except the role is the same, so that it is described that the device 1 and the device 2 are in the same IRF as the device 3 and the device 4 before, and therefore, the device 1 and the device 2 do not need to be restarted, the status identifiers of the device 1 and the device 2 are changed from the temporary identifiers to the election identifiers, so that the device 1 and the device 2 finally enter the slave device status through the election, and thus the newly joined original device in the whole stack recovery process does not need to be restarted, therefore, the speed of network fault recovery is greatly accelerated.

If the topology information of the device 1 and the device 2 is different from the topology information of the device 3 and the device 4, it indicates that the rejoined IRF group 1 is not a member device of the original IRF, the device 1 and the device 2 are triggered to restart the devices, so as to initialize the state identifiers of the device 1 and the devices, re-enter the contention state election process, and finally enter a stable slave device state.

Through the above processing, the IRF merging speed after fault recovery is very high, the restart of newly added equipment is not required to be triggered, and the stable state is quickly entered after election and election, so that the network fault recovery time is shortened.

Based on the same concept, the present disclosure also provides an apparatus for combining after IRF splitting, which may be implemented by software, or may be implemented by hardware or a combination of hardware and software. Taking a software implementation as an example, the IRF split and merge device of the present disclosure is a logical device, and is formed by reading a corresponding computer program instruction in a memory and then running the computer program instruction by a CPU of a device in which the IRF split and merge device is located.

Referring to fig. 4a, an IRF split and merge apparatus 400 in an exemplary embodiment of the present disclosure is applied to a routing device in an IRF network, where the basic operating environment of the apparatus includes a CPU, a memory, and other hardware, and from a logic level, a logic structure of the apparatus 400 is as shown in fig. 4b, where the logic structure includes:

a merging judgment unit 401, configured to judge whether a second IRF group is a first IRF group in which the routing device is located before splitting, if the routing device applies for joining the second IRF group;

a first changing unit 402, configured to update the status identifier of the local routing device to a slave election identifier if the second IRF group is the first IRF group in which the local routing device is located before splitting, where the slave election identifier is used to participate in the slave election of the second IRF group.

As an embodiment, the merging judgment unit 401 is specifically configured to acquire, by the routing device, topology information of member devices of a second IRF group according to interaction with the member devices of the second IRF group; and comparing the topology information of the routing device before splitting reserved by the routing device with the acquired topology information of the member device, and if the topology information of the routing device before splitting is the same as the acquired topology information of the member device, determining that the second IRF group is the first IRF group in which the routing device before splitting is located.

As an embodiment, the apparatus further comprises:

a second changing unit 403, configured to update the state identifier of the routing device to a temporary identifier if the routing device determines to split from the first IRF group where the routing device is located.

As an embodiment, the second changing unit 403 is specifically configured to determine that the routing device is split from the first IRF group if it is determined that the routing device is interrupted in communication with other devices in the first IRF group according to the received multiple activity detection MAD message.

As an embodiment, the apparatus further comprises:

a device restarting unit 404, configured to restart the routing device and initialize the state identifier of the routing device if the second IRF group is not the first IRF group in which the routing device is located before splitting.

The present disclosure further provides a computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, and when the computer program is executed by a processor, the computer program implements any step of the above method for combining after IRF splitting provided by the embodiment of the present disclosure.

In summary, the present disclosure may enable the routing device to determine whether the second IRF group is the first IRF group where the routing device is located before splitting when applying for joining the second IRF group; and if so, updating the state identifier of the routing equipment into a slave election identifier, wherein the slave election identifier is used for participating in the election of the slave equipment in the second IRF group. By applying the method, the split equipment does not need to be restarted when being added into the IRF group again after the fault is recovered, and can enter a stable state quickly after election and election, so that the network fault recovery period is shortened.

The implementation process of the functions and actions of each unit in the above device is specifically described in the implementation process of the corresponding step in the above method, and is not described herein again.

For the device embodiments, since they substantially correspond to the method embodiments, reference may be made to the partial description of the method embodiments for relevant points. The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the scheme of the application. One of ordinary skill in the art can understand and implement it without inventive effort.

The above description is only exemplary of the present disclosure and should not be taken as limiting the disclosure, as any modification, equivalent replacement, or improvement made within the spirit and principle of the present disclosure should be included in the scope of the present disclosure.

Claims

1. A method for merging after IRF splitting of an intelligent resilient framework is applied to a routing device in an IRF network, and comprises the following steps:

2. The method according to claim 1, wherein determining whether the second IRF group is the first IRF group in which the routing device is located before splitting includes:

the routing equipment acquires the topology information of the member equipment of a second IRF group according to the interaction with the member equipment of the second IRF group;

and comparing the topology information of the routing device before splitting reserved by the routing device with the acquired topology information of the member device, and if the topology information of the routing device before splitting is the same as the acquired topology information of the member device, determining that the second IRF group is the first IRF group in which the routing device before splitting is located.

3. The method of claim 1, further comprising:

and if the routing equipment determines to split from the first IRF group, updating the state identifier of the routing equipment to a temporary identifier.

4. The method according to claim 3, wherein determining that the routing device is split from the first IRF group specifically includes:

and if the routing equipment is determined to be interrupted in communication with other equipment in the first IRF group according to the received multi-activity detection MAD message, determining that the routing equipment is split from the first IRF group.

5. The method of claim 1, further comprising:

and if the second IRF group is not the first IRF group where the routing equipment is located before the splitting, restarting the routing equipment and initializing the state identifier of the routing equipment.

6. An apparatus for combining after IRF splitting, the apparatus being applied to a routing device in an IRF network, the apparatus comprising:

7. The apparatus of claim 6,

the merging judgment unit is specifically configured to acquire, by the routing device, topology information of member devices of a second IRF group according to interaction with the member devices of the second IRF group; and comparing the topology information of the routing device before splitting reserved by the routing device with the acquired topology information of the member device, and if the topology information of the routing device before splitting is the same as the acquired topology information of the member device, determining that the second IRF group is the first IRF group in which the routing device before splitting is located.

8. The apparatus of claim 6, further comprising:

and the second changing unit is used for updating the state identifier of the routing device to the temporary identifier if the routing device determines to be split from the first IRF group.

9. The apparatus according to claim 8, wherein determining that the routing device is split from the first IRF group specifically includes:

10. The apparatus of claim 6, further comprising:

and the device restarting unit is used for restarting the routing device and initializing the state identifier of the routing device if the second IRF group is not the first IRF group where the routing device is located before splitting.