CN104426794B - A kind of message forwarding method and device - Google Patents

Abstract

The invention discloses a message forwarding method for saving address pool resources. The method includes: allocating M address pool blocks corresponding to the first single board that has failed to other single boards in the load sharing group; The single board identification of the single board corresponding to the block identification is correspondingly updated to the assigned single board identification, and the second load sharing table is obtained; the second load sharing table is issued to each single board in the load sharing group, so that a single board in the load sharing group assigns a source public network address to the received message according to the second load sharing table, and forwards the message to the received message according to the destination public network address carried in the message to retweet. The invention also discloses another message forwarding method and a corresponding device.

Description

A message forwarding method and device

technical field

The invention relates to the technical field of communications, in particular to a message forwarding method and device.

Background technique

CGN, that is, carrier-level NAT (Network Address Translation, Network Address Translation). CGN equipment is located inside the operator, at the boundary between the public IPv4 Internet network and the private IPv4 address network, and is used to provide high-performance translation from IPv4 private addresses to public addresses.

In the prior art, a CGN single board is generally reserved in a load sharing group and does not work. When the working CGN single board fails, the reserved CGN single board is enabled to join the load sharing group to work.

Disadvantages of this solution: Because one CGN single board is reserved for non-working, all load sharing members cannot participate in the load sharing work, resulting in waste of hardware resources. When the working CGN board fails, the reserved CGN board joins the load sharing group to work, and the address pool resources of the faulty CGN board cannot be released, which actually causes a waste of address pool resources.

Contents of the invention

Embodiments of the present invention provide a message forwarding method and device, which are used to solve the technical problem of waste of address pool resources that may occur in a load sharing group in the prior art.

A first aspect of the present invention provides a message forwarding method, the method comprising the following steps:

Allocating the M address pool blocks corresponding to the first single board that has failed to other single boards in the load sharing group;

correspondingly update the single board identifications of the single boards corresponding to the M address pool block identifications of the M address pool blocks in the first load sharing table to the assigned single board identifications respectively, to obtain the second load sharing table;

sending the second load sharing table to each board in the load sharing group, so that a board in the load sharing group distributes the received message according to the second load sharing table source public network address, and forward the message according to the destination public network address carried in the message.

In combination with the first aspect, in a first possible implementation manner, before allocating the M address pool blocks corresponding to the first single board that has failed to other single boards in the load sharing group, the method further includes:

dividing the address pool into at least one address pool block;

Allocating the at least one address pool block to each board in the load sharing group.

With reference to the first aspect or the first possible implementation manner, in the second possible implementation manner, after sending the second load sharing table to each board in the load sharing group, the method further includes:

After the failure of the first single board is recovered, at least one target single board is selected from other single boards in the load sharing group, and part of the address pool block corresponding to each target single board is allocated to the first single board, and causing each target single board to send address pool block information corresponding to the address pool block allocated to the first single board to the first single board;

updating the single board identification of the single board corresponding to the address pool block identifier of the address pool block assigned to the first single board in the second load sharing table to the first single board identification, to obtain a third load sharing table ;

Sending the third load sharing table to each board in the load sharing group.

With reference to the second possible implementation, in the third possible implementation, after sending the third load sharing table to each board in the load sharing group, the method further includes:

Adding the first single board to the load sharing group corresponding to the third load sharing table, so that the first single board starts working.

A second aspect of the present invention provides a message forwarding method, the method comprising the following steps:

After obtaining the second load sharing table, obtain a message from an electronic device, where the message at least carries an active private network address and a destination public network address;

Processing the source private network address according to a first operation rule to obtain a corresponding address pool block identifier; wherein, the first operation rule is: when the input information is the same, the output information is also the same;

According to the second load sharing table, assigning a source public network address to the message through the address pool block corresponding to the address pool block identifier, and forwarding the message according to the destination public network address.

With reference to the second aspect, in a first possible implementation manner, according to the second load sharing table, the address pool block corresponding to the address pool block identifier is used to assign a source public network address to the packet, and according to the The destination public network address forwards the message, including:

Determine according to the second load sharing table whether the board ID of the board corresponding to the address pool block ID is the board ID of the current board or the board ID of other boards;

When the board ID of the board corresponding to the address pool block ID is the board ID of the current board, assigning a source public network address to the message through the address pool block corresponding to the address pool block ID, and forwarding the message according to the destination public network address;

When the board ID of the board corresponding to the address pool block ID is the board ID of other boards, forward the message to the board corresponding to the other board ID, so that the The address pool block corresponding to the address pool block identifier allocates a source public network address for the message, and forwards the message according to the destination public network address.

A third aspect of the present invention provides a message forwarding device, the device comprising:

The first allocation unit is configured to allocate the M address pool blocks corresponding to the first single board that has failed to other single boards in the load sharing group;

The first update unit is configured to update the board IDs of the boards corresponding to the M address pool block IDs of the M address pool blocks in the first load sharing table to the assigned board IDs respectively, and obtain the second load sharing table;

The first sending unit is configured to send the second load sharing table to each single board in the load sharing group, so that a single board in the load sharing group, according to the second load sharing table, Allocating a source public network address to the received message, and forwarding the message according to the destination public network address carried in the message.

With reference to the third aspect, in a first possible implementation manner, the device further includes a dividing unit and a second allocating unit;

The division unit is used to divide the address pool into at least one address pool block;

The second allocation unit is configured to allocate the at least one address pool block to each board in the load sharing group.

With reference to the third aspect or the first possible implementation manner, in a second possible implementation manner, the device further includes a third allocating unit and a second updating unit;

The third allocation unit is configured to select at least one target single board from other single boards in the load sharing group after the first single board fails to recover, and allocate part of the address pool block corresponding to each target single board to the first single board, and make each target single board send address pool block information corresponding to the address pool block allocated to the first single board to the first single board;

The second updating unit is configured to update the single board identifier of the single board corresponding to the address pool block identifier of the address pool block allocated to the first single board in the second load sharing table to the first single board identifier, obtaining a third load sharing table;

The first sending unit is further configured to send the third load sharing table to each board in the load sharing group.

With reference to the second possible implementation manner, in a third possible implementation manner, the device further includes an operation unit configured to add the first single board to the load sharing group corresponding to the load sharing table, so that The first veneer starts to work.

A fourth aspect of the present invention provides a message forwarding device, the device comprising:

An obtaining unit, configured to obtain a message from an electronic device after obtaining the second load sharing table, the message carrying at least an active private network address and a destination public network address;

A processing unit, configured to process the source private network address according to a first operation rule to obtain a corresponding address pool block identifier; wherein, the first operation rule is: when the input information is the same, the output information is also the same;

The fourth allocating unit is configured to allocate a source public network address to the message according to the address pool block corresponding to the address pool block identifier according to the second load sharing table, and allocate the source public network address to the message according to the destination public network address The message is forwarded.

With reference to the fourth aspect, in a first possible implementation manner, the fourth allocating unit is specifically configured to: determine, according to the second load sharing table, that the board identifier of the board corresponding to the address pool block identifier is this The single board identification of the single board is also the single board identification of other single boards; when the single board identification of the single board corresponding to the address pool block identification is the single board identification of this single board, all The address pool block allocates a source public network address for the message, and forwards the message according to the destination public network address; Forwarding the message to the board corresponding to the other board ID, so that the board allocates the message according to the address pool block corresponding to the address pool block ID source public network address, and forward the message according to the destination public network address.

A fifth aspect of the present invention provides a message forwarding device, the device comprising:

The first processor is configured to allocate M address pool blocks corresponding to the first single board that has failed to other single boards in the load sharing group; and allocate M address pool blocks of the M address pool blocks in the first load sharing table The board IDs of the boards corresponding to the address pool block IDs are respectively updated to the assigned board IDs to obtain the second load sharing table;

The first sending interface is configured to send the second load sharing table to each board in the load sharing group, so that a board in the load sharing group can, according to the second load sharing table, Allocating a source public network address to the received message, and forwarding the message according to the destination public network address carried in the message.

With reference to the fifth aspect, in a first possible implementation manner, the first processor is further configured to: divide the address pool into at least one address pool block; allocate the at least one address pool block to the load balancing Each board in the group.

With reference to the fifth aspect or the first possible implementation manner, in a second possible implementation manner, the first processor is further configured to: after the failure of the first single board recovers, the load sharing group Select at least one target single board from other single boards, allocate part of the address pool block corresponding to each target single board to the first single board, and make each target single board allocate to the first single board The address pool block information corresponding to the address pool block is sent to the first single board; the second update unit is used to identify the address pool block of the address pool block allocated to the first single board in the second load sharing table The single-board identification of the corresponding single board is updated to the first single-board identification, and a third load sharing table is obtained;

The first sending interface is further configured to deliver the third load sharing table to each board in the load sharing group.

With reference to the second possible implementation manner, in a third possible implementation manner, the first processor is further configured to add the first single board to the load sharing group corresponding to the load sharing table, so that all The above-mentioned first board starts to work.

A sixth aspect of the present invention provides a message forwarding device, the device comprising:

Obtaining an interface, configured to obtain a message from an electronic device after obtaining the second load sharing table, the message carrying at least an active private network address and a destination public network address;

The second processor is configured to process the source private network address according to a first operation rule to obtain a corresponding address pool block identifier; wherein, the first operation rule is: when the input information is the same, the output information is also the same ; according to the second load sharing table, assigning a source public network address to the message through the address pool block corresponding to the address pool block identifier, and forwarding the message according to the destination public network address.

With reference to the sixth aspect, in a first possible implementation manner, the second processor is specifically configured to: determine, according to the second load sharing table, that the board ID of the board corresponding to the address pool block ID is this The single board identification of the single board is also the single board identification of other single boards; when the single board identification of the single board corresponding to the address pool block identification is the single board identification of this single board, all The address pool block allocates a source public network address for the message, and forwards the message according to the destination public network address; Forwarding the message to the board corresponding to the other board ID, so that the board allocates the message according to the address pool block corresponding to the address pool block ID source public network address, and forward the message according to the destination public network address.

In the embodiment of the present invention, when the first single board fails, the M address pool blocks corresponding to the first single board can be allocated to other single boards in the load sharing group, for example, can be allocated to other A single board, or may also be assigned to other multiple single boards, and the single board identifiers of the single boards corresponding to the M address pool block identifiers of the M address pool blocks in the load sharing table are correspondingly updated to be assigned board ID, so that when the message needs to be forwarded, the address pool block ID corresponding to the message can be obtained, and then according to the corresponding relationship between the address pool block ID and the board ID in the load sharing table The message can be allocated to the correct single board to forward the message, so that when a single board fails, the address pool block corresponding to the single board can be allocated to other single boards, which improves the The utilization rate of address pool blocks avoids the waste of address pool resources as much as possible. And, since the load sharing table is also updated correspondingly when the address pool block is allocated, subsequent message forwarding can be carried out according to the updated load sharing table, which can ensure that the message is correctly forwarded and avoid message loss as much as possible. Reduce information loss rate.

In the embodiment of the present invention, after sending the second load sharing table to each single board, the single board allocates the source public network address and port number for the message, and the control plane does not need to perform the public network IP address and port number. The allocation of port numbers improves the allocation efficiency and can meet the demand for high-speed processing of information without adding burden to the control plane.

Description of drawings

Fig. 1 is the main flowchart of a kind of message forwarding method in the embodiment of the present invention;

Fig. 2 is the main flowchart of another message forwarding method in the embodiment of the present invention;

FIG. 3 is a structural diagram of a message forwarding device in an embodiment of the present invention;

FIG. 4 is a structural diagram of another message forwarding device in an embodiment of the present invention;

FIG. 5 is a schematic structural diagram of a message forwarding device in an embodiment of the present invention;

FIG. 6 is a schematic structural diagram of another message forwarding device in an embodiment of the present invention.

Detailed ways

An embodiment of the present invention provides a message forwarding method, the method may include: allocating the M address pool blocks corresponding to the first single board that has failed to other single boards in the load sharing group; The single board identifications of the single boards corresponding to the M address pool block identifications of the M address pool blocks in the above are respectively updated to the assigned single board identifications to obtain the second load sharing table; sending to each board in the load sharing group, so that a board in the load sharing group assigns a source public network address to the received message according to the second load sharing table, and according to the The destination public network address carried in the message forwards the message.

In the embodiment of the present invention, when the first single board fails, the M address pool blocks corresponding to the first single board can be allocated to other single boards in the load sharing group, for example, can be allocated to other A single board, or may also be assigned to other multiple single boards, and the single board identifiers of the single boards corresponding to the M address pool block identifiers of the M address pool blocks in the load sharing table are correspondingly updated to be assigned board ID, so that when the message needs to be forwarded, the address pool block ID corresponding to the message can be obtained, and then according to the corresponding relationship between the address pool block ID and the board ID in the load sharing table The message can be allocated to the correct single board to forward the message, so that when a single board fails, the address pool block corresponding to the single board can be allocated to other single boards, which improves the The utilization rate of address pool blocks avoids the waste of address pool resources as much as possible. And, since the load sharing table is also updated correspondingly when the address pool block is allocated, subsequent message forwarding can be carried out according to the updated load sharing table, which can ensure that the message is correctly forwarded and avoid message loss as much as possible. Reduce information loss rate.

In the embodiment of the present invention, after sending the second load sharing table to each single board, the single board allocates the source public network address and port number for the message, and the control plane does not need to perform the public network IP address and port number. The allocation of port numbers improves the allocation efficiency and can meet the demand for high-speed processing of information without adding burden to the control plane.

In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments It is a part of embodiments of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.

The technology described in this article can be used in various communication systems, such as current 2G, 3G communication systems and next-generation communication systems, such as Global System for Mobile Communications (GSM, Global System for Mobile communications), Code Division Multiple Access (CDMA, Code Division Multiple Access) system, Time Division Multiple Access (TDMA, Time Division Multiple Access) system, Wideband Code Division Multiple Access (WCDMA, Wideband Code Division Multiple Access Wireless), Frequency Division Multiple Access (FDMA, Frequency Division Multiple Addressing) system, Orthogonal Frequency Division Multiple Access Orthogonal Frequency-Division Multiple Access (OFDMA, Orthogonal Frequency-Division Multiple Access) system, Single Carrier FDMA (SC-FDMA) system, General Packet Radio Service (GPRS, General Packet Radio Service) system, Long Term Evolution (LTE, Long Term Evolution) system, and others such communication systems.

The electronic device in the embodiment of the present invention can be a wireless device or a wired device, and the wireless device can be a device that provides voice and/or data connectivity to the user, a handheld device with a wireless connection function, or a wireless modem other processing equipment. The wireless device can communicate with one or more core networks via a radio access network (for example, RAN, Radio Access Network). The wireless device can be a mobile device, such as a mobile phone (or called a "cellular" phone) and a mobile phone with a mobile device Computers, for example, may be portable, pocket, handheld, built-in, or vehicle-mounted mobile devices that exchange voice and/or data with the radio access network. For example, Personal Communication Service (PCS, Personal Communication Service) phones, cordless phones, Session Initiation Protocol (SIP) phones, Wireless Local Loop (WLL, Wireless Local Loop) stations, Personal Digital Assistant (PDA, Personal Digital Assistant) and other equipment . A wireless device can also be called a system, a subscriber unit (Subscriber Unit), a subscriber station (SubscriberStation), a mobile station (Mobile Station), a mobile station (Mobile), a remote station (Remote Station), an access point (Access Point), a remote Terminal (Remote Terminal), Access Terminal (Access Terminal), User Terminal (User Terminal), User Agent (User Agent), User Equipment (User Device), or User Equipment (User Equipment).

A wired device may be a device that provides voice and/or data connectivity to a user, a handheld device with a wired connection. Wired devices can communicate with one or more core networks, and wired devices can be fixed devices such as landline phones and desktop computers. Wired equipment can also be called system, subscriber unit (Subscriber Unit), subscriber station (Subscriber Station), remote station (Remote Station), access point (Access Point), remote terminal (Remote Terminal), access terminal (Access Terminal ), a user terminal (User Terminal), a user agent (User Agent), a user device (User Device), or a user equipment (User Equipment).

Additionally, the terms "system" and "network" are often used herein interchangeably. The term "and/or" in this article is just an association relationship describing associated objects, which means that there can be three relationships, for example, A and/or B can mean: A exists alone, A and B exist simultaneously, and there exists alone B these three situations. In addition, the character "/" in this article generally indicates that the contextual objects are an "or" relationship.

The embodiments of the present invention will be further described in detail below in conjunction with the accompanying drawings.

Embodiment one

Referring to Fig. 1, an embodiment of the present invention provides a message forwarding method, the main flow of the method may be as follows:

Step 101: Allocate the M address pool blocks corresponding to the first single board in failure to other single boards in the load sharing group.

In the embodiment of the present invention, before step 101, the address pool may be evenly distributed into several address pool blocks by the control plane, and each address pool may be numbered, wherein, in the embodiment of the present invention, one address pool block corresponds to The number corresponding to the address pool block can be called the address pool block identifier corresponding to the address pool block, for example, the number corresponding to the first address pool block can be called the first address pool block identifier, and so on.

Wherein, the address pool may include several public network IP addresses and corresponding port numbers.

At the same time or after numbering each address pool, each board included in a CGN load sharing group can be numbered, and then each address pool block can be delivered to each CGN board. After each CGN board receives its own address pool block, it can subsequently assign the public network IP addresses and port numbers in each address pool block to different messages, and these messages come from different electronic devices, and also It is equivalent to assigning public network IP addresses and port numbers to different electronic devices.

In the embodiment of the present invention, the message may carry a source address and a destination address, wherein the source address may be the private network address of the electronic device sending the message, so the source address may also be called the source private network address , the destination address may be a public network address of the electronic device to receive the message, so the destination address may also be called a destination public network address. In the embodiment of the present invention, allocating a public network IP address and a port number for a message is equivalent to allocating a source public network address for the message according to the source private network address of a message. After allocating a source public network address for a message, The message may be forwarded according to the destination public network address.

Allocating the source public network address to the message is equivalent to converting the source private network address of a message to the source public network address, so that the message can be transmitted in the public network.

In the embodiment of the present invention, if the maximum number of CGN single boards that can be inserted in a CGN load sharing group is the first number, then when the address pool is evenly distributed into several address pool blocks, the address pool can be evenly distributed as all The first number of blocks.

When allocating address pool blocks for each CGN single board in a CGN load balancing group, the allocation can be made according to specific conditions.

For example, a CGN load balancing group supports a maximum of 4 CGN boards inserted, so the address pool can be evenly distributed into 4 address pool blocks.

If the number of actually inserted CGN boards in the CGN load balancing group is 4, then an address pool block can be assigned to each of the 4 CGN boards.

If two CGN boards are actually inserted in the CGN load sharing group, two address pool blocks can be assigned to each of the two CGN boards.

If one CGN board is actually inserted in the CGN load balancing group, all the four address pool blocks can be allocated to the CGN board.

If the number of actually inserted CGN boards in the CGN load balancing group is 3, you can assign one address pool block to two of the CGN boards, and two address pool blocks to the remaining CGN board. Specifically assigning two address pool blocks to which CGN board can be randomly selected, or can also be selected according to other rules. As long as the distribution method conforms to the concept of the present invention, it is within the protection scope of the present invention.

Wherein, in the embodiment of the present invention, for convenience of description, a CGN load sharing group may be referred to as the load sharing group, and a CGN single board may be directly referred to as a single board.

In the embodiment of the present invention, in a load sharing group, when no fault occurs on each single board, each single board can work normally.

In this embodiment of the present invention, in a load sharing group, when a single board fails, for example, the faulty single board is the first single board, then the control plane can set the corresponding Address pool blocks are allocated to other boards in the load sharing group. For example, the M address pool blocks can be allocated to another board in the load sharing group, or the M address pool blocks can also be allocated to other boards in the load sharing group. The blocks are allocated to other multiple boards in the load sharing group, that is, at least the M address pool blocks can be allocated to the second boards in the load sharing group, wherein the first board and The second board is not the same board.

For example, the first single board corresponds to the M address pool blocks, then, the M address pool blocks can be allocated to the second single board, that is, the first single board can be assigned to All address pool blocks are allocated to the second single board, so as to avoid waste of address pool resources as much as possible.

Or for example, the first single board corresponds to the M address pool blocks, then, part of the address pool blocks in the M address pool blocks can be allocated to the second single board, and the M address pool blocks can be assigned to the second single board. The remaining address pool blocks in the address pool blocks are allocated to the third board, so as to avoid waste of address pool resources as much as possible.

Or for example, the first single board corresponds to the M address pool blocks, then the M address pool blocks may be allocated to other multiple single boards.

Specifically assigning each address pool block of the first veneer that has failed to several veneers may be arbitrarily determined, and specifically assigning each address pool block of the first veneer to which veneer may also be It can be determined arbitrarily, theoretically, as long as the selected board does not fail.

Step 102: Update the board IDs of the boards corresponding to the M address pool block IDs of the M address pool blocks in the first load sharing table to the assigned board IDs respectively, to obtain the second load sharing table.

In the embodiment of the present invention, the load sharing table may exist on each board, may be sent to each board by the control plane, and the control plane may also store the load sharing table. For example, if there are a second number of single boards in a load sharing group, each single board in the second number of single boards may store a load sharing table.

In the embodiment of the present invention, what is stored in the load sharing table may be the correspondence between the board identifiers in a load sharing group and the address pool block identifiers. For example, if in a load sharing group, the address pool block A is located on board 1, then the board corresponding to the address pool block A is the board 1, then the load sharing table can store the The corresponding relationship between the address pool block ID A of the address pool block A and the board ID 1 of the board 1 .

In the embodiment of the present invention, after the M address pool blocks are allocated to other boards in the load sharing group, the first load sharing table may be updated.

For example, in the first load sharing table, the ID corresponding to the first board is the first board ID, the ID corresponding to the second board is the second board ID, and the M address pool blocks It may correspond to the M address pool block identifiers in total.

Then, if all the M address pool blocks are allocated to the second board, the difference between the first board ID and the M address pool block IDs in the first load sharing table can be The first corresponding relationship among them is updated to the second corresponding relationship between the second board identifier and the M address pool block identifiers. That is to say, let the M address pool blocks correspond to the second board instead of corresponding to the first board.

For example, in the first load sharing table, the ID corresponding to the first board is the first board ID, the ID corresponding to the second board is the second board ID, and the third board The identifier corresponding to the board is the third board identifier, and the M address pool blocks may correspond to the M address pool block identifiers in total.

Then, if m address pool blocks in the M address pool blocks are allocated to the second board, and the remaining n address pool blocks in the M address pool blocks are allocated to the For the third single board, the first corresponding relationship between the first single board identifier and the M address pool block identifiers in the first load sharing table may be respectively updated as: the second single board identifier A second correspondence between the m address pool block identifiers, and a third correspondence between the third single board identifier and the n address pool blocks. That is to say, let the M address pool blocks correspond to the second single board and the third single board respectively, instead of corresponding to the first single board.

In this embodiment of the present invention, each board in a load sharing group may store the first load sharing table.

Step 103: Send the second load sharing table to each board in the load sharing group, so that a board in the load sharing group is the received The message is assigned a source public network address, and the message is forwarded according to the destination public network address carried in the message.

After the control plane updates the first load sharing table, it can obtain the second load sharing table, and the control plane can send the second load sharing table to each single board for each single board. The board updates the first load sharing table stored locally. Therefore, when updating, it is equivalent to updating the first load sharing table stored on each single board.

Specifically, after the control plane sends the second load sharing table to each board, any faultless board in the load sharing group may receive the message.

For example, if a board (for example, board 1) in the load sharing group receives the message from an electronic device, then the board 1 may first determine the first board corresponding to the message. The first address pool block identifier of an address pool block, after determining the first address pool block identifier, can determine the single board corresponding to the first address pool block identifier according to the second load sharing table Whether the logo is the board logo of the current board or the board logo of other boards. For example, it is determined that the board ID of the board corresponding to the first address pool block ID is 2, and the board corresponding to the board ID is, for example, board 2. If the board 1 and the board 2 are the same board board, that is, determine that the board ID of the board corresponding to the first address pool block ID is the board ID of this board, that is, determine that the first address pool block corresponding to the first address pool block ID is located in This single board, then the single board 1 can allocate the source public network address for the message according to the local first address pool block, after allocating the source public network address for the message, The single board 1 may forward the message according to the destination public network address carried in the message.

For example, if a board (for example, board 1) in the load sharing group receives the message from an electronic device, then the board 1 may first determine the first board corresponding to the message. The first address pool block identifier of an address pool block, after determining the first address pool block identifier, can determine the single board corresponding to the first address pool block identifier according to the second load sharing table Whether the ID is the board ID of the current board or the board ID of other boards. For example, it is determined that the board ID of the board corresponding to the first address pool block ID is 2, and the board corresponding to the board ID is, for example, board 2, if the board 1 is different from the board 2 A single board, that is, determine that the single board identification of the single board corresponding to the first address pool block identification is the single board identification of other single boards, that is, determine the first address pool block corresponding to the first address pool block identification located on other boards, then the board 1 can forward the message to the board 2.

After the message arrives at the single board 2, the single board 2 can allocate a public network IP address and a port number for the message according to the first address pool block, that is, allocate the required port number for the message. The source public network address.

In the embodiment of the present invention, a single board can assign the same public network IP address and the same port number to messages from the same electronic device, that is to say, an electronic device can be assigned a fixed public network IP address and a fixed The port number is to assign a fixed source public network address to an electronic device.

Because the message initially carries the source private network address of the electronic device, when the message arrives at the board 2, the board 2 can assign A public network IP address and a port number are assigned to the above-mentioned message, that is to say, a fixed public network IP address and a fixed port number can be assigned to a private network IP address. This can not only avoid information confusion during transmission, avoid information loss, but also save public network IP address resources as much as possible.

Specifically, in this embodiment of the present invention, when determining the first address pool block identifier of the first address pool block corresponding to the message, a first operation rule may be used for determination.

Specifically, in the embodiment of the present invention, the first operation rule may be: when the input information is the same, the output information is also the same.

Preferably, in the embodiment of the present invention, the first operation rule may be a hash (Hash) operation rule, or the first operation rule may also be other operation rules.

Specifically, in the embodiment of the present invention, if the first operation rule is a hash operation rule, then the first operation rule may specifically be: using the source private network address as the input value of the hash function, the The output value of the hash function is the block identifier of the first address pool. Among them, for the hash function, when the input information is the same, the output information is also the same, that is, when the input source private network addresses are the same, the output address pool block identifiers are also the same.

For example, if the first operation rule is a hash operation rule, then the source private network address carried in the message can be used as the input value of the hash function, and the output value of the hash function can be the corresponding Address pool block ID.

For example, in the embodiment of the present invention, after the board 1 determines the first address pool block ID, if it is determined that the board ID of the board corresponding to the first address pool block ID is 2, the board ID The corresponding single board is single board 2, and the single board 2 and the single board 1 are different single boards, then the single board 1 can forward the message to the The veneer 2. Wherein, the first address pool block may be the address pool block originally located in the first single board, or the first address pool block may not be the address pool block originally located in the first single board.

In the embodiment of the present invention, after forwarding the message to the veneer 2, the veneer 2 can allocate the source public network address for the message, and the veneer 2 can The destination public network address carried in the message forwards the message.

Further, in another embodiment of the present invention, after the second load sharing table is sent to each board in the load sharing group, if the first board recovers from failure, then the After a single board fails and recovers, the control plane may reallocate an address pool block for the first single board.

In this embodiment of the present invention, after the failure of the first single board is restored, the control plane may select at least one target single board from other single boards in the load sharing group except the first single board, and may use The selected part of the address pool block corresponding to each target board is allocated to the first board, and each target board sends the address pool block information corresponding to the address pool block allocated to the first board to the The first single board, updating the single board identification of the single board corresponding to the address pool block identifier of the address pool block allocated to the first single board in the second load sharing table to the first single board identification , obtain a third load sharing table, and issue the third load sharing table to each board in the load sharing group.

For example, if the address pool block in board 3 is selected to be allocated to the first board, then it must be ensured that there is at least one address pool block in board 3 after allocation, that is, the address pool block in board 3 is allocated to all The first single board refers to its own part of the address pool block.

Preferably, in this embodiment of the present invention, after allocating part of the address pool blocks corresponding to each selected target board to the first board, the addresses allocated to the first board in each target board can be deleted. The address pool block of the board. For example, if address pool block 1 and address pool block 2 of board B are allocated to the first board, then address pool block 1 and address pool block 2 in board B may be deleted . Because when a message is forwarded to the board B, the board B will allocate a source public network address for the message according to its own address pool block, and delete the address assigned to the first board. The address pool block 1 and the address pool block 2 can prevent the single board B from allocating the source public network address for the received message according to the address pool block 1 or the address pool block 2 again.

Optionally, after the first single board recovers from failure, when reassigning address pool blocks to the first single board, the M address pool blocks originally owned by the first single board may be reassigned to The first single board may reassign to the first single board the M address pool blocks allocated to other single boards in the load sharing table when the first single board fails.

Optionally, when the address pool block is reassigned to the first board after the fault recovery of the first board, other boards in the load sharing group except the first board may Randomly select target boards, and allocate part of address pool blocks in each selected target board to the first board.

For example, the load sharing group shares the first board, the second board, the third board and the fourth board, and when the first board fails, all the The M address pool blocks are all allocated to the second single board.

After the failure of the first single board is recovered, a target single board can be selected in the load sharing group, and the optional target single boards are the second single board, the third single board and the fourth single board plate. For example, if the second board has three address pool blocks, the third board has two address pool blocks, and the fourth board has one address pool block, then the fourth board cannot serve as The target single board, where the optional target single boards are the second single board and the third single board.

At this time, one address pool block from the second board can be randomly assigned to the first board, or two address pool blocks can be selected from the second board to be assigned to the first board. single board, or select an address pool block from the third single board to allocate to the first single board, or select an address pool block from the second single board to allocate to the first single board A single board, and an address pool block selected from the third single board is allocated to the first single board.

That is to say, when the first single board recovers from a fault, only one address pool block may be allocated to the first single board, or a second number of address pool blocks may be allocated to the first single board, and the The second number may be the number of address pool blocks that the first single board had before failure, or a third number of address pool blocks may be allocated to the first single board, and the third number may be greater than The number of address pool blocks that the first single board has before failure. The specific distribution method is not limited in the present invention.

Moreover, when the address pool block is allocated to the first single board after the fault recovery of the first single board, the address pool block may only be allocated to the first single board from the second single board, or may be allocated from other The first board may be allocated among the boards, or the first board may be allocated from the second board and other boards at the same time. The second single board is: a single board to which its own address pool block is allocated when the first single board fails. The specific distribution method is not limited in the present invention.

Preferably, in the embodiment of the present invention, the second load sharing table may be updated at the same time or after the address pool block is reassigned to the first single board. For example, the single board identifiers of the single boards corresponding to the address pool block identifiers of the address pool blocks reassigned to the first single board in the second load sharing table may be updated to the first single board identifier, That is, in the second load sharing table, the address pool block identifiers of the address pool blocks reassigned to the first single board are all corresponding to the first single board identifier, indicating that these address pool blocks are located in the first single board. In a single board, after updating the second load sharing table, a third load sharing table may be obtained.

For example, when redistributing, the address pool block A in the second single board in the load sharing group is allocated to the first single board, and the address pool block A in the third single board in the load sharing group is allocated to the first single board. The address pool block B in the board is allocated to the first single board. Then, the second load sharing table may be updated at the same time as or after the allocation. In the second load sharing table, the address pool block ID A of the address pool block A corresponds to the second board ID of the second board, and the address pool block ID of the address pool block B B corresponds to the third board identifier of the third board, then in the third load sharing table obtained after updating the second load sharing table, the address pool block of the address pool block A The identifier A corresponds to the first board identifier of the first board, and the address pool block identifier B of the address pool block B also corresponds to the first board identifier of the first board.

Preferably, in the embodiment of the present invention, after allocating part of the address pool blocks corresponding to each selected target board to the first board, the address pool blocks assigned to the first board can also be The corresponding address pool block information is sent to the first single board.

In the embodiment of the present invention, the address pool block information may at least include user table information and/or flow table information.

In the embodiment of the present invention, a single board can establish a user table information and at least one flow table information for an electronic device. For example, the board 1 allocates a source public network address to a message, then the board 1 can establish a user table information and at least one flow table information for the electronic device sending the message.

That is, for a single board, it may store multiple user table information and corresponding flow table information. In the embodiment of the present invention, the address pool block information may include at least one piece of user table information and/or corresponding flow table information.

For example, the user table information may store information corresponding to the corresponding electronic device, for example, the corresponding relationship between the corresponding source private network address and the corresponding source public network address may be stored, and a source private network address may correspond to on an electronic device.

For example, the flow table information may include not only information corresponding to a corresponding electronic device, but also flow information corresponding to each electronic device, and so on.

When each single board is working, it can work according to the corresponding address pool block information.

Then, in the embodiment of the present invention, after allocating part of the address pool blocks corresponding to each selected target board to the first board, the address pool blocks corresponding to the first board can also be assigned Address pool block information is sent to the first veneer, so as to ensure that the first veneer continues to work normally according to the corresponding address pool block information. For users, they basically do not feel any changes in the veneer. A single board failure will basically not affect the user's use, which can significantly improve the user experience.

For example, the selected target board is the second board, and address pool block A and address pool block B in the second board are allocated to the first board, then the address pool After the block A and the address pool block B are allocated to the first board, the address pool block information A corresponding to the address pool block A originally stored in the second board and the address The address pool block information B corresponding to the pool block B is sent to the first single board.

For example, the selected target boards are the second board and the third board, the address pool block A in the second board is allocated to the first board, and the address pool block A in the second board is assigned Address pool block B in the three boards is allocated to the first board. Then, after the address pool block A and the address pool block B are allocated to the first board, the address corresponding to the address pool block A originally stored in the second board can also be The pool block information A is sent to the first board, and the address pool block information B corresponding to the address pool block B originally stored in the third board is sent to the first board.

When the M address pool blocks are allocated to the second board, because the first board has failed, the M address pool block information corresponding to the M address pool blocks cannot be sent to the The second single board, therefore, after obtaining the M address pool blocks, the second single board may re-establish corresponding address pool block information for the M address pool blocks.

Preferably, in the embodiment of the present invention, after sending the address pool block information corresponding to the address pool block allocated to the first board to the first board, the control plane may send the first The single board rejoins the load sharing group corresponding to the third load sharing table, so that the first single board starts working again. Generally speaking, a load sharing table may correspond to a load sharing group, and the load sharing table may include the correspondence between each board ID and each address pool block ID in the load sharing group.

Preferably, in the embodiment of the present invention, after the address pool block information corresponding to the address pool block allocated to the first board is sent to the first board, the allocation in each target board can also be deleted. Address pool block information corresponding to each address pool block of the first single board.

For example, the selected target board is the second board, and address pool block A and address pool block B in the second board are allocated to the first board, then the address pool After the block A and the address pool block B are allocated to the first board, the address pool block information A corresponding to the address pool block A originally stored in the second board and the address The address pool block information B corresponding to the pool block B is sent to the first single board. After sending both the address pool block information A and the address pool block information B to the first board, the address pool block information A and the address pool may be deleted in the second board Block information B, to save storage space in the second single board.

For example, the selected target boards are the second board and the third board, the address pool block A in the second board is allocated to the first board, and the address pool block A in the second board is assigned Address pool block B in the three boards is allocated to the first board. Then, after the address pool block A and the address pool block B are allocated to the first board, the address corresponding to the address pool block A originally stored in the second board can also be The pool block information A is sent to the first board, and the address pool block information B corresponding to the address pool block B originally stored in the third board is sent to the first board. After sending both the address pool block information A and the address pool block information B to the first board, the address pool block information A may be deleted in the second board, and the The address pool block information B is deleted from the third board to save storage space on the second board and the third board.

Embodiment two

Referring to Fig. 2, the embodiment of the present invention provides a message forwarding method, the main process of the method is as follows:

Step 201: After obtaining the second load sharing table, obtain a message from an electronic device, where the message at least carries an active private network address and a destination public network address.

In this embodiment of the present invention, in a load sharing group, when a single board fails, for example, the faulty single board is the first single board, then the control plane can set the corresponding M address pool blocks are allocated to other boards in the load sharing group, and after the M address pool blocks are allocated to other boards in the load sharing group, the control plane may update the first In the load sharing table, for example, the board IDs of the boards corresponding to the M address pool block IDs of the M address pool blocks in the first load sharing table may be respectively updated to the assigned board IDs, and the first Two load sharing table.

After the first load sharing table is updated, the second load sharing table can be obtained. In addition, the control plane may issue the second load sharing table to each board in the load sharing group.

Each single board in the load sharing table can continue to receive messages after obtaining the second load sharing table.

For example, after board 1 in the load sharing group receives the message after obtaining the second load sharing table, the message may carry the source private information corresponding to the electronic device sending the message. network address, and the destination public network address corresponding to the electronic device to receive the message.

Step 202: Process the source private network address according to a first operation rule to obtain a corresponding address pool block identifier; wherein, the first operation rule is: when the input information is the same, the output information is also the same.

In the embodiment of the present invention, after receiving the message, the single board 1 may process the source private network address according to the first operation rule.

Preferably, in the embodiment of the present invention, the first operation rule may be a hash (Hash) operation rule, or the first operation rule may also be other operation rules.

Specifically, in the embodiment of the present invention, if the first operation rule is a hash operation rule, then the first operation rule may specifically be: using the source private network address as the input value of the hash function, the The output value of the hash function is the block identifier of the first address pool. Among them, for the hash function, when the input information is the same, the output information is also the same, that is, when the input source private network addresses are the same, the output address pool block identifiers are also the same.

In the embodiment of the present invention, the hash operation rule can be selected as the first operation rule, because for the hash operation rule, as long as the input values are the same, the output values will be the same. For example, if there are two messages, as long as the two messages come from an electronic device, the source private network addresses of the electronic devices corresponding to the two messages are the same, and the two source private network addresses are respectively passed through the hash After processing by the Greek operation rule, the two obtained output values will be the same, that is, the obtained address pool block identifiers will be the same. This can ensure that the message from an electronic device can always be forwarded to the same address pool block for processing, that is, it can ensure that an electronic device only needs to be assigned a public network IP address and a port number, saving the public network IP as much as possible address.

Therefore, in the embodiment of the present invention, when selecting the first operation rule, a selection condition may be to ensure that the first operation rule can obtain the same output value when the input value is the same, and all those that meet this condition The above-mentioned first operation rules are all within the protection scope of the present invention.

For example, in this embodiment of the present invention, after the board 1 determines the first address pool block ID, it can determine the unit number of the board corresponding to the first address pool block ID according to the second load sharing table. The board identifier is 2, and the board corresponding to the board identifier is board 2, then the board 1 may determine that the board corresponding to the message is the board 2.

Wherein, the first address pool block may be the address pool block originally located in the first single board, or the first address pool block may not be the address pool block originally located in the first single board.

Step 203: According to the second load sharing table, allocate a source public network address for the message through the address pool block corresponding to the address pool block identifier, and forward the message according to the destination public network address .

After the single board 1 determines the first address pool block identifier, it may determine according to the second load sharing table that the single board identifier of the single board corresponding to the first address pool block identifier is the single board of this single board The identification is still the single board identification of other single boards. If it is determined that the single board identification of the single board corresponding to the first address pool block identification is the single board identification of this single board, it is equivalent to determining that the first address pool block is located in this single board. board, then the single board 1 may allocate the source public network address for the message according to the first address pool block, and after allocating the source public network address, it may The destination public network address forwards the message.

After the single board 1 determines the first address pool block identifier, it may determine according to the second load sharing table that the single board identifier of the single board corresponding to the first address pool block identifier is the single board of this single board The ID is also the board ID of other boards. If it is determined that the board ID of the board corresponding to the first address pool block ID is the board ID of other boards, for example, the board corresponding to the first address pool block ID The veneer ID of the veneer is 2, and the veneer corresponding to the veneer identification is veneer 2, which means that it can be determined that the first address pool block is located on the veneer 2, then the veneer 1 can transmit the message forwarded to the veneer 2.

In the embodiment of the present invention, after forwarding the message to the veneer 2, the veneer 2 can allocate the source public network address for the message, and the veneer 2 can The destination public network address carried in the message forwards the message.

The following uses a specific embodiment to introduce the information forwarding method in the present invention.

Embodiment three

In the embodiment of the present invention, the load sharing group can support the insertion of 4 single boards, so the control plane evenly distributes the address pool into 4 address pool blocks, and numbers the 4 address pool blocks as 1, 2, 3, 4. The number of each address pool block may be referred to as the address pool block identifier of the address pool block. Moreover, in the embodiment of the present invention, the actually inserted four boards are also numbered, for example, the numbers of these four boards are respectively a, b, c, and d, and the number of each board can be referred to as the single board Board ID of the board.

After numbering the four address pool blocks and the four boards, a corresponding address pool block can be assigned to each board. In the embodiment of the present invention, since there are 4 address pool blocks and 4 single boards in total, one address pool block can be allocated to each single board. The first load sharing table can be constructed according to the distribution result, for example, the first load sharing table can be as shown in Table 1:

Table 1

Correspondence address pool block veneer Correspondence 1 1 a Correspondence 2 2 b Correspondence 3 3 c Correspondence 4 4 d

The correspondence 1 in Table 1 indicates the correspondence between the board ID a of the board a and the address pool block ID 1 of the address pool block 1, and the correspondence 2 in Table 1 indicates the board ID b of the board b and the address pool The corresponding relationship between the address pool block ID 2 of block 2, the corresponding relationship 3 in Table 1 indicates the corresponding relationship between the single board ID c of the single board c and the address pool block ID 3 of the address pool block 3, and the corresponding relationship 4 in Table 1 Indicates the correspondence between the board ID d of the board d and the address pool block ID 4 of the address pool block 4 .

That is, in the embodiment of the present invention, address pool block 1 is assigned to board a, address pool block 2 is assigned to board b, address pool block 3 is assigned to board c, and address pool block 4 is assigned Gave veneer d.

At the first moment, the single board d fails, and at this time, the address pool block 4 corresponding to the single board d can be allocated to other single boards. For example, in this embodiment of the present invention, the address pool block 4 may be allocated to the single board c.

After allocation, the first load sharing table may be updated. The corresponding relationship between the single board identifier d and the address pool block 4 identifier in the first load sharing table shown in Table 1 may be updated to the single board identifier c and the address pool block 4 identifier The corresponding relationship between. The second load sharing table obtained after the update can be shown in Table 2:

Table 2

Correspondence address pool block veneer Correspondence 1 1 a Correspondence 2 2 b Correspondence 3 3 c Correspondence 4 4 c

It can be seen from Table 2 that after the first load sharing table is updated, the address pool block 4 corresponds to the single board c and no longer corresponds to the single board d.

The control plane sends the second load sharing table to each board in the load sharing group, that is, to the board a, the board b, and the board c.

After the first load sharing table is updated, at the first moment, the single board a receives a message A, the message A comes from an electronic device A, and the message A carries the The source private network address of the electronic device A, and the destination public network address corresponding to the electronic device C to receive the message A, for example, the source private network address may be the private network IP address of the electronic device A, For example, the destination public network address may be the public network IP address of the electronic device C.

After receiving the message A, the board a may process the source private network address according to a first operation rule, for example, the first operation rule in the embodiment of the present invention may be a hash operation rule.

In the embodiment of the present invention, the source private network address may be an input value of a hash function, and after the hash operation is performed according to the source private network address, the output value of the hash function may be an address pool block identifier, For example, the address pool block identifier may be the address pool block identifier 2 shown in Table 2. Then it can continue to determine whether the board ID of the board corresponding to the address pool block ID 2 is the board ID of this board or the board ID of other boards. In the embodiment of the present invention, the address pool block ID 2 can be determined The board ID of the corresponding board is 2, and the board corresponding to the board ID is board b, so it can be determined that the message A needs to be forwarded to the board b.

The board a may forward the message A to the board b, and after receiving the message A, the board b may allocate an address for the message A according to the address pool block 2. A public network IP address and a port number are to allocate a source public network address for the packet A.

Wherein, if the electronic device A has not sent other messages before, and the message A is the message sent by the electronic device A for the first time, then a new source public network can be assigned to the message A address. If the electronic device A has sent other messages before, it is likely that the board b has allocated a source public network address for the electronic device A, and the address pool block information stored locally on the board b The corresponding relationship between the source private network address and the source public network address of the electronic device A may have been stored, then when the single board b receives the message A sent by the electronic device A, it can The previously allocated source public network address may be directly allocated to the packet A.

After the source public network address is allocated, the single board b may forward the message A according to the destination public network address carried in the message.

At a second moment after the first moment, the board a receives a message B, the message B also comes from the electronic device A, and the message B carries the electronic device A The source private network address of , and the destination public network address corresponding to the electronic device B to receive the message B.

After the single board a receives the message B, it may process the source private network address according to a hash operation rule.

In the embodiment of the present invention, the source private network address may be an input value of a hash function, and after the hash operation is performed according to the source private network address, the output value of the hash function may be an address pool block identifier, The address pool block identifier may still be the address pool block identifier 2 shown in Table 2. Then it can continue to determine whether the board ID of the board corresponding to the address pool block ID 2 is the board ID of this board or the board ID of other boards. In the embodiment of the present invention, the address pool block ID 2 can be determined The board ID of the corresponding board is 2, and the board corresponding to the board ID is board b, so it can be determined that the message B still needs to be forwarded to the board b.

The message B may be forwarded to the board b, and after receiving the message B, the board b may allocate a source public network address for the message B according to the address pool block 2 .

Wherein, because the electronic device A has sent the message A before, the source public network address has been allocated to the electronic device A, and the message B sent by the electronic device A is received this time, Similarly, the previously allocated source public network address may be directly allocated to the message B.

After the source public network address is allocated, the single board b may forward the message B according to the destination public network address.

At a third moment after the second moment, the single board c receives a message C, the message C is from an electronic device D, and the message C carries the source of the electronic device D The private network address, and the destination public network address corresponding to the electronic device E to receive the message C.

After receiving the message C, the single board c may process the source private network address according to a hash operation rule.

In the embodiment of the present invention, the source private network address may be an input value of a hash function, and after the hash operation is performed according to the source private network address, the output value of the hash function may be an address pool block identifier, For example, the address pool block identifier may be the address pool block identifier 4 shown in Table 2. Then it can continue to determine whether the board ID of the board corresponding to the address pool block ID 4 is the board ID of this board or the board ID of other boards. In the embodiment of the present invention, the address pool block ID 4 can be determined The board ID of the corresponding board is 3, and the board corresponding to the board ID is board c, so it can be determined that the board ID of the board corresponding to the address pool block ID 4 is the board ID of the board , that is, the single board c does not need to forward the message, but can directly process it on this single board. According to the second load sharing table, it can be determined that the message C needs to be processed locally on the board c, and the message C will not be forwarded to the board d where the address pool block 4 originally resides , which can avoid errors in information forwarding and reduce the information loss rate.

The single board c may allocate a source public network address for the message C according to the address pool block 4 .

Wherein, if the electronic device D has not sent other messages before, and the message C is the message sent by the electronic device D for the first time, then a new source public network can be assigned to the message C address. If the electronic device D has sent other messages before, it is likely that the source public network address has been allocated to the electronic device D, and the address pool block information stored locally on the board c may have been The corresponding relationship between the source private network address and the source public network address of the electronic device D is stored, and when the single board c receives the message C sent by the electronic device D, it can directly The source public network address that has been allocated before the packet C is allocated can be used.

After the source public network address is allocated, the single board c may forward the message C according to the destination public network address carried in the message.

At the fourth moment after the third moment, the fault of the single board d is recovered, then, after the fault of the single board d is recovered, the control plane can start from the single board in the load sharing group At least one target single board is selected from the board a, the single board b, and the single board c, and part of the address pool block corresponding to each selected target single board may be allocated to the first single board.

In the embodiment of the present invention, when the control plane selects target boards, it needs to ensure that each target board has at least one address pool block after allocating part of the address pool blocks to the board d. Then it can be seen that the single board a and the single board b have only one address pool block respectively, obviously neither can be used as the target single board.

Therefore, the target board selected in the embodiment of the present invention is the board c. Because it is necessary to ensure that the board c itself has at least one address pool block after allocating part of the address pool blocks to the board d, it is possible to assign an address pool block in the board c to the board d. board d.

In the embodiment of the present invention, the address pool block corresponding to the single board d before the fault occurs is the address pool block 4, and at this time, an address pool block in the single board c is allocated to the single board When d, the address pool block 4 originally corresponding to the board d may be reassigned to the board d, or the address pool block 3 may be allocated to the board d. For example, in the embodiment of the present invention, the address pool block 4 is reassigned to the single board d.

In the embodiment of the present invention, at the same time or after reassigning the address pool block 4 to the single board d, the second load sharing table as shown in Table 2 may also be updated to obtain a third load sharing table . The board ID 3 of the board corresponding to the address pool block ID 4 of the address pool block 4 in the second load sharing table may be updated to the board ID d, and the third load sharing table may continue as follows Table 1 shows.

After obtaining the third load sharing table, the control plane may issue the third load sharing table to each board in the load sharing group, that is, to the board a, the veneer b, the veneer c and the veneer d.

In this embodiment of the present invention, at the same time or after the control plane reassigns the address pool block 4 to the board d, the board c may be ordered to allocate the address pool block 4 stored in its local The corresponding address pool block information 4 is sent to the single board d.

In the embodiment of the present invention, after sending the address pool block information 4 to the board d, the board c may delete the address pool block information 4 locally.

In the embodiment of the present invention, after the board c sends the address pool block information 4 to the board d, the control plane may add the board d to the load sharing group again.

Embodiment four

Referring to FIG. 3 , an embodiment of the present invention provides a message forwarding device, and the device may include a first distribution unit 301 , a first update unit 302 and a first sending unit 303 . In the embodiment of the present invention, the packet forwarding device may be the control plane.

Preferably, the apparatus may further include a dividing unit 304 , a second allocating unit 305 , a third allocating unit 306 and a second updating unit 307 .

The first allocation unit 301 may be configured to allocate the M address pool blocks corresponding to the first single board that has failed to other single boards in the load sharing group.

Preferably, the M address pool blocks may be all address pool blocks corresponding to the first single board, which can better avoid waste of address pool block resources.

The first updating unit 302 may be configured to update the board IDs of the boards corresponding to the M address pool block IDs of the M address pool blocks in the first load sharing table to the assigned board IDs respectively, Obtain the second load sharing table.

In the embodiment of the present invention, what is stored in a load sharing table may be the corresponding relationship between the board identifiers in a load sharing group and the address pool block identifiers. For example, if in a load sharing group, the board corresponding to address pool block A is board 1, then the address pool block ID A and all Describe the correspondence between board IDs 1 of boards 1.

The first sending unit 303 may be configured to send the second load sharing table to each board in the load sharing group, so that a board in the load sharing group The distribution table allocates the source public network address for the received message, and forwards the message according to the destination public network address carried in the message.

If a single board in the load sharing group receives the message from an electronic device, it may first determine the first address pool block identifier of the first address pool block corresponding to the message.

Specifically, for a single board in the load sharing group, after receiving the message, it can process the source private network address carried in the message according to the first calculation rule to obtain the corresponding address pool block identifier; wherein, the first operation rule is: when the input information is the same, the output information is also the same. After obtaining the address pool block identifier, the single board corresponding to the address pool block identifier can be determined according to the second load sharing table, so that it can be determined whether the message should be left on this single board for processing or forwarded to other veneer handling.

In the embodiment of the present invention, the first operation rule may be a hash operation rule. Specifically, the first operation rule may be: using the source private network address as the input value of the hash function, the hash The output value of the function is the block identifier of the first address pool. Wherein, the source private network address may be a private network IP address of the electronic device.

In the embodiment of the present invention, the hash operation rule can be selected as the first operation rule, because for the hash operation rule, as long as the input values are the same, the output values will be the same. For example, if there are two messages, as long as the two messages come from an electronic device, the source private network addresses of the electronic devices corresponding to the two messages are the same, and the two source private network addresses are respectively passed through the hash After processing by the Greek operation rule, the two obtained output values will be the same, that is, the obtained address pool block identifiers will be the same. This can ensure that the message from an electronic device can always be forwarded to the same address pool block for processing, that is, it can ensure that an electronic device only needs to be assigned a public network IP address and a port number, saving the public network IP as much as possible address.

Therefore, in the embodiment of the present invention, when selecting the first operation rule, a selection condition may be to ensure that the first operation rule can obtain the same output value when the input value is the same, and all those that meet this condition The above-mentioned first operation rules are all within the protection scope of the present invention.

The division unit 304 may be configured to divide the address pool into at least one address pool block.

The second allocation unit 305 may be configured to allocate the at least one address pool block to each board in the load sharing group.

The third allocating unit 306 may be configured to select at least one target single board from other single boards in the load sharing group after the first single board fails and recovers, and allocate part of the address pool corresponding to each target single board to allocate blocks to the first single board, and make each target single board send address pool block information corresponding to the address pool block allocated to the first single board to the first single board.

Optionally, after the first single board recovers from failure, when reassigning address pool blocks to the first single board, the M address pool blocks originally owned by the first single board may be reassigned to The first single board may re-allocate to the first single board the M address pool blocks allocated to the second single board when the first single board fails.

Optionally, when the address pool block is reassigned to the first board after the fault recovery of the first board, other boards in the load sharing group except the first board may Randomly select target boards, and allocate part of address pool blocks in each selected target board to the first board.

The second updating unit 307 may be configured to update the board identifier of the board corresponding to the address pool block identifier of the address pool block allocated to the first board in the second load sharing table to the first The board ID is used to obtain the third load sharing table.

For example, when redistributing, the address pool block A in the second single board in the load sharing group is allocated to the first single board, and the address pool block A in the third single board in the load sharing group is allocated to the first single board. The address pool block B in the board is allocated to the first single board. Then, the load sharing table can be updated at the same time as or after the allocation. In the load sharing table before updating, the address pool block ID A of the address pool block A corresponds to the second board ID of the second board, and the address pool block B of the address pool block ID B corresponds to the third board ID of the third board, then after the load sharing table is updated, the address pool block ID A of the address pool block A corresponds to the ID of the first board. The first board identifier, the address pool block identifier B of the address pool block B also corresponds to the first board identifier of the first board.

The first sending unit 303 may also be configured to send the third load sharing table to each board in the load sharing group.

In the embodiment of the present invention, the address pool block information may at least include user table information and/or flow table information.

In this embodiment of the present invention, after each target board sends the address pool block information corresponding to the address pool block allocated to the first board to the first board, the address pool block information in each target board can also be deleted. Address pool block information corresponding to each address pool block allocated to the first single board.

In the embodiment of the present invention, after the first single board recovers from failure, the first single board may continue to be added to the load sharing group corresponding to the load sharing table, so that the first single board starts working.

Specifically, in this embodiment of the present invention, after the address pool block information corresponding to the address pool block allocated to the first board is sent to the first board, the first board may be added to the the load sharing group corresponding to the load sharing table, so that the first single board starts working again.

Embodiment five

Referring to FIG. 4 , an embodiment of the present invention provides a packet forwarding device, and the device may include an acquiring unit 401 , a processing unit 402 and a fourth distributing unit 403 . In the embodiment of the present invention, the packet forwarding device may be any single board in the load sharing group.

The obtaining unit 401 may be configured to obtain a message from an electronic device after obtaining the second load sharing table, and the message at least carries an active private network address and a destination public network address.

In this embodiment of the present invention, in a load sharing group, when a single board fails, for example, the faulty single board is the first single board, then the control plane can set the corresponding M address pool blocks are allocated to other boards in the load sharing group, and after the M address pool blocks are allocated to other boards in the load sharing group, the control plane may update the first In the load sharing table, for example, the board IDs of the boards corresponding to the M address pool block IDs of the M address pool blocks in the first load sharing table may be respectively updated to the assigned board IDs, and the first Two load sharing table.

After the first load sharing table is updated, the second load sharing table can be obtained. In addition, the control plane may issue the second load sharing table to each board in the load sharing group.

The processing unit 402 may be configured to process the source private network address according to a first operation rule to obtain a corresponding address pool block identifier; wherein, the first operation rule is: when the input information is the same, the output information is also same.

Preferably, in the embodiment of the present invention, the first operation rule may be a hash (Hash) operation rule, or the first operation rule may also be other operation rules.

The fourth allocating unit 403 may be configured to allocate a source public network address for the packet according to the address pool block corresponding to the address pool block identifier according to the second load sharing table, and allocate a source public network address according to the destination public network address forward the message.

The fourth allocating unit 403 may be specifically configured to determine, according to the second load sharing table, whether the board ID of the board corresponding to the address pool block ID is the board ID of this board or the board ID of other boards ; When the single board identification of the single board corresponding to the address pool block identification is the single board identification of this single board, assign the source public network address for the message through the address pool block corresponding to the address pool block identification , and forward the message according to the destination public network address; when the board ID of the board corresponding to the address pool block ID is the board ID of other boards, forward the message to other boards A single board corresponding to the single board identification, so that the single board allocates a source public network address for the message according to the address pool block corresponding to the address pool block identification, and sends the message according to the destination public network address The text is forwarded.

For example, after the single board 1 in the load sharing group determines the first address pool block identifier, it can determine the single board identifier of the single board corresponding to the first address pool block identifier according to the second load sharing table Whether it is the board ID of the current board or the board ID of other boards is to determine which board the first address pool block is located on. If it is determined that the board ID of the board corresponding to the first address pool block ID is the board ID of this board, it is equivalent to determining that the first address pool block is located on this board, then the board 1 The source public network address may be allocated to the message according to the first address pool block, and after the source public network address is allocated, the source public network address may be allocated according to the destination public network address carried in the message The message is forwarded.

For example, after the single board 1 in the load sharing group determines the first address pool block identifier, it can determine the unit of the single board corresponding to the first address pool block identifier according to the second load sharing table. Whether the board ID is the board ID of the current board or the board ID of other boards is to determine which board the first address pool block is located on. If it is determined that the board ID of the board corresponding to the first address pool block ID is the board ID of another board, for example, if it is determined that the board ID of the board corresponding to the first address pool block ID is 2, the board The veneer corresponding to the board identifier 2 is veneer 2, which is equivalent to determining that the first address pool block is located on the veneer 2, and the veneer 2 and the veneer 1 are different veneers, then the veneer Board 1 may forward the message to board 2.

In the embodiment of the present invention, after the message is allocated to the veneer 2, the veneer 2 may allocate the source public network address for the message, and the veneer 2 may allocate the source public network address according to the The destination public network address carried in the message forwards the message.

Embodiment six

Referring to FIG. 5 , an embodiment of the present invention provides a packet forwarding device, and the device may include a first processor 501 and a first sending interface 502 .

The first processor 501 may be configured to allocate the M address pool blocks corresponding to the first single board that has failed to other single boards in the load sharing group; the M address pool blocks in the first load sharing table The board IDs of the boards corresponding to the M address pool block IDs are updated to the assigned board IDs respectively, and the second load sharing table is obtained.

The first processor 501 may also be configured to divide the address pool into at least one address pool block; and allocate the at least one address pool block to each board in the load sharing group.

The first processor 501 may also be configured to select at least one target board from other boards in the load sharing group after the first board recovers from failure, and set the partial address corresponding to each target board to A pool block is allocated to the first veneer, and each target veneer is configured to send address pool block information corresponding to the address pool block allocated to the first veneer to the first veneer; the second updating unit It is used to update the single board identifier of the single board corresponding to the address pool block identifier of the address pool block allocated to the first single board in the second load sharing table to the first single board identifier, to obtain a third load contribution table.

The first processor 501 may also be configured to add the first single board to a load sharing group corresponding to the load sharing table, so that the first single board starts working.

The first sending interface 502 may be used to send the second load sharing table to each board in the load sharing group, so that a board in the load sharing group The distribution table allocates the source public network address for the received message, and forwards the message according to the destination public network address carried in the message.

The first sending interface 502 may also be configured to send the third load sharing table to each board in the load sharing group.

Embodiment seven

Referring to FIG. 6 , an embodiment of the present invention provides a packet forwarding device, and the device may include an acquisition interface 601 and a second processor 602 .

The obtaining interface 601 may be used to obtain a message from an electronic device after obtaining the second load sharing table, and the message at least carries an active private network address and a destination public network address.

The second processor 602 may be configured to process the source private network address according to a first operation rule to obtain a corresponding address pool block identifier; wherein, the first operation rule is: when the input information is the same, the output The information is also the same; according to the second load sharing table, assign the source public network address to the message through the address pool block corresponding to the address pool block identification, and process the message according to the destination public network address Forward.

The second processor 602 may be specifically configured to determine, according to the second load sharing table, whether the board ID of the board corresponding to the address pool block ID is the board ID of this board or the board ID of other boards. ; When the single board identification of the single board corresponding to the address pool block identification is the single board identification of this single board, assign the source public network address for the message through the address pool block corresponding to the address pool block identification , and forward the message according to the destination public network address; when the board ID of the board corresponding to the address pool block ID is the board ID of other boards, forward the message to the board The single board corresponding to the other single board identification, so that the single board allocates the source public network address for the message according to the address pool block corresponding to the address pool block identification, and allocates the source public network address according to the destination public network address The message is forwarded.

An embodiment of the present invention provides a message forwarding method, the method may include: allocating the M address pool blocks corresponding to the first single board that has failed to other single boards in the load sharing group; The single board identifications of the single boards corresponding to the M address pool block identifications of the M address pool blocks in the above are respectively updated to the assigned single board identifications to obtain the second load sharing table; sending to each board in the load sharing group, so that a board in the load sharing group assigns a source public network address to the received message according to the second load sharing table, and according to the The destination public network address carried in the message forwards the message.

In the embodiment of the present invention, when the first single board fails, the M address pool blocks corresponding to the first single board can be allocated to other single boards in the load sharing group, for example, can be allocated to other A single board, or may also be assigned to other multiple single boards, and the single board identifiers of the single boards corresponding to the M address pool block identifiers of the M address pool blocks in the load sharing table are correspondingly updated to be assigned board ID, so that when the message needs to be forwarded, the address pool block ID corresponding to the message can be obtained, and then according to the corresponding relationship between the address pool block ID and the board ID in the load sharing table The message can be allocated to the correct single board to forward the message, so that when a single board fails, the address pool block corresponding to the single board can be allocated to other single boards, which improves the The utilization rate of address pool blocks avoids the waste of address pool resources as much as possible. And, since the load sharing table is also updated correspondingly when the address pool block is allocated, subsequent message forwarding can be carried out according to the updated load sharing table, which can ensure that the message is correctly forwarded and avoid message loss as much as possible. Reduce information loss rate.

In the embodiment of the present invention, after sending the second load sharing table to each single board, the single board allocates the source public network address and port number for the message, and the control plane does not need to perform the public network IP address and port number. The allocation of port numbers improves the allocation efficiency and can meet the demand for high-speed processing of information without adding burden to the control plane.

Those skilled in the art can clearly understand that for the convenience and brevity of description, only the division of the above-mentioned functional modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional modules according to needs. The internal structure of the device is divided into different functional modules to complete all or part of the functions described above. For the specific working process of the above-described system, device, and unit, reference may be made to the corresponding process in the foregoing method embodiments, and details are not repeated here.

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

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

In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, each unit may exist separately physically, or two or more units may be integrated into one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

If the integrated unit is realized in the form of a software function unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or part of the contribution 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 to make a computer device (which may be a personal computer, a server, or a network device, etc.) or a processor (processor) execute all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disk or optical disk and other media that can store program codes. .

As mentioned above, the above embodiments are only used to introduce the technical solutions of the present application in detail, but the descriptions of the above embodiments are only used to help understand the method and core idea of the present invention, and should not be construed as limiting the present invention. Any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention shall fall within the protection scope of the present invention.

Claims (12)

1. a kind of message forwarding method, which is characterized in that the described method comprises the following steps：

The corresponding M address pool block of the first veneer to break down is distributed to other veneers in load balancing group；

M address pool block of M address pool block described in the first load balancing table is identified to the veneer sign phase of corresponding veneer It should be updated to assigned veneer sign respectively, obtain the second load balancing table；

The second load balancing table is handed down to each veneer in the load balancing group, so that in the load balancing group One veneer is identified according to the second load balancing table in the first address pool block of corresponding first address pool block of message received In the case that the veneer sign of corresponding veneer is the veneer sign of the veneer, the message to receive distributes source public network address, And be forwarded the message according to the purpose public network address that the message carries, in corresponding first address pool of the message In the case that the veneer sign of the corresponding veneer of the first address pool block mark of block is other veneer signs, the message is forwarded To other veneers, wherein, the veneer handles the source private net address that the message carries according to the first operation rule, obtains The first address pool block mark of corresponding first address pool block of the message is obtained, first operation rule is：When input information When identical, output information is also identical.

2. the method as described in claim 1, which is characterized in that by the corresponding M address pool block of the first veneer to break down Before distributing to other veneers in load balancing group, further include：

Address pool is divided at least one address pool block；

At least one address pool block is distributed into each veneer in the load balancing group.

3. method as claimed in claim 1 or 2, which is characterized in that described negative the second load balancing table is handed down to After carrying each veneer shared in group, further include：

After first single board default recovery, at least one target list is selected from other veneers of the load balancing group The corresponding partial address pond block of each target single board is distributed to first veneer by plate, and enables each target single board that will distribute First veneer is sent to the corresponding address pool block information of address pool block of first veneer；

The address pool block that the address pool block of first veneer will be distributed in the second load balancing table identifies corresponding list The veneer sign of plate is updated to first veneer sign, obtains third load balancing table；

The third load balancing table is handed down to each veneer in the load balancing group.

4. a kind of message forwarding method, applied to veneer, which is characterized in that the described method comprises the following steps：

After the second load balancing table is obtained, the message from an electronic equipment is obtained, active private is at least carried in the message Net address and purpose public network address；

The source private net address according to the first operation rule is handled, obtains corresponding first address pool block of the message First address pool block identifies；Wherein, first operation rule is：When input information is identical, output information is also identical；

According to the second load balancing table, identify and correspond in the first address pool block of corresponding first address pool block of the message Veneer veneer sign be the veneer sign of the veneer in the case of, pass through the corresponding address pool of described address pond block identification Block distributes source public network address for the message and is forwarded the message according to the purpose public network address, in the report The veneer sign that first address pool block of corresponding first address pool block of text identifies corresponding veneer is the feelings of other veneer signs Under condition, the message is forwarded to other veneers.

5. method as claimed in claim 4, which is characterized in that according to the second load balancing table, pass through described address pond The corresponding address pool block of block identification distributes source public network address and according to the purpose public network address by the message for the message It is forwarded, including：

The veneer sign that block identification corresponding veneer in described address pond is determined according to the second load balancing table is this veneer The veneer sign of veneer sign or other veneers；

When the veneer sign of the corresponding veneer of described address pond block identification is the veneer sign of this veneer, pass through described address pond The corresponding described address pond block of block identification distributes source public network address and according to the purpose public network address by described in for the message Message is forwarded；

When the veneer sign of the corresponding veneer of described address pond block identification is the veneer sign of other veneers, the message is turned Be sent to the corresponding veneer of other veneer signs, with by the veneer according to the corresponding described address of described address pond block identification Pond block distributes source public network address for the message and is forwarded the message according to the purpose public network address.

6. a kind of apparatus for forwarding message, which is characterized in that described device includes：

First allocation unit, the corresponding M address pool block of the first veneer for that will break down are distributed in load balancing group Other veneers；

First updating unit, for the M address pool block mark of M address pool block described in the first load balancing table is corresponding The veneer sign of veneer is accordingly updated to assigned veneer sign respectively, obtains the second load balancing table；

First transmitting element, for the second load balancing table to be handed down to each veneer in the load balancing group, so that A veneer in the load balancing group is according to the second load balancing table, in corresponding first address pool block of message received The first address pool block identify corresponding veneer veneer sign be the veneer sign of the veneer in the case of, for the report received Literary distribution source public network address and according to the message carry purpose public network address the message is forwarded, in the report The veneer sign that first address pool block of corresponding first address pool block of text identifies corresponding veneer is the feelings of other veneer signs Under condition, the message is forwarded to other veneers, wherein, the source private net address that the veneer carries the message is according to first Operation rule is handled, and obtains the first address pool block mark of corresponding first address pool block of the message, first fortune Calculating rule is：When input information is identical, output information is also identical.

7. device as claimed in claim 6, which is characterized in that described device further includes division unit and the second allocation unit；

The division unit is used to address pool being divided at least one address pool block；

Second allocation unit is used to distributing at least one address pool block into each veneer in the load balancing group.

8. device as claimed in claims 6 or 7, which is characterized in that described device further includes third allocation unit and second more New unit；

The third allocation unit is used for after first single board default recovery, from other veneers of the load balancing group At least one target single board is selected, the corresponding partial address pond block of each target single board is distributed into first veneer, and enable The corresponding address pool block information of the address pool block for distributing to first veneer is sent to first list by each target single board Plate；

Second updating unit is used for the address of address pool block for first veneer being distributed in the second load balancing table The veneer sign of the corresponding veneer of pond block identification is updated to first veneer sign, obtains third load balancing table；

First transmitting element is additionally operable to each veneer being handed down to the third load balancing table in the load balancing group.

9. a kind of apparatus for forwarding message, which is characterized in that described device includes：

Acquiring unit, for after the second load balancing table is obtained, obtaining the message from an electronic equipment, in the message extremely Active private net address and purpose public network address are carried less；

For the source private net address to be handled according to the first operation rule, it is corresponding to obtain the message for processing unit The first address pool block mark of first address pool block；Wherein, first operation rule is：When input information is identical, output Information is also identical；

4th allocation unit, for according to the second load balancing table, the of corresponding first address pool block of the message In the case that the veneer sign of the corresponding veneer of one address pool block mark is the veneer sign of the apparatus for forwarding message, pass through institute It is that the message distributes source public network address and incited somebody to action according to the purpose public network address to state address pool block and identify corresponding address pool block The message is forwarded, and the list of corresponding veneer is identified in the first address pool block of corresponding first address pool block of the message In the case that panel sign is other veneer signs, the message is forwarded to other veneers.

10. device as claimed in claim 9, which is characterized in that the 4th allocation unit is specifically used for：According to described second Load balancing table determine the veneer sign of the corresponding veneer of described address pond block identification be this veneer veneer sign or other The veneer sign of veneer；When the veneer sign of the corresponding veneer of described address pond block identification is the veneer sign of this veneer, lead to It crosses the corresponding described address pond block of described address pond block identification and distributes source public network address and according to purpose public affairs for the message The message is forwarded by net address；When the veneer sign of the corresponding veneer of described address pond block identification is the list of other veneers During panel sign, the message is forwarded to the corresponding veneer of other veneer signs, with by the veneer according to described address pond block It identifies corresponding described address pond block and distributes source public network address and according to the purpose public network address by the report for the message Text is forwarded.

11. a kind of apparatus for forwarding message, which is characterized in that described device includes：

First processor, the corresponding M address pool block of the first veneer for that will break down are distributed in load balancing group Other veneers；M address pool block of M address pool block described in the first load balancing table is identified to the veneer mark of corresponding veneer It is sensible to be updated to assigned veneer sign respectively, obtain the second load balancing table；

First transmission interface, for the second load balancing table to be handed down to each veneer in the load balancing group, so that A veneer in the load balancing group is according to the second load balancing table, in corresponding first address pool block of message received The first address pool block identify corresponding veneer veneer sign be the veneer sign of the veneer in the case of, for the report received Literary distribution source public network address and according to the message carry purpose public network address the message is forwarded, in the report The veneer sign that first address pool block of corresponding first address pool block of text identifies corresponding veneer is the feelings of other veneer signs Under condition, the message is forwarded to other veneers, wherein, the source private net address that the veneer carries the message is according to first Operation rule is handled, and obtains the first address pool block mark of corresponding first address pool block of the message, first fortune Calculating rule is：When input information is identical, output information is also identical.

12. a kind of apparatus for forwarding message, which is characterized in that described device includes：

Interface is obtained, for after the second load balancing table is obtained, obtaining the message from an electronic equipment, in the message extremely Active private net address and purpose public network address are carried less；

Second processor for the source private net address to be handled according to the first operation rule, obtains the message and corresponds to The first address pool block the first address pool block mark；Wherein, first operation rule is：It is defeated when input information is identical It is also identical to go out information；According to the second load balancing table, in the first address pool of corresponding first address pool block of the message In the case that the veneer sign of the corresponding veneer of block identification is the veneer sign of the apparatus for forwarding message, pass through described address pond The corresponding address pool block of block identification distributes source public network address and according to the purpose public network address by the message for the message It is forwarded, the veneer sign of corresponding veneer is identified in the first address pool block of corresponding first address pool block of the message is In the case of other veneer signs, the message is forwarded to other veneers.