CN106230717B - Route obtaining method and device in cluster system - Google Patents

Abstract

The application provides a method and a device for acquiring a route in a cluster system, wherein the method is applied to a designated cluster device running a link state routing protocol in the cluster system, and comprises the following steps: after establishing a neighbor relation with a specified cluster device of a directly connected opposite terminal, acquiring address information of a CPU on the cluster device in a machine frame where the device is located; the address information consists of an ID of a machine frame where the cluster equipment is located, an ID of a slot position where the cluster equipment is located and an ID of a CPU (Central processing Unit), the cluster equipment is an MPU (micro processing Unit), a CCU (Central processing Unit) or an LPU (Low Power Unit), and the designated cluster equipment running a link state routing protocol is the MPU or the CCU; sending a link state routing protocol message carrying the acquired address information; receiving a link state routing protocol message carrying address information; and calculating the route from the equipment to the CPU identified by the address information by using an SPF (SPF) protocol according to each address information carried in the received link state routing protocol message.

Description

Route obtaining method and device in cluster system

Technical Field

The present application relates to the field of network communication technologies, and in particular, to a method and an apparatus for obtaining a route in a cluster system.

Background

The cluster (cluster) technology is a newer technology, and can meet the requirements of high-speed service growth, network performance and capacity improvement, network construction cost and maintenance cost reduction and the like by using a cheaper means on the premise of convenient maintenance and no increase of network complexity, and has a wide development space. The cluster technology can enable a plurality of routing devices to form a cluster system, and enables all the routing devices in the cluster system to work together well through centralized and integrated control management, thereby greatly expanding the routing capacity. The cluster system can be regarded as a routing device, so that the network topology and the routing strategy become simple and clear, and the maintenance is more convenient and faster.

Fig. 1 is a schematic structural diagram of a cluster system, as can be seen from fig. 1, the cluster system includes: a plurality of subracks, each subrack may include therein: an MPU (Master Control Unit), a CCU (Central Control Unit), and an LPU (Line Process Unit, interface board), wherein CPUs (Central Processing units) running application programs may be disposed on the MPU, the CCU, and the LPU. Different machine frames are connected through a CCU. The subrack can be divided into a central switching box and a line card box according to different functions.

In the cluster system shown in fig. 1, the CCU1, the CCU2, the CCU3, and the CCU4 form a ring structure, and when STP (Spanning Tree Protocol) is used, some ports are blocked to avoid a loop. Thus, there is only one forwarding path between the CPU5 on the MPU5 and the CPU4 on the MPU4, and the forwarding path is, for example, MPU 5-CCU 3-CCU 4-CCU 2-MPU 4. Subsequently, when the forwarding path is interrupted, for example, the link between the CCU3 and the CCU4 is interrupted, the blocked port can be opened only after waiting for a period of time, and a new forwarding path is calculated, so that the convergence rate is slow.

Disclosure of Invention

In view of this, the present application provides a method and an apparatus for route acquisition in a cluster system.

Specifically, the method is realized through the following technical scheme:

in one aspect, a method for obtaining a route in a cluster system is provided, where the method is applied to a designated cluster device running a link state routing protocol in the cluster system, and the method includes:

after establishing a neighbor relation with a specified cluster device of a directly connected opposite terminal, acquiring address information of a CPU on the cluster device in a machine frame where the device is located; the address information consists of an ID of a machine frame where the cluster equipment is located, an ID of a slot position where the cluster equipment is located and an ID of a CPU (Central processing Unit), the cluster equipment is an MPU (micro processing Unit), a CCU (Central processing Unit) or an LPU (Low Power Unit), and the designated cluster equipment running a link state routing protocol is the MPU or the CCU;

sending a link state routing protocol message carrying the acquired address information;

receiving a link state routing protocol message carrying address information;

and calculating the route from the equipment to the CPU identified by the address information by using an SPF (SPF) protocol according to each address information carried in the received link state routing protocol message.

On the other hand, a route obtaining device in a cluster system is also provided, the device is applied to a designated cluster device running a link state routing protocol in the cluster system, and the device comprises:

the neighbor establishing module is used for establishing a neighbor relation with the directly connected opposite terminal appointed cluster equipment;

the address acquisition module is used for acquiring the address information of the CPU on the cluster equipment in the machine frame where the equipment is located after the neighbor establishment module establishes a neighbor relation with the opposite-end appointed cluster equipment, wherein the address information consists of the ID of the machine frame where the cluster equipment is located, the ID of a slot position where the cluster equipment is located and the ID of the CPU; the cluster equipment is MPU, CCU or LPU, and the designated cluster equipment operating the link state routing protocol is MPU or CCU;

a sending module, configured to send a link state routing protocol packet carrying the address information obtained by the address obtaining module;

the receiving module is used for receiving a link state routing protocol message carrying address information;

and the route calculation module is used for calculating the route from the equipment to the CPU identified by the address information by using an SPF (specific pathogen free) protocol according to each address information carried in the link state routing protocol message received by the receiving module.

By the technical scheme, a link state routing protocol is run on MPUs and CCUs (collectively referred to as designated cluster equipment) in a cluster system, the local end designated cluster equipment acquires the address information of CPUs (central processing units) on the cluster equipment (MPUs, CCUs and LPUs) in a machine frame where the local equipment is located after establishing a neighbor relation with the directly connected opposite end designated cluster equipment, and the address information is carried in a link state routing message and flooded to all other designated cluster equipment in the cluster system, wherein the address information consists of IDs (identity) of the machine frame where the cluster equipment is located, IDs of slot positions where the cluster equipment is located and IDs of the CPUs; and receiving a link state routing protocol message carrying address information sent by other specified cluster equipment, and calculating a route from the equipment to the CPU identified by the address information by using an SPF (shortest path first) protocol according to each address information carried in the received link state routing protocol message, thereby realizing mutual access between the CPUs.

Because the link state routing protocol is expanded with address information consisting of machine frame ID, slot ID and CPU ID, and the link state routing protocol is run on MPU and CCU, the appointed cluster devices running the link state routing protocol can synchronize the address information of the CPU on all cluster devices in the machine frame to other appointed cluster devices, thereby realizing the synchronization of the address information of the CPU on the cluster devices in each machine frame; then, the shortest path from the present device to each CPU can be calculated using the SPF protocol. The routes calculated by the SPF protocol do not form a loop, and when the routes are disconnected, new routes can be quickly recalculated, so that the topology convergence speed is high.

Drawings

FIG. 1 is a schematic structural diagram of a cluster system;

FIG. 2 IS a flow chart illustrating establishing IS-IS neighbor relationships in accordance with an exemplary embodiment of the present application;

FIG. 3 is a flow chart illustrating the calculation of a route according to an exemplary embodiment of the present application;

FIG. 4 is a flowchart illustrating an exemplary embodiment of the present application for obtaining address information for CPUs on all cluster devices in a subrack;

FIG. 5 is a flowchart illustrating an exemplary embodiment of the present application for obtaining address information of a CPU on a changed cluster device when the cluster device in a subrack changes;

fig. 6 is a schematic structural diagram of a route obtaining apparatus in a cluster system according to an exemplary embodiment of the present application;

fig. 7 is a schematic structural diagram of another route obtaining apparatus in a cluster system according to an exemplary embodiment of the present application;

fig. 8 is a schematic structural diagram of a route obtaining apparatus in a cluster system according to an exemplary embodiment of the present application.

Detailed Description

Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present application, as detailed in the appended claims.

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. As used in this application and the appended claims, the singular forms "a", "an", and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.

It is to be understood that although the terms first, second, third, etc. may be used herein to describe various information, such information should not be limited to these terms. These terms are only used to distinguish one type of information from another. For example, first information may also be referred to as second information, and similarly, second information may also be referred to as first information, without departing from the scope of the present application. The word "if" as used herein may be interpreted as "at … …" or "when … …" or "in response to a determination", depending on the context.

The following embodiments of the present application provide a method for obtaining a route in a cluster system, and a device for obtaining a route that can apply the method.

The embodiment of the present application is applied to the cluster system shown in fig. 1, where the cluster system includes: a plurality of subracks, each subrack may include therein: MPU, CCU and LPU, wherein, MPU, CCU and LPU can set up the CPU who runs the application program. Different machine frames are connected through a CCU. The subrack can be divided into a central switching box and a line card box according to different functions. For convenience of description, the MPU, CCU, and LPU are hereinafter collectively referred to as a cluster device.

In the embodiment of the application, in link state routing protocols such as an IS-IS (Intermediate System-to-Intermediate System) protocol and an OSPF (Open Shortest Path First) protocol, address information IS extended, and the address information IS composed of an ID of a machine frame, an ID of a slot position and an ID of a CPU.

The interface connected with the CCU on the MPU runs a link state routing protocol, and all the interfaces on the CCU run the link state routing protocol. For convenience of description, the MPU and the CCU running the link state routing protocol are collectively referred to as designated cluster devices hereinafter, and a neighbor relationship is established between two designated cluster devices directly connected in the same cluster system.

In the following embodiments of the present application, a link-state routing protocol IS described as an IS-IS protocol, and obviously, the link-state routing protocol may also be OSPF or the like, which IS not limited in this embodiment of the present application.

Because an interface connected with a CCU on the MPU runs an IS-IS protocol, the MPU can generate a system ID (System ID) of the equipment, wherein the system ID IS a bridge MAC (media access control) address of the equipment or an ID of a machine frame where the equipment IS located and an ID of a slot position where the equipment IS located; then, the MPU sends a Hello (handshake) message through the interface connection running the IS-IS protocol, so as to establish an IS-IS neighbor relationship with the directly connected opposite-end CCU.

Similarly, because all interfaces on the CCU run the IS-IS protocol, the CCU can generate the system ID of the equipment, wherein the system ID IS the bridge MAC address of the equipment, or the ID of the machine frame where the equipment IS located and the ID of the slot position where the equipment IS located; and then, the CCU sends a Hello message through all interfaces so as to establish an IS-IS neighbor relation with an opposite-end MPU or an opposite-end CCU which are directly connected.

In an actual implementation process, when the system ID is an ID of a subrack where the device is located and an ID of a slot where the device is located, the form of the system ID may be, for example: ID.00 of slot ID. of region ID. subrack, where one cluster system has one region ID.

Specifically, as shown in fig. 2, the method for establishing the IS-IS neighbor relationship between the local designated cluster device and the directly connected opposite designated cluster device IS as follows:

step S101, sending a Hello message (for convenience of description, called as a first Hello message) carrying a system ID of the device and specific indication information to an opposite terminal appointed cluster device, wherein the specific indication information is used for indicating that the device supports address information consisting of an ID of a machine frame, an ID of a slot position and an ID of a CPU;

similarly, the opposite-end designated cluster device also executes step S101 to send a first Hello packet to the home-end designated cluster device.

Step S102, after receiving a first Hello message sent by an opposite-end appointed cluster device, sending a Hello message (for convenience of description, marked as a second Hello message) carrying a system ID of the device, the system ID of the opposite-end appointed cluster device and the specific indication information to the opposite-end appointed cluster device;

similarly, the opposite-end designated cluster device also executes step S102 to send a second Hello packet to the home-end designated cluster device.

Step S103, after receiving the second Hello message sent by the opposite terminal appointed cluster equipment, establishing an IS-IS neighbor relation with the opposite terminal appointed cluster equipment.

In step S103, the home-end-designated cluster device records the system ID, the link status, and HoldTime (hold time) of the peer-end-designated cluster device in the neighbor relation table, where the link status is set to UP (normal).

Similarly, the opposite-end designated cluster device also executes step S103, so that the local-end designated cluster device establishes an IS-IS neighbor relationship with the opposite-end designated cluster device, where when the local-end designated cluster device IS an MPU, the opposite-end designated cluster device IS a CCU, and when the local-end designated cluster device IS a CCU, the opposite-end designated cluster device IS an MPU or a CCU.

After establishing the IS-IS neighbor relationship with the directly connected peer-to-peer designated cluster device, the peer-to-peer designated cluster device needs to perform the operations shown in fig. 3:

step S201, acquiring address information of CPUs on all cluster equipment in a machine frame where the equipment is located, wherein the address information consists of an ID of the machine frame where the cluster equipment is located, an ID of a slot where the cluster equipment is located and an ID of the CPU;

as shown in fig. 4, the specific acquisition method is as follows:

step S301, acquiring information of all MPUs in the subrack where the equipment is located, wherein the information comprises: the ID of the machine frame where the MPU is located and the ID of the slot position where the MPU is located;

step S302, acquiring IDs of all CPUs on each MPU;

thus, in step S301 and step S302, the address information of the CPU on the MPU is obtained as the ID of the subrack in which the MPU is located, the ID of the slot in which the MPU is located, and the ID of the CPU.

Step S303, obtaining information of all CCUs in the subrack where the device is located, where the information includes: the ID of the machine frame where the CCU is located and the ID of the slot position where the CCU is located;

step S304, obtaining the IDs of all CPUs on each CCU;

thus, in step S303 and step S304, the address information of the CPU in the CCU is obtained as the ID of the subrack in which the CCU is located, the ID of the slot in which the CCU is located, and the ID of the CPU.

Step S305, obtaining information of all LPUs in the subrack where the device is located, where the information includes: the ID of the machine frame where the LPU is located and the ID of the slot position where the LPU is located;

step S306, obtaining the IDs of all CPUs on each LPU;

thus, in step S305 and step S306, the address information of the CPU in the LPU is obtained as the ID of the subrack in which the LPU is located, the ID of the slot in which the LPU is located, and the ID of the CPU.

Obviously, if the MPU, CCU or LPU does not exist in the subrack where the present device is located, the corresponding acquisition step does not need to be performed.

Step S202, sending LSP (Link State PDUs) messages carrying the acquired address information;

in an actual implementation process, the address information acquired in steps S301 to S306 needs to be added to a TLV (Type Length Value), and then the TLV is added to an LSP packet and sent out, and the LSP packet is stored in an LSDB (Link State DataBase).

In the IS-IS protocol, the LSP packet IS sent in a flooding manner, so that the LSP packet can be flooded to all other designated cluster devices in the cluster system. And other appointed cluster devices also send LSP messages carrying the address information acquired by the other appointed cluster devices.

Step S203, receiving LSP messages with address information sent by other specified cluster devices.

In step S203, the received LSP packet is also stored in the LSDB.

Step S204, for each address information carried in the received LSP packet, an SPF (Shortest Path First) protocol is used to calculate a route from the device to the CPU identified by the address information.

When performing routing calculation, for each address information carried in a received LSP packet, a shortest forwarding path from the device to the CPU identified by the address information is calculated, and if at least two shortest forwarding paths are calculated, that is, costs of the at least two shortest forwarding paths are equal, one of the shortest forwarding paths that satisfies a predetermined condition is selected as a route from the device to the CPU, and a corresponding routing table entry is added to the routing forwarding table shown in table 1.

TABLE 1

Destination address information	Next hop information

The next hop information in table 1 includes: the present device connects the interface (i.e., the outgoing interface) of the next-hop device and the MAC address of the next-hop device, etc.

The predetermined condition may be, for example: the ID of the egress interface is maximum or minimum.

Because the CPU is located on the cluster device, the calculated route from the device to the CPU is the route from the CPU on the device to the CPU, and is also the route from the device to the cluster device where the CPU is located.

Therefore, the CPU on the designated cluster equipment can access other CPUs in different machine frames by utilizing the route calculated by the method of the embodiment of the application, and can also access other CPUs in the same machine frame. In addition, since default routes to other cluster devices in the same subrack are stored in the cluster device in the prior art, when CPUs on an MPU and an LPU that do not run the IS-IS protocol need to access CPUs on cluster devices in different subracks (called destination cluster devices for convenience of description), the default route may be first passed through to a designated cluster device in the same subrack, and the designated cluster device then passes through the route calculated by the method in the embodiment of the present application to reach the CPU on the destination cluster device. Therefore, the mutual access among the CPUs in the cluster system is finally realized.

Subsequently, when the cluster device in the subrack changes, for example, the cluster device is added or reduced, at this time, as shown in fig. 5, the local designated cluster device executes the following steps:

step S401, judging whether cluster equipment is added or reduced in a machine frame where the equipment is located, if the cluster equipment is added, executing step S402, and if the cluster equipment is reduced, executing step S404;

the newly added cluster device may be an MPU, a CCU, or an LPU, and the newly added cluster device may be an MPU, a CCU, or an LPU.

Step S402, acquiring address information of all CPUs on the newly added cluster equipment;

step S403, sending LSP message carrying address information of CPU on the newly added cluster device;

in an actual implementation process, the address information of the CPU on the newly added cluster device may be added to the TLV, and then the TLV is added to the LSP packet and sent out, so that all other designated cluster devices in the cluster system may be flooded. And the other appointed cluster equipment receives the LSP message and stores the LSP message into the LSDB.

Step S404, acquiring the address information of the CPU on the reduced cluster equipment;

step S405, searching sent LSP messages carrying reduced address information of CPUs on the cluster equipment;

step S406, after deleting the address information of the CPU on the reduced cluster device from the sent LSP packet, sending the sent LSP packet.

In an actual implementation process, TLVs carrying reduced address information of CPUs on the cluster device may be searched from the sent LSP message, the reduced address information of the CPUs on the cluster device is deleted from the searched TLVs, and then the sent LSP message is sent out, so that all other designated cluster devices in the cluster system can be flooded. After receiving the LSP packet (for convenience of description, denoted as LSP packet 1), the other designated cluster devices find that there is a received LSP packet in the LSDB that is the same as the packet ID (i.e., fragment number) of the LSP packet, and update the received LSP packet to LSP packet 1.

When detecting that the topology of the cluster system changes, for example, the IS-IS neighbor relationship between the directly connected opposite-end designated cluster device IS deleted, the IS-IS neighbor relationship IS established with a directly connected new opposite-end designated cluster device, and the link IS interrupted or restored, the cluster device IS newly added or reduced in the machine frame where the local device IS located, a new LSP message or the like sent from other designated cluster devices IS received, the local-end designated cluster device recalculates the route from the local device to the CPU identified by the address information according to each address information carried in the received LSP message in the LSDB, the calculation method IS the same as above, and the route forwarding table shown in table 1 IS updated.

In the method of this embodiment of the present application, a link state routing protocol is run on an MPU and a CCU (collectively referred to as designated cluster devices) in a cluster system, and after a neighbor relation is established between a local designated cluster device and a directly connected opposite designated cluster device, address information of a CPU on the cluster device (MPU, CCU, LPU) in a machine frame where the local designated cluster device is located is obtained and carried in a link state routing packet to be flooded to all other designated cluster devices in the cluster system, where the address information is composed of an ID of the machine frame where the cluster device is located, an ID of a slot where the cluster device is located, and an ID of the CPU; and receiving a link state routing protocol message carrying address information sent by other specified cluster equipment, and calculating a route from the equipment to the CPU identified by the address information by using an SPF (shortest path first) protocol according to each address information carried in the received link state routing protocol message, thereby realizing mutual access between the CPUs.

Because the link state routing protocol is expanded with address information consisting of machine frame ID, slot ID and CPU ID, and the link state routing protocol is run on MPU and CCU, the appointed cluster devices running the link state routing protocol can synchronize the address information of the CPU on all cluster devices in the machine frame to other appointed cluster devices, thereby realizing the synchronization of the address information of the CPU on the cluster devices in each machine frame; then, the shortest path from the present device to each CPU can be calculated using the SPF protocol. The routes calculated by the SPF protocol do not form a loop, and when the routes are disconnected, new routes can be quickly recalculated, so that the topology convergence speed is high.

The above method is described in detail by taking the cluster system shown in fig. 1 as an example. In fig. 1, the central exchange box 1 has CCU1, CCU3, MPU1, and MPU2, wherein MPU1 has a CPU1, and MPU2 has a CPU 2; the central exchange frame 2 is provided with a CCU2, a CCU4, an MPU3 and an MPU4, wherein the MPU3 is provided with a CPU3, and the MPU4 is provided with a CPU 4; the wire card frame 3 is provided with an MPU5, an MPU6 and an LPU1, wherein the MPU5 is provided with a CPU5, the MPU6 is provided with a CPU6, and the LPU1 is provided with a CPU 9; the wire card frame 4 is provided with an MPU7, an MPU8 and an LPU2, wherein the MPU7 is provided with a CPU7, the MPU8 is provided with a CPU8, and the LPU2 is provided with a CPU 10.

In the cluster system shown in fig. 1, the CPUs in the central switch box may exchange accesses with each other, the CPUs in the line card box may exchange accesses with each other, and the CPUs in the central switch box and the CPUs in the line card box may also exchange accesses with each other. The specific implementation process is as follows:

the interface connected with the CCU1 on the MPU1 runs an IS-IS protocol, the interface connected with the CCU2 on the MPU4 runs an IS-IS protocol, the interface connected with the CCU3 on the MPU5 runs an IS-IS protocol, the interface connected with the CCU4 on the MPU8 runs an IS-IS protocol, and all the interfaces on the CCU1, the CCU2, the CCU3 and the CCU4 run an IS-IS protocol.

In the center box exchange box 1, the MPU1 (slot ID of 1) generates a system ID of 01.0000.0001.0001.00; the CCU1 (slot ID 3) generates a system ID of 01.0000.0001.0003.00, and the CCU3 (slot ID 4) generates a system ID of 01.0000.0001.0004.00;

in the center box exchange box 2, the MPU4 (slot ID of 2) generates a system ID of 01.0000.0002.0002.00, the CCU2 (slot ID of 3) generates a system ID of 01.0000.0002.0003.00, and the CCU4 (slot ID of 4) generates a system ID of 01.0000.0002.0004.00;

in the line card box 3, the MPU5 (slot ID of 1) generates a system ID of 01.0000.0003.0001.00;

in the line card box 4, the MPU8 (slot ID of 2) generates a system ID of 01.0000.0004.0002.00;

the MPU1 and the CCU1 establish an IS-IS neighbor relation through interacting Hello messages; the MPU4 and the CCU2 establish an IS-IS neighbor relation through interacting Hello messages; meanwhile, the CCU1, the CCU2, the CCU3 and the CCU4 establish an IS-IS neighbor relation through mutual Hello messages; the MPU5 and the CCU3 establish an IS-IS neighbor relation through interacting Hello messages; the MPU8 and the CCU4 establish IS-IS neighbor relation through mutual Hello messages.

Then, the MPU1 transmits an LSP packet carrying address information 1 (address information of CPU 1) and address information 2 (address information of CPU 2) to thereby flood CCUs 1 to CCU4, MPU4, MPU5, and MPU8, where the address information 1 is 01 (chassis ID) +01 (slot ID of MPU 1) +01 (ID of CPU 1), and the address information 2 is 01 (chassis ID) +02 (slot ID of MPU 2) +02 (ID of CPU 2); similarly, CCU1 and CCU3 also send LSP packets carrying address information 1 and address information 2.

The MPU4 transmits an LSP packet carrying the following address information 3 (address information of the CPU 3) and address information 4 (address information of the CPU 4), thereby flooding to the CCUs 1 to 4, the MPU1, the MPU5, and the MPU 8; wherein, the address information 3 is 02 (machine frame ID) +01 (slot ID of MPU 3) +01 (ID of CPU 3), and the address information 4 is 02 (machine frame ID) +02 (slot ID of MPU 4) +02 (ID of CPU 4); similarly, CCU2 and CCU4 also send LSP packets carrying address information 3 and address information 4.

The MPU5 transmits an LSP packet carrying the following address information 5 (address information of the CPU 5), address information 6 (address information of the CPU 6), and address information 7 (address information of the CPU 9), thereby flooding to the CCUs 1 to 4, MPU1, MPU4, and MPU 8; wherein, address information 5 is 03 (machine frame ID) +01 (slot ID of MPU 5) +01 (ID of CPU 5), address information 6 is 03 (machine frame ID) +02 (slot ID of MPU 6) +02 (ID of CPU 6), and address information 7 is 03 (machine frame ID) +03 (slot ID of LPU 1) +03 (ID of CPU 9);

the MPU8 transmits an LSP packet carrying the following address information 8 (address information of the CPU 7), address information 9 (address information of the CPU 8), and address information 10 (address information of the CPU 10), thereby flooding to the CCUs 1 to 4, MPU1, MPU4, and MPU 5; wherein, address information 8 is 04 (machine frame ID) +01 (slot ID of MPU 7) +01 (ID of CPU 7), address information 9 is 04 (machine frame ID) +02 (slot ID of MPU 8) +02 (ID of CPU 8), and address information 10 is 04 (machine frame ID) +03 (slot ID of LPU 2) +03 (ID of CPU 10);

the MPU1, MPU4, MPU5, MPU8, and CCU1 to CCU4 each specify a cluster device, and calculate a route from the own device to the CPU identified by the address information for each piece of address information carried in the received LSP message in the LSDB. For example, the next hop of the route from the device to the CPU4 identified by the address information 4 calculated by the MPU5 is CCU3, the next hop of the route from the device to the CPU4 identified by the address information 4 calculated by the CCU3 is CCU4, the next hop of the route from the device to the CPU4 identified by the address information 4 calculated by the CCU4 is CCU2, and the next hop of the route from the device to the CPU4 identified by the address information 4 calculated by the CCU2 is MPU4, so that when the CPU5 accesses the CPU4, the message can pass through the routing MPUs 5-CCU 3-CCU 4-CCU 2-MPU 4 and finally reaches the CPU 4.

When the link between the CCU3 and the CCU4 fails, the CCU3 recalculates the route, and the next hop of the recalculated route from the device to the CPU4 identified by the address information 4 is the CCU 1. Thereafter, when the CPU5 wants to access the CPU4, the message may traverse the path MPU 5-CCU 3-CCU 1-CCU 2-MPU 4, eventually reaching the CPU 4.

In addition, when the CPU9 wants to access the CPU3, the LPU1 may send a message to the MPU5 via a default route, the MPU5 sends the message to the MPU4 via the routes MPU 5-CCU 3-CCU 4-CCU 2-MPU 4, and the MPU4 finally sends the message to the CPU3 on the MPU3 via the default route.

Corresponding to the foregoing embodiment of the route obtaining method in the cluster system, the present application also provides an embodiment of a route obtaining apparatus in the cluster system. The route acquisition device can be applied to specified cluster equipment running a link state route protocol in a cluster system, wherein the cluster equipment is an MPU (micro processing unit), a CCU (central processing unit) or an LPU (low power unit), and the specified cluster equipment is the MPU or the CCU.

Referring to fig. 6, a route obtaining apparatus according to an embodiment of the present application includes the following modules: a neighbor establishing module 601, an address obtaining module 602, a sending module 603, a receiving module 604 and a route calculating module 605, wherein:

a neighbor establishing module 601, configured to establish a neighbor relationship with a directly connected opposite-end designated cluster device;

an address obtaining module 602, configured to obtain address information of a CPU on cluster equipment in a subrack where the equipment is located after a neighbor relation is established between the neighbor establishing module 601 and an opposite-end specified cluster equipment, where the address information is composed of an ID of the subrack where the cluster equipment is located, an ID of a slot where the cluster equipment is located, and an ID of the CPU;

a sending module 603, configured to send a link state routing protocol packet carrying the address information obtained by the address obtaining module 602;

a receiving module 604, configured to receive a link state routing protocol packet carrying address information;

a route calculating module 605, configured to calculate, by using an SPF protocol, a route from the device to the CPU identified by the address information, for each address information carried in the link state routing protocol packet received by the receiving module 604.

The interface connected with the CCU on the MPU runs a link state routing protocol, and all the interfaces on the CCU run the link state routing protocol; when the device is an MPU, the opposite end designates the cluster device as a CCU; and when the equipment is the CCU, the opposite end designates the cluster equipment as the MPU or the CCU.

When the link state routing protocol IS an IS-IS protocol, the neighbor establishing module 601 IS specifically configured to establish a neighbor relationship with an opposite-end-designated cluster device that IS directly connected, through the following steps:

generating a system ID of the equipment, wherein the system ID is a bridge MAC address of the equipment, or the ID of a machine frame where the equipment is located and the ID of a slot position where the equipment is located;

sending a first Hello message carrying a system ID of the equipment and specific indication information to opposite-end appointed cluster equipment, wherein the specific indication information is used for indicating that the equipment supports address information consisting of the ID of a machine frame, the ID of a slot position and the ID of a CPU;

after receiving a first Hello message sent by an opposite terminal appointed cluster device, sending a second Hello message carrying a system ID of the device, the system ID of the opposite terminal appointed cluster device and specific indication information to the opposite terminal appointed cluster device;

and after receiving a second Hello message sent by the opposite-end appointed cluster equipment, establishing an IS-IS neighbor relation with the opposite-end appointed cluster equipment.

As shown in fig. 7, the route acquisition device further includes: a detection module 606, wherein:

a detection module 606, configured to detect whether cluster devices are added or subtracted in a subrack where the device is located;

the address obtaining module 602 is further configured to, when the detecting module 606 detects that a cluster device is newly added to the machine frame where the device is located, obtain address information of a CPU on the newly added cluster device;

a sending module 603, configured to send a link state routing protocol packet carrying the address information of the CPU on the newly added cluster device, where the address information is obtained by the address obtaining module 602;

the receiving module 604 is further configured to receive a link state routing protocol packet carrying address information of a CPU on the newly added cluster device.

As shown in fig. 8, the route acquisition device further includes: find delete module 607, wherein:

the address obtaining module 602 is further configured to, when the detecting module 606 detects that cluster equipment is reduced in the subrack where the equipment is located, obtain address information of CPUs on the reduced cluster equipment;

a searching and deleting module 607, configured to search for a sent link state routing protocol packet carrying address information of the CPU on the cluster device that is reduced, delete the address information of the CPU on the cluster device that is reduced from the sent link state routing protocol packet, and output the message to the sending module 603;

the sending module 603 is further configured to send the sent LSP packet input by the searching and deleting module 607;

the receiving module 604 is further configured to update the received link state routing protocol packet to the received link state routing protocol packet after receiving the link state routing protocol packet carrying the reduced address information of the CPU on the cluster device.

The route calculation module 605 is specifically configured to calculate a route from the local device to the CPU identified by the address information by the following steps:

calculating the shortest forwarding path from the equipment to the CPU identified by the address information to determine an interface;

when at least two shortest forwarding paths exist, selecting the shortest forwarding path meeting a preset condition as a route from the equipment to the CPU, wherein the preset condition comprises the following steps: the ID of the egress interface is maximum or minimum.

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 application and should not be taken as limiting the present application, as any modification, equivalent replacement, or improvement made within the spirit and principle of the present application should be included in the scope of protection of the present application.

Claims (12)

1. A method for acquiring a route in a cluster system is applied to a designated cluster device running a link state routing protocol in the cluster system, and the method comprises the following steps:

after establishing a neighbor relation with a specified cluster device of a directly connected opposite terminal, acquiring address information of Central Processing Units (CPUs) on all cluster devices in a machine frame where the device is located; the address information consists of an identifier ID of a machine frame where the cluster equipment is located, an ID of a slot position where the cluster equipment is located and an ID of the CPU, the cluster equipment is a main control board MPU, a central control unit CCU or an interface board LPU, and the designated cluster equipment running the link state routing protocol is the MPU or the CCU;

sending a link state routing protocol message carrying the acquired address information;

receiving a link state routing protocol message carrying address information;

and aiming at each address information carried in the received link state routing protocol message, calculating the route from the equipment to the CPU identified by the address information by using a Shortest Path First (SPF) protocol.

2. The method of claim 1, wherein the interface on the MPU to which the CCU is connected runs the link state routing protocol, and all interfaces on the CCU run the link state routing protocol;

when the device is an MPU, the opposite end designates the cluster device as a CCU;

and when the equipment is a CCU, the opposite end designates the cluster as an MPU or a CCU.

3. The method of claim 1, wherein when the link state routing protocol IS an intermediate system to intermediate system IS-IS protocol, the method for establishing a neighbor relation with a directly connected peer-designated cluster device comprises:

generating a system ID of the equipment, wherein the system ID is a bridge MAC address of the equipment, or the ID of a machine frame where the equipment is located and the ID of a slot position where the equipment is located;

sending a first Hello message carrying a system ID of the equipment and specific indication information to the opposite-end appointed cluster equipment, wherein the specific indication information is used for indicating that the equipment supports address information consisting of the ID of a machine frame, the ID of a slot position and the ID of a CPU;

after receiving a first Hello message sent by the opposite-end appointed cluster device, sending a second Hello message carrying a system ID of the opposite-end appointed cluster device, the system ID of the opposite-end appointed cluster device and the specific indication information to the opposite-end appointed cluster device;

and after receiving a second Hello message sent by the opposite-end appointed cluster equipment, establishing an IS-IS neighbor relation with the opposite-end appointed cluster equipment.

4. The method of claim 1, further comprising:

when detecting that cluster equipment is newly added in a machine frame where the equipment is positioned, acquiring address information of a CPU (central processing unit) on the newly added cluster equipment, and sending a link state routing protocol message carrying the address information of the CPU on the newly added cluster equipment;

and receiving a link state routing protocol message carrying the address information of the CPU on the newly added cluster equipment.

5. The method of claim 4, further comprising:

when detecting that cluster equipment is reduced in a machine frame where the equipment is located, acquiring address information of CPUs (central processing units) on the reduced cluster equipment, searching for a sent link state routing protocol message carrying the address information of the CPUs on the reduced cluster equipment, deleting the address information of the CPUs on the reduced cluster equipment from the sent link state routing protocol message, and then sending the sent link state routing protocol message;

after receiving the link state routing protocol message carrying the address information of the CPU on the cluster equipment, updating the corresponding received link state routing protocol message into the received link state routing protocol message.

6. The method according to any one of claims 1 to 5, wherein the method of calculating the route from the own device to the CPU identified by the address information includes:

calculating the shortest forwarding path from the equipment to the CPU identified by the address information to determine an interface;

when at least two shortest forwarding paths exist, selecting the shortest forwarding path meeting a predetermined condition as a route from the device to the CPU, wherein the predetermined condition comprises: the ID of the egress interface is maximum or minimum.

7. A device for acquiring a route in a cluster system, wherein the device is applied to a designated cluster device running a link state routing protocol in the cluster system, and the device comprises:

the neighbor establishing module is used for establishing a neighbor relation with the directly connected opposite terminal appointed cluster equipment;

an address obtaining module, configured to obtain address information of CPUs on all cluster devices in a subrack of the device after the neighbor establishing module establishes a neighbor relationship with the opposite-end specified cluster device, where the address information is composed of an identifier ID of the subrack where the cluster device is located, an ID of a slot where the cluster device is located, and an ID of the CPU; the cluster equipment is a main control board MPU, a central control unit CCU or an interface board LPU, and the designated cluster equipment running the link state routing protocol is the MPU or the CCU;

a sending module, configured to send a link state routing protocol packet carrying the address information obtained by the address obtaining module;

the receiving module is used for receiving a link state routing protocol message carrying address information;

and the route calculation module is used for calculating the route from the equipment to the CPU identified by the address information by using a Shortest Path First (SPF) protocol according to each address information carried in the link state routing protocol message received by the receiving module.

8. The apparatus of claim 7, wherein the interface on the MPU to which the CCU is connected runs the link state routing protocol, and all interfaces on the CCU run the link state routing protocol;

when the device is an MPU, the opposite end designates the cluster device as a CCU;

and when the equipment is CCU, the opposite end designates the cluster equipment as MPU or CCU.

9. The apparatus according to claim 7, wherein when the link state routing protocol IS an intermediate system to intermediate system IS-IS protocol, the neighbor establishing module IS specifically configured to establish a neighbor relationship with a directly connected peer-designated cluster device by:

generating a system ID of the equipment, wherein the system ID is a bridge MAC address of the equipment, or the ID of a machine frame where the equipment is located and the ID of a slot position where the equipment is located;

sending a first Hello message carrying a system ID of the equipment and specific indication information to the opposite-end appointed cluster equipment, wherein the specific indication information is used for indicating that the equipment supports address information consisting of the ID of a machine frame, the ID of a slot position and the ID of a CPU;

after receiving a first Hello message sent by the opposite-end appointed cluster equipment, sending a second Hello message carrying a system ID of the opposite-end appointed cluster equipment, the system ID of the opposite-end appointed cluster equipment and the specific indication information to the opposite-end appointed cluster equipment;

and after receiving a second Hello message sent by the opposite-end appointed cluster equipment, establishing an IS-IS neighbor relation with the opposite-end appointed cluster equipment.

10. The apparatus of claim 7, further comprising: a detection module, wherein:

the detection module is used for detecting whether cluster equipment is added or reduced in a machine frame where the equipment is located;

the address acquisition module is further configured to acquire address information of a CPU on the newly added cluster device when the detection module detects that the cluster device is newly added to the machine frame where the device is located;

the sending module is further configured to send a link state routing protocol packet carrying the address information of the CPU on the newly added cluster device, which is obtained by the address obtaining module;

the receiving module is further configured to receive a link state routing protocol packet carrying address information of a CPU on the newly added cluster device.

11. The apparatus of claim 10, further comprising: a find delete module, wherein:

the address acquisition module is further configured to acquire address information of a CPU on reduced cluster equipment when the detection module detects that cluster equipment is reduced in the subrack where the equipment is located;

the searching and deleting module is configured to search for a sent link state routing protocol packet carrying address information of the CPU on the reduced cluster device, delete the address information of the CPU on the reduced cluster device from the sent link state routing protocol packet, and output the deleted address information to the sending module;

the sending module is further configured to send the sent link state routing protocol packet input by the searching and deleting module;

the receiving module is further configured to update the received link state routing protocol packet to the received link state routing protocol packet after receiving the link state routing protocol packet carrying the reduced address information of the CPU on the cluster device.

12. The apparatus according to any one of claims 7 to 11, wherein the route calculation module is specifically configured to calculate the route from the local device to the CPU identified by the address information by:

calculating the shortest forwarding path from the equipment to the CPU identified by the address information to determine an interface;

when at least two shortest forwarding paths exist, selecting the shortest forwarding path meeting a predetermined condition as a route from the device to the CPU, wherein the predetermined condition comprises: the ID of the egress interface is maximum or minimum.

Images