JP2011182272A - Multicast relay system, multicast relay apparatus and method for recovering relay control information of multicast relay apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a technology for promptly recovering relay control information of a multicast relay apparatus. <P>SOLUTION: A multicast relay system is equipped with: a first multicast relay apparatus; and a second multicast relay apparatus connected to the first multicast relay apparatus. The first multicast relay apparatus is equipped with: a relay processing section for performing relay processing of a multicast packet based on relay control information; a transmission section for transmitting duplicate creation information to the second multicast relay apparatus; a distribution request section for requesting the distribution of a duplicate of the relay control information to the second multicast relay apparatus; and a setting section for setting relay control information based on the duplicate of the relay control information. The second multicast relay apparatus is equipped with: a duplicate creation section for creating a duplicate of relay control information based on the duplicate creation information; and a distribution section for distributing the duplicate of the relay control information to the first multicast relay apparatus. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a multicast relay system, a multicast relay device, and a method for restoring relay control information of the multicast relay device.

  Multicast is a router that accommodates multicast recipients on the Internet. The router copies and outputs the packets to multiple lines to which the recipients are connected. Technology to send to. In multicast, it is possible to send a packet to a plurality of recipients with a smaller amount of packets than when a packet sender sends a packet to each recipient individually. Therefore, multicast is particularly suitable for real-time multimedia communication that requires large-capacity traffic represented by streaming and video conferencing. As typical multicast routing protocols, PIM-SM (Protocol Independent Multicast Sparse-Mode: see Non-Patent Document 1) and PIM-SSM (Protocol Independent Multicast-Source Specific Multicast: see Non-Patent Document 1) are known. Yes.

  In general, a multicast process is performed for each interface to which an IP address is assigned. If the interface to which the IP address is assigned is a VLAN (Virtual LAN) that accommodates multiple lines, the packet is copied to all the lines belonging to the VLAN and relay processing is performed, so no reception request is made Packets are also relayed to the line. That is, if even one of the plurality of lines belonging to the VLAN makes a packet reception request, the packet is copied and relayed to all other lines belonging to the VLAN. At this time, the packet relayed to the line for which no reception request has been made is an unnecessary packet, and the communication band is wasted.

  As a technique for efficiently relaying multicast packets, an IGMP / MLD snooping technique (IGMP / MLD snooping: see Non-Patent Document 2) is known. In IGMP / MLD snooping, the contents of a reception request received from a line belonging to a VLAN are snooped (analyzed), and the received line information is acquired. By controlling the relay of the multicast packet based on the acquired line information and the multicast group address, the multicast packet is relayed only to the line for which the reception request is made. As a result, multicast communication to a line other than the line that has requested reception is suppressed, and multicast relay can be performed efficiently.

  In IGMP / MLD snooping, relay control information is created based on a reception request (IGMP / MLD subscription request message) from a multicast receiver. Generally, an IGMP / MLD inquiry message transmission device (IGMP / MLD querier) set in the network periodically transmits an IGMP / MLD inquiry message, and a multicast receiver connected to each VLAN receives this IGMP. An IGMP / MLD subscription request message is responded to the / MLD inquiry message. Each multicast receiver responds to the IGMP / MLD subscription request message, so that the subscription status of the multicast receiver connected to each VLAN can be grasped. The IGMP / MLD inquiry message can be set to an arbitrary cycle time for each network, but is generally transmitted at a cycle of 125 seconds.

  When a failure occurs in a multicast relay apparatus that implements IGMP / MLD snooping and is restored by restarting or the like, all relay control information of IGMP / MLD snooping is also initialized. In the state in which the relay control information immediately after the recovery is initialized, the state is the same as the state in which no multicast receiver exists, and thus the received multicast packet is not relayed anywhere. At this time, even if the multicast receiver exists at the time of failure recovery, the presence of the multicast receiver cannot be grasped, and the relay is stopped. After recovery from the failure, an IGMP / MLD inquiry message is sent from the IGMP / MLD querier, and the multicast receiver recognizes the multicast receiver by responding to the IGMP / MLD join request message. Communication resumes.

  As described above, since an IGMP / MLD inquiry message is transmitted from the IGMP / MLD querier at a cycle of 125 seconds, multicast communication may be stopped for a maximum of 125 seconds even after recovery from a failure. In addition, when an IGMP / MLD inquiry message is flooded in the VLAN and receives an IGMP / MLD subscription request message from all multicast receivers connected to each line, a large number of multicast receivers are connected. However, the IGMP / MLD subscription request messages are concentrated and the communication band is compressed. In addition, since a large number of IGMP / MLD subscription request messages are received and relay control information creation processing is performed, it takes more time to create all relay control information and resume relaying of all multicast receivers. Can be considered.

  When a device having an IGMP snooping function has an IGMP querier function (see Non-Patent Document 3) as a method for quickly creating relay control information at the time of failure recovery, an IGMP / MLD inquiry message simultaneously with the initial activation of the IGMP snooping function The method of transmitting is mentioned. Further, as a similar technique, when IGMP / MLD snooping relay control information is erased due to a failure of another protocol, a technique is known in which relay control information is reproduced by immediately transmitting an IGMP / MLD inquiry message. (See Non-Patent Document 1).

  In the case of a method of recovering communication by distributing the IGMP / MLD inquiry message at an early stage, to transmit the IGMP / MLD inquiry message from the own apparatus without waiting for the transmission of the IGMP / MLD inquiry message from the IGMP / MLD querier. The IGMP / MLD subscription request message can be received relatively quickly. However, IGMP / MLD subscription request messages are typically responded at intervals to avoid centralized reception (maximum response time of 10 seconds, response time determined randomly). Therefore, there is a possibility that the IGMP / MLD subscription request message is replied after a maximum of 10 seconds, and the multicast communication may be stopped for a maximum of 10 seconds. Furthermore, since the IGMP / MLD inquiry message is flooded in the VLAN, there is a possibility that a huge IGMP / MLD subscription request message will be replied, and a reception process of a huge IGMP / MLD subscription request message may occur. . Therefore, the delay of the relay control information creation process and the compression of the communication band due to centralized reception are not improved.

  As described above, in a multicast relay device performing multicast communication using IGMP / MLD snooping, when a failure occurs in the multicast relay device and the multicast relay device is recovered, multicast communication stops for a maximum of 125 seconds. Furthermore, even if an IGMP / MLD inquiry message is transmitted from the IGMP / MLD querier, if there are a large number of multicast receivers, it takes time to receive the IGMP / MLD subscription request message and generate the relay control information. Therefore, restart of multicast communication is delayed. Further, it is not possible to neglect the compression of the communication band caused by centralized reception of IGMP / MLD subscription request messages in response to IGMP / MLD inquiry messages.

  As described above, even if the IGMP / MLD inquiry message is transmitted early at the time of recovery, the multicast communication is resumed relatively quickly, but the centralized reception of the IGMP / MLD subscription request message is not eliminated, and the rapid relay control is performed. It is not a reproduction of information. In addition, compression of the communication band due to centralized reception of IGMP / MLD subscription request messages is not improved.

JP 2007-324688 A

IETF, “Protocol Independent Multicast-Sparse Mode (PIM-SM): Protocol Specification (Revised)”, RFC4601 IETF, “Considerations for Internet Group Management Protocol (IGMP) and Multicast Listener Discovery (MLD) Snooping Switches”, RFC4541 AX3600S Software Manual Ver. 11.1 Configuration Guide Vol.1 23.3.4 IGMP Querier Function

  The present invention has been made to solve the above-described conventional problems, and an object of the present invention is to provide a technique capable of quickly recovering relay control information of a multicast relay apparatus.

  In order to solve at least a part of the problems described above, the present invention can take the following forms or application examples.

[Application Example 1]
A first multicast relay device that relays a multicast packet to a line connected to a recipient of the multicast packet in a logically independent network, and a second multicast relay device that is connected to the first multicast relay device A multicast relay system comprising:
The first multicast relay device is:
A relay processing unit that performs relay processing of the multicast packet based on relay control information including information of a line to which a recipient of the multicast packet is connected;
A transmission unit that transmits information for creating a copy, which is information used to create a copy of the relay control information, to the second multicast relay device;
A distribution request unit that requests distribution of a copy of the relay control information to the second multicast relay apparatus at a predetermined timing;
A setting unit configured to set the relay control information based on a copy of the relay control information distributed from the second multicast relay device,
The second multicast relay device is
A copy creation unit for creating a copy of the relay control information based on the copy creation information transmitted from the first multicast relay device;
A multicast relay system comprising: a distribution unit that distributes a copy of the relay control information to the first multicast relay device in response to a request from the first multicast relay device.

  According to the application example 1, the setting unit of the first multicast relay device sets the relay control information based on the copy of the relay control information distributed from the second multicast relay device. The relay control information can be quickly restored without flooding. Moreover, since the inquiry message is flooded in the VLAN and the response of the subscription request message from all the multicast receivers is not received, the communication band is compressed and relayed by receiving a large number of subscription request messages in a concentrated manner. Control information entry creation processing delay can also be reduced.

[Application Example 2]
The multicast relay system according to application example 1, further comprising:
A plurality of the first multicast relay devices;
The second multicast relay device is connected to the plurality of multicast relay devices;
The copy creating unit of the second multicast relay device is configured to use the first multicast relay device based on the copy creation information transmitted from each of the plurality of connected first multicast relay devices. A copy of the relay control information for
The distribution unit of the second multicast relay apparatus distributes a copy of the relay control information to the first multicast relay apparatus in response to a request from the first multicast relay apparatus. Multicast relay system.

  According to Application Example 2, since the second multicast relay device creates a copy of the relay control information of the plurality of first multicast relay devices, the relay control information of the plurality of first multicast relay devices can be quickly recovered. Can be made.

[Application Example 3]
The multicast relay system according to Application Example 1 or 2, further comprising:
A third multicast relay device connected to the second multicast relay device;
The second multicast relay device further includes:
A relay processing unit that performs relay processing of the multicast packet based on second relay control information including information of a line to which a receiver of the multicast packet is connected;
A transmitter for transmitting second copy creation information, which is information used for creating a copy of the second relay control information, to the third multicast relay device;
A distribution requesting unit that requests distribution of a copy of the second relay control information to the third multicast relay device at a predetermined timing;
A setting unit configured to set the second relay control information based on a copy of the second relay control information distributed from the third multicast relay device;
The third multicast relay device is
A copy creation unit for creating a copy of the second relay control information based on the second copy creation information transmitted from the second multicast relay device;
A multicast relay system comprising: a distribution unit that distributes a copy of the second relay control information to the second multicast relay device in response to a request from the second multicast relay device.

  According to the third application example, the relay control information of the first multicast relay device can be quickly restored by duplicating the relay control information created by the replica creation unit of the second multicast relay device, and the third multicast The relay control information of the second multicast relay device can be quickly restored by duplicating the relay control information created by the replica creation unit of the relay device.

[Application Example 4]
The multicast relay system according to any one of Application Examples 1 to 3,
The multicast relay system, wherein the transmission unit transmits the copy creation information in addition to a join request message or a withdrawal request message transmitted from the multicast receiver.

  According to the application example 4, it is possible to create a copy of the relay control information without increasing the number of packets for transmitting the copy creation information. Further, since the relay control information is updated by the join request message or the withdrawal request message, the update of the copy of the relay control information can be synchronized with the update of the relay control information.

[Application Example 5]
The multicast relay system according to any one of Application Examples 1 to 4,
The distribution request unit transmits an inquiry message to the second multicast relay device as the distribution request,
In response to the inquiry message, the distribution unit adds a copy of the relay control information to a subscription request message and distributes it.

  According to the application example 5, since the packet in the existing format is used, a part of the processing procedure of the existing packet can be used.

[Application Example 6]
The multicast relay device according to any one of Application Examples 1 to 5,
A multicast relay system, wherein the relay control information distributed by the distribution unit includes a plurality of pieces of line information.

  According to the application example 6, even when a plurality of pieces of line information are included in the relay control information, the relay control information can be quickly recovered by transmitting a small number of packets.

[Application Example 7]
A first multicast relay device that relays a multicast packet to a line connected to a recipient of the multicast packet in a logically independent network, and a second multicast relay device that is connected to the first multicast relay device A multicast relay system comprising:
The first multicast relay device is:
A relay processing unit that performs relay processing of the multicast packet based on first relay control information including information of a line to which a receiver of the multicast packet is connected;
A transmission unit that transmits information for creating a copy, which is information used to create a copy of the first relay control information, to the second multicast relay device;
The second multicast relay device is
A copy creation unit for creating a copy of the first relay control information based on the copy creation information transmitted from the first multicast relay device;
An operational state detector that detects that the second multicast relay device has transitioned from a standby state to an operational state;
When the operation state detection unit detects that the second multicast relay device has transitioned from the standby state to the operation state, the second multicast relay device in the second multicast relay device is based on the copy of the first relay control information. A setting unit for setting the relay control information of
A multicast processing system, comprising: a relay processing unit that performs relay processing of the multicast packet based on the second relay control information.

  According to Application Example 7, when the second multicast relay device transitions from the standby state to the operational state, the setting unit of the second multicast relay device performs the second operation based on the copy of the first relay control information. Therefore, the second multicast relay device can take over the relay control information of the first multicast relay device and quickly restore the multicast relay.

[Application Example 8]
Information provided in a multicast relay device for relaying a multicast packet to a line to which a multicast packet receiver is connected in a logically independent network, the information about the line to which the multicast packet receiver is connected. A method for restoring relay control information including:
The multicast relay device transmitting copy creation information, which is information used to create a copy of the relay control information, to an external multicast relay device;
The external multicast relay device creating a copy of the relay control information based on the copy creation information transmitted from the multicast relay device;
The multicast relay device requesting the external multicast relay device to distribute the copy of the relay control information at a predetermined timing;
The external multicast relay device delivering a copy of the relay control information to the multicast relay device in response to a request from the multicast relay device;
The multicast relay device includes a step of setting the relay control information based on a copy of the relay control information distributed from the external multicast relay device.

  According to the application example 8, the setting unit of the first multicast relay apparatus sets the relay control information based on the copy of the relay control information distributed from the second multicast relay apparatus. The relay control information can be quickly restored without flooding.

[Application Example 9]
A multicast relay device that relays a multicast packet to a line to which a receiver of the multicast packet is connected in a logically independent network,
A relay processing unit that performs relay processing of the multicast packet based on relay control information including information of a line to which a recipient of the multicast packet is connected;
A transmission unit for transmitting information for creating a copy, which is information used for creating a copy of the relay control information, to an external multicast relay device connected to the multicast relay device;
A distribution request unit that requests distribution of a copy of the relay control information at a predetermined timing to the external multicast relay device;
A multicast relay device comprising: a setting unit configured to set the relay control information based on a copy of the relay control information distributed from the external multicast relay device.

  According to the application example 9, since the relay control information is set based on the copy of the relay control information distributed from the external multicast relay device, the relay control information can be quickly recovered.

[Application Example 10]
A multicast relay device connected to an external multicast relay device that relays a multicast packet to a line connected to a receiver of the multicast packet in a logically independent network,
A copy creation unit for creating a copy of the relay control information based on the copy creation information transmitted from the external multicast relay device;
A multicast relay apparatus comprising: a distribution unit that distributes a copy of the relay control information to the external multicast relay apparatus in response to a request from the external multicast relay apparatus.

  According to Application Example 10, a copy of the relay control information is distributed to the external multicast relay device in response to a request from the external multicast relay device, so that the relay control information of the external multicast relay device can be quickly restored. Can do.

  Note that the present invention can be realized in various modes. For example, the present invention can be realized in the form of a method and apparatus for restoring relay control information, an integrated circuit for realizing the functions of the method or apparatus, a computer program, a recording medium on which the computer program is recorded, and the like.

It is explanatory drawing which shows the structure of the network system as one Example of this invention. It is explanatory drawing which shows the process for recovering the multicast relay control table before failure occurrence, when the multicast relay apparatus 1000 is recovered from failure. 2 is an explanatory diagram showing an internal configuration of a multicast relay apparatus 1000. FIG. It is explanatory drawing which shows the content of the various information set in the multicast relay apparatus 1000, and a table. 2 is an explanatory diagram showing an internal configuration of a multicast router 2000. FIG. 6 is an explanatory diagram showing various information set in a multicast router 2000 and the contents of a table. FIG. It is a flowchart which shows a process in case the multicast relay apparatus 1000 receives an IGMP joining request message. It is a flowchart which shows a process in case the multicast relay apparatus 1000 receives an IGMP leaving request message. FIG. 10 is an explanatory diagram showing a processing flow when the multicast relay apparatus 1000 receives an IGMP join request message from a multicast receiver 40. It is explanatory drawing which shows a mode that the entry management table T1131 is updated. 6 is an explanatory diagram showing a flow of processing when a multicast relay apparatus 1000 receives an IGMP join request message from a multicast receiver 30. FIG. FIG. 10 is an explanatory diagram showing a processing flow when the multicast relay apparatus 1000 receives an IGMP leave request message from a multicast receiver 30. 6 is an explanatory diagram showing a flow of processing when the multicast relay apparatus 1000 receives an IGMP leave request message from a multicast receiver 40. FIG. 10 is a flowchart showing control packet relay processing in the multicast relay apparatus 1000. It is explanatory drawing which shows the format of the IGMP joining request / leaving request message to which additional information is added. 6 is an explanatory diagram showing a flow of relay processing of an IGMP join request message in the multicast relay device 1000. FIG. 6 is an explanatory diagram showing a flow of relay processing of an IGMP leave request message in the multicast relay device 1000. FIG. 6 is an explanatory diagram showing a flow of relay processing of an IGMP inquiry message in the multicast relay device 1000. FIG. 20 is a flowchart showing an update process of a replication entry table T2430 when the multicast router 2000 receives an IGMP join request message. 18 is a flowchart showing an update process of a replication entry table T2430 when the multicast router 2000 receives an IGMP leave request message. It is explanatory drawing which shows the flow of the update process of the replication entry table | surface T2430 when the multicast router 2000 receives an IGMP joining request message. FIG. 6 is an explanatory diagram showing a state of multicast communication before a failure occurs in the multicast relay apparatus 1000 and an explanatory diagram showing a state of the entry management table T1131 before the failure occurs in the multicast relay apparatus 1000. It is explanatory drawing which shows the condition of the multicast communication immediately after the multicast relay apparatus 1000 is recovered | restored from a failure, and explanatory drawing which shows the state of the entry management table T1131 immediately after the multicast relay apparatus 1000 is recovered from a failure. 10 is a flowchart showing processing for requesting distribution of a replication entry when the multicast relay apparatus 1000 recovers from a failure. 10 is a flowchart showing a replication entry distribution process in the multicast router 2000. 10 is a flowchart showing a process of receiving an IGMP join request message distributed from a multicast router 2000 and recreating the entry management table T1131 before the occurrence of a failure by the multicast relay apparatus 1000. FIG. 10 is an explanatory diagram showing a flow of processing for requesting distribution of a duplicate entry when the multicast relay apparatus 1000 recovers from a failure. FIG. 10 is an explanatory diagram showing a flow of replication entry distribution processing in the multicast router 2000; FIG. 10 is an explanatory diagram showing a flow of processing in which the multicast relay apparatus 1000 receives an IGMP join request message distributed from the multicast router 2000 and recreates the entry management table T1131 before the failure occurs. It is explanatory drawing which shows the mode after the reproduction of an entry. It is explanatory drawing which shows the restoration method of the entry management table of the multicast relay apparatus 1000 in a comparative example. It is explanatory drawing which shows the network structure in 2nd Example. It is explanatory drawing which shows the various information of the multicast relay apparatus 1000, and the content of VLAN management table T1111. FIG. 11 is an explanatory diagram showing various information of a multicast relay device 3000 and the contents of a VLAN management table T3111. 6 is an explanatory diagram showing various information of a multicast router 2000 and the contents of various tables. FIG. A process of generating a duplicate entry table T2430 when the multicast receiver 20 belonging to the VLAN 101 transmits an IGMP join request message for the multicast group G1 and the multicast receiver 50 belonging to the VLAN 102 transmits an IGMP join request message for the multicast group G2 is shown. It is explanatory drawing and duplicate drawing table T2430 updated based on the IGMP joining request message transmitted from the multicast receivers 20 and 50. It is explanatory drawing which shows the network structure in 3rd Example. It is explanatory drawing which shows the internal structure of the multicast relay apparatus 1000 in 3rd Example. FIG. 11 is an explanatory diagram showing various information of a multicast relay device 3000 and the contents of a VLAN management table T3111. It is explanatory drawing which shows the content of the various information of a multicast relay apparatus 1000, and various tables. It is explanatory drawing which shows the content of the replication entry production | generation information D2120 of the multicast router 2000, and the replication entry management table T2410. It is explanatory drawing which shows the flow of the control packet in the network structure of 3rd Example. It is explanatory drawing which shows the IGMP subscription request message to which additional information was added. It is explanatory drawing which shows a replication entry table. FIG. 10 is an explanatory diagram showing a processing flow when the multicast relay apparatus 1000 receives an IGMP join request message to which additional information is added by the multicast relay apparatus 3000. It is explanatory drawing which shows the network structure which employ | adopted the redundant structure in 4th Example. It is explanatory drawing which shows a state when a failure generate | occur | produces in the multicast relay apparatus 4000A of an operation state. It is explanatory drawing which shows a mode that the replication entry table of the multicast relay apparatus 4000A of an operation state is produced | generated in the multicast relay apparatus 4000B of a standby state. It is explanatory drawing which shows the internal structure of multicast relay apparatus 4000A. It is explanatory drawing which shows the content of the various information in the multicast relay apparatus 4000A, and various tables. It is explanatory drawing which shows the content of the various information in the multicast relay apparatus 4000B, and various tables. It is explanatory drawing which shows the flow of a process when the multicast relay apparatus 4000A which is an operation state receives an IGMP joining request message. It is explanatory drawing which shows the flow of a process when the multicast relay apparatus 4000B in a standby state receives the IGMP joining request message to which additional information is added. It is explanatory drawing which shows the flow of a process when a failure generate | occur | produces in the multicast relay apparatus 4000A, and the multicast relay apparatus 4000B will be in an operation state. 18 is a flowchart showing a processing flow of an operation state determination unit 4500 when the multicast relay device 4000B transitions from a standby state to an operation state. It is a flowchart which shows the flow of a process after multicast relay apparatus 4000B receives an operation state transition notification from an operation state determination part until it transmits an IGMP joining request message.

Next, embodiments of the present invention will be described in the following order based on examples.
A. First embodiment:
A1. Network system overview:
A2. Configuration of each device:
A3. Update processing of entry management table T1131 in multicast relay apparatus 1000:
A4. Relay process of IGMP control packet in multicast relay apparatus 1000:
A5. Updating the replication entry table T2430 in the multicast router 2000:
A6. State before failure:
A7. Status immediately after recovery from failure:
A8. Recovery method:
A9. How to restore the entry management table in the comparative example:
B. Second embodiment:
C. Third embodiment:
D. Fourth embodiment:
E. Variations:

A. First embodiment:
A1. Network system overview:
FIG. 1 is an explanatory diagram showing the configuration of a network system as an embodiment of the present invention. In FIG. 1, a general multicast communication and IGMP snooping function will be described. The network system according to the present embodiment includes a multicast relay device 1000 that implements an IGMP snooping function, and a multicast router 2000 that accommodates the multicast sender 10. The IGMP snooping function is a function that determines a multicast receiver and performs multicast communication only to the multicast receiver that has made a reception request. For this reason, the IGMP snooping function is generally a function used in the multicast relay apparatus 1000 to which a plurality of multicast receivers are connected.

  A multicast sender 10 is connected to the multicast router 2000 via a port 2001. The multicast sender 10 distributes multicast packets to the multicast router 2000 (arrow F100). The multicast router 2000 can perform multicast communication between the VLAN 100 that accommodates the multicast sender 10 and the VLAN 101 that accommodates the multicast receivers 20, 30, and 40 by using the multicast protocol.

  In general, a multicast router often serves as an IGMP querier. Therefore, the IGMP querier in this embodiment is assumed to be a multicast router 2000. In general, in the IGMP snooping function, a port connected to a multicast router is defined as a multicast router port. Therefore, the multicast router port in this embodiment is assumed to be port 1001.

  The IGMP snooping function in the multicast relay apparatus 1000 is transmitted from the multicast receivers 20, 30, 40 to notify the upstream multicast router 2000 of the participation status of the downstream multicast receivers 20, 30, 40 in the multicast group. Relay request (IGMP join request message) and leave request (IGMP leave request message) to the multicast router port 1001.

  The multicast router 2000 receives the IGMP join request / leave request message transmitted from the multicast receivers 20, 30, and 40, thereby recognizing the downstream multicast receivers 20, 30, and 40, and performs multicast communication from the upstream. Can be controlled. Specifically, for example, when an IGMP join request message is not transmitted from the multicast receivers 20, 30, and 40, even when the multicast sender 10 continues to distribute multicast packets (arrow F100). Since there is no recipient of the multicast packet, the multicast packet is not relayed anywhere.

  When the multicast receiver 40 wants to receive a multicast packet distributed from the multicast sender 10, the multicast receiver 40 transmits an IGMP join request message including a multicast group address desired to join (arrow F200). The multicast relay apparatus 1000 that has received the IGMP join request message analyzes the IGMP join request message using the IGMP snooping function, and acquires the multicast group address and reception port information that is information regarding the port that has received the IGMP join request message. . The multicast relay apparatus 1000 sets a multicast packet relay control table that determines a relay destination of the multicast packet based on the acquired multicast group address and reception port information. This multicast packet relay control table is stored in a memory (not shown). The same applies to the other tables described below.

  The multicast relay apparatus 1000 relays the received IGMP join request message to the multicast router 2000 via the multicast router port 1001 (arrow F201). When receiving the relayed IGMP join request message, the multicast router 2000 performs multicast relay between the VLAN 100 and the VLAN 101. That is, the multicast router 2000 relays the multicast packet transmitted from the multicast sender 10 of the VLAN 100 to the VLAN 101. The multicast packet relayed to the VLAN 101 is relayed only to the multicast receiver 40 based on the multicast packet relay control table set in the multicast relay device 1000 (arrow F101).

  In this embodiment, since the multicast router 2000 is an IGMP querier, the multicast router 2000 generally transmits an IGMP inquiry message at a period of 125 seconds. When receiving the IGMP inquiry message, the multicast relay apparatus 1000 floods the IGMP inquiry message in the VLAN.

  Each multicast receiver, when receiving the IGMP inquiry message, transmits an IGMP join request message to the subscribed multicast group. The multicast router 2000, which is an IGMP querier, periodically delivers an IGMP inquiry message and each multicast receiver responds, so that the multicast router 2000 can monitor the group participation status of the multicast receiver.

  FIG. 2 is an explanatory diagram showing processing for recovering the multicast relay control table before the failure when the multicast relay device 1000 recovers from the failure. When a failure occurs in the multicast relay apparatus 1000 and then recovered, the multicast relay control table is initialized. In this embodiment, the multicast relay apparatus 1000 creates a copy of the multicast relay control table in an adjacent apparatus (multicast router 2000 in this embodiment) in order to quickly restore the multicast relay control table. When the multicast relay apparatus 1000 recovers from the failure, the multicast relay apparatus 1000 recovers its own multicast relay control table based on a copy of the multicast relay control table created in the multicast router 2000.

  Specifically, the multicast relay apparatus 1000 is used to generate a copy of the multicast relay control table in the IGMP join request / leave request message (arrow F215, arrow F216, arrow F217) transmitted from each multicast receiver. Information (hereinafter, also referred to as “additional information”) is added. The multicast relay apparatus 1000 relays the IGMP join request / leave request message to which the additional information is added to the multicast router 2000 via the multicast router port 1001 according to the IGMP snooping process.

  The multicast router 2000 creates a duplicate entry table T2430 that is a duplicate of the multicast packet relay control table of the multicast relay apparatus 1000 based on the IGMP join request / leave request message to which the additional information is added. In this specification, “replication” means not only a complete replication of a replication target but also information including information that can restore the replication target to its original state.

  When a failure occurs in the multicast relay device 1000 and then recovers from the failure, the multicast relay control table entry in the multicast relay device 1000 is initialized. Therefore, after the multicast relay apparatus 1000 recovers from the failure, it requests the multicast router 2000 for an entry in the replication entry table T2430 (arrow F218).

  The multicast router 2000 promptly responds to the entry from the replication entry table T2430 in response to the request from the multicast relay apparatus 1000 (arrow F219). The multicast relay apparatus 1000 reproduces the entry in the multicast packet relay control table before the occurrence of the failure based on the entry of the response to the duplicate entry table T2430.

  In this way, entries in the multicast packet relay control table before the occurrence of a failure can be quickly recreated, and multicast communication can be resumed promptly. In addition, compared to a comparative example described later, since the IGMP join request message responding to the multicast relay apparatus 1000 can be suppressed to a minimum, the communication bandwidth can be reduced.

  Further, since the multicast packet relay control table is updated by the join request message or the leave request message, if the additional information is added to the join request message or the leave request message and relayed, a copy that is a duplicate of the multicast packet relay control table The update of the entry table T2430 can be synchronized with the update of the multicast packet relay control table. Below, the structure and operation | movement detail of each apparatus are demonstrated.

A2. Configuration of each device:
FIG. 3 is an explanatory diagram showing the internal configuration of the multicast relay apparatus 1000. The multicast relay apparatus 1000 includes a device configuration definition unit 1200, an IGMP snooping processing unit 1100, a multicast packet relay processing unit 1300, and four ports 1001, 1002, 1003, and 1004. The multicast relay apparatus 1000 includes a CPU (not shown), a ROM, a RAM, and the like, and the CPU can read and execute a program expanded in the RAM, thereby executing the functions indicated therein. . The same applies to other devices described below.

  The device configuration definition unit 1200 stores VLAN information D1210 and IGMP snooping information D1220. The VLAN information D1210 and the IGMP snooping information D1220 stored in the device configuration definition unit 1200 can be set from the outside.

  The VLAN information D1210 includes a VLAN name that is a VLAN name set in the multicast relay apparatus 1000, an IP address set in the VLAN, and port information belonging to the VLAN. Since the VLAN name is for identifying the VLAN, it may be anything that can identify the VLAN, such as a VLAN number.

  The IGMP snooping information D1220 includes a VLAN name of the VLAN in which the IGMP snooping function is used, multicast router port information, and a VLAN for which a replication entry table is created (hereinafter also referred to as a “duplication entry target”). And duplicate entry information for determining whether or not.

  The IGMP snooping processing unit 1100 is a processing unit for realizing the IGMP snooping function, and includes a VLAN management unit 1110, a control packet transmission / reception processing unit 1120, and an IGMP snooping entry generation unit 1130. Inside the control packet transmission / reception processing unit 1120 is provided an additional information processing unit for adding additional information to the IGMP join request / leaving request message relayed to the multicast router 2000. The entry generation unit 1130 generates an entry management table T1131 (relay control information) for realizing the IGMP snooping function.

  The multicast packet relay processing unit 1300 performs multicast packet relay processing and IGMP control packet (IGMP join request / leave request message) reception processing. The multicast packet relay processing unit 1300 includes a VLAN registration slip T1310 and a multicast packet relay control table T1320.

  When the VLAN information D1210 and the IGMP snooping information D1220 are externally set for the device configuration definition unit 1200, the IGMP snooping processing unit 1100 acquires the VLAN information D1210 and the IGMP snooping information D1220 from the device configuration definition unit 1200 (arrows) F300, arrow F301), and is registered in the VLAN management table T1111 in the VLAN management unit 1110.

  When the VLAN information D1210 and the IGMP snooping information D1220 are registered in the VLAN management table T1111, the IGMP snooping processing unit 1100 notifies the multicast packet relay processing unit 1300 of the contents of the VLAN management table T1111 and performs multicast packet relay processing. Registered in the VLAN registration table T1310 in the unit 1300 (arrow F302). That is, the contents of the VLAN registration slip T1310 are the same as the contents of the VLAN management table T1111.

  When the multicast packet relay processing unit 1300 receives an IGMP control packet (IGMP join request / leaving request / inquiry message) from the VLAN registered in the VLAN registration table T1310, the multicast packet relay processing unit 1300 stores the IGMP control packet in the IGMP snooping processing unit 1100. The control packet transmission / reception processing unit 1120 is notified.

  When the IGMP snooping information D1220 of the device configuration definition unit 1200 is updated and a VLAN that uses the IGMP snooping function is newly added to the VLAN management table T1111, the IGMP snooping processing unit 1100 includes the multicast packet relay processing unit 1300. Is notified of the added VLAN and registered in the VLAN registration table T1310 in the multicast packet relay processing unit 1300 (arrow F302).

  Also in this case, when receiving the IGMP control packet from the added VLAN, the multicast packet relay processing unit 1300 notifies the control packet transmission / reception processing unit 1120 in the IGMP snooping processing unit 1100 of the IGMP control packet. The processing in the IGMP snooping processing unit 1100 will be described later.

  FIG. 4 is an explanatory diagram showing various information set in the multicast relay apparatus 1000 and the contents of the table. FIG. 4A is an explanatory diagram showing an example of the VLAN information D1210. In the VLAN information D1210, information related to the VLAN set in the multicast relay apparatus 1000 is set. The VLAN information D1210 includes a VLAN name, an IP address, and port information. In the example shown in FIG. 4A, “VLAN101” is set in the VLAN name field 1211 of the VLAN information D1210, “IP_addr1” that is the address set in the VLAN 101 is set in the IP address field 1212, In the port information field 1213, “1001, 1002, 1003, 1004” which are ports belonging to the VLAN 101 are set.

  FIG. 4B is an explanatory diagram showing an example of IGMP snooping information D1220. In the IGMP snooping information D1220, information on a VLAN that uses IGMP snooping is set. The IGMP snooping information D1220 includes a VLAN name, a multicast router port, and copy entry information. In the example shown in FIG. 4B, “VLAN101” is set in the VLAN name field 1221, and “1001” that is the port to which the multicast router 2000 is connected is set in the multicast router port field 1222. In the duplicate entry information field 1223, “YES” indicating that it is a duplicate entry target is set.

  FIG. 4C is an explanatory diagram showing an example of the VLAN management table T1111. The VLAN management table T1111 is created based on the VLAN information D1210 and the IGMP snooping information D1220. In the example shown in FIG. 4C, “VLAN101” is set in the VLAN name field 1112, “IP_addr1” is set in the IP address field 1113, and “1001, 1002, 1003” are set in the port information field 1114. , 1004 ”,“ 1001 ”is set in the multicast router port field 1115, and“ YES ”is set in the duplicate entry information field 1223.

  FIG. 4D is an explanatory diagram showing an example of the entry management table T1131. The entry management table T1131 includes a VLAN name, a multicast group address, and port information. In the example shown in FIG. 4D, “VLAN101” is set in the VLAN name field 11311, “G1” is set in the multicast group address field 11312, and “1004” is set in the port information field 11313. Has been. In this case, the multicast packet relay processing unit 1300 relays the multicast packet addressed to the multicast group address “G1” only to the multicast receiver 40 connected to the port 1004.

  FIG. 5 is an explanatory diagram showing the internal configuration of the multicast router 2000. The multicast router 2000 includes an apparatus configuration definition unit 2100, a multicast protocol processing unit 2200, a multicast packet relay processing unit 2300, and a duplicate entry processing unit 2400.

  The device configuration definition unit 2100 stores multicast information D2110 and duplicate entry generation information D2120. The multicast information D2110 and the duplicate entry creation information D2120 stored in the device configuration definition unit 2100 can be set from the outside.

  The duplicate entry creation information D2120 includes the VLAN name of the VLAN for which the duplicate entry table T2430 is to be created, the IP address set in the VLAN of the multicast relay device that is the duplicate entry target, and the multicast relay device that is the duplicate entry target. And port information to which is connected.

  The multicast protocol processing unit 2200 acquires VLAN information using multicast from the multicast information D2110 stored in the device configuration definition unit 2100, and registers the VLAN information in the VLAN registration table T2310 in the multicast packet relay processing unit 2300. To do. Thus, when receiving the IGMP control packet from the VLAN, the multicast packet relay processing unit 2300 notifies the multicast protocol processing unit 2200 of the IGMP control packet.

  The duplicate entry processing unit 2400 includes a duplicate entry management table T2410, a control packet transmission / reception processing unit 2420, and a duplicate entry table T2430. The replication entry processing unit 2400 acquires the replication entry generation information D2120 from the device configuration definition unit 2100 and registers it in the replication entry management table T2410 (arrow F316). Therefore, the contents of the duplicate entry management table T2410 are the same as the contents of the duplicate entry generation information D2120.

  When the replication entry processing unit 2400 acquires the VLAN to be replicated entry from the replication entry generation information D2120, the replication entry processing unit 2400 registers the acquired VLAN in the VLAN registration table T2310 in the multicast packet relay processing unit 2300 (arrow F317).

  When the multicast packet relay processing unit 2300 receives the IGMP join request / leave request message from the VLAN registered in the VLAN registration table T2310, it notifies the duplicate entry processing unit 2400. The process of the duplicate entry processing unit 2400 when the IGMP join request / leave request message is received will be described later.

  When receiving the IGMP join request message from the port 2002, the multicast packet relay processing unit 2300 notifies the multicast protocol processing unit 2200 of the IGMP join request message (arrow F201).

  When receiving the notification of the IGMP join request message, the multicast protocol processing unit 2200 performs multicast route calculation. The multicast protocol processing unit 2200 registers the calculated multicast route in the multicast packet relay control table T2320. Thereby, the multicast packet relay processing unit 2300 can relay the multicast packet from the VLAN 100 to the VLAN 101 (arrow F101). A description of the multicast route creation processing here is omitted.

  Through the above processing, the multicast packet distributed from the multicast sender 10 is relayed to the multicast router 2000 and the multicast relay apparatus 1000, and the IGMP snooping entry registered in the multicast packet relay control table 1320 of the multicast relay apparatus 1000 uses the IGMP packet. It will be relayed only to the multicast receiver that sent the join request message.

  FIG. 6 is an explanatory diagram showing various information set in the multicast router 2000 and the contents of the table. FIG. 6A is an explanatory diagram showing an example of the duplicate entry generation information D2120. In the example shown in FIG. 6, “VLAN101”, which is the VLAN targeted for replication entry, is set in the VLAN name field 2121 of the replication entry generation information D2120, and the multicast relay device 1000 of the replication entry target in the IP address field 2122. “IP_addr1” that is an IP address set in the VLAN 101 is set, and “2002” that is a port to which the multicast relay apparatus 1000 that is a replication entry target is connected is set in the port information field 2123.

  FIG. 6B is an explanatory diagram showing an example of the duplicate entry management table T2410. The contents of the duplicate entry management table T2410 are the same as the contents of the duplicate entry generation information D2120 described above.

  FIG. 6C is an explanatory diagram showing an example of the replication entry table T2430. In the example shown in FIG. 6C, “VLAN101” is set in the VLAN name field 2431, and “IP_addr1” that is the IP address set in the VLAN 101 of the multicast relay apparatus 1000 is set in the IP address field 2432. Has been. In the multicast group address field 2433, “G1” which is the multicast group address of the multicast group to which the receiver 40 has joined is set, and in the port information field 2434, the multicast relay device 1000 acquired from the additional information of the IGMP join request message. The port “1004” is set. A method for creating and using the duplicate entry table T2430 will be described later.

  As will be described later, the multicast relay apparatus 1000 relays the IGMP join request / leave request message to which the additional information is added to the multicast router port. Therefore, the multicast router 2000 can create relay control information (contents of the duplicate entry table T2430) having the same content as the relay control information (contents of the entry management table T1131) of the multicast relay device 1000.

A3. Update processing of entry management table T1131 in multicast relay apparatus 1000:
When the multicast relay apparatus 1000 receives an IGMP join request message or an IGMP leave request message in the IGMP control packet, the multicast relay apparatus 1000 performs an update process of the entry management table T1131 (relay control information update process). Hereinafter, update processing of the entry management table T1131 will be described.

  FIG. 7 is a flowchart illustrating processing when the multicast relay apparatus 1000 receives an IGMP join request message. When the multicast relay apparatus 1000 receives an IGMP join request message from a multicast receiver (step S6100), the control packet transmission / reception processing unit 1120 analyzes the IGMP join request message and acquires a multicast group address and a reception port (step S6101). ). The acquired multicast group address and reception port are notified to the entry generation unit 1130.

  The entry generation unit 1130 searches the entry management table T1131 using the acquired multicast group address as a key (step S6102), and determines whether there is an entry that matches the multicast group address (step S6103).

  If there is an entry that matches the acquired multicast group address (step S6103: Yes), the entry generation unit 1130 determines whether the acquired port has already been registered in the entry. (Step S6105).

  If the port has already been registered (step S6105: Yes), the entry generation unit 1130 ends the process (step S6108). If the port is not registered, the entry generation unit 1130 adds the port to the port information field 11313 of the entry (step S6106).

  On the other hand, when there is no entry that matches the acquired multicast group address (step S6103: No), the entry generation unit 1130 newly creates an entry of the acquired multicast group in the entry management table T1131 (step S6103). In step S6104, the acquired port is registered in the port information field 11313 of the entry management table 1131 (step S6106).

  When the entry management table T1131 is updated as described above, the entry generation unit 1130 notifies the multicast packet relay processing unit 1300 of the updated entry (step S6107).

  FIG. 8 is a flowchart showing processing when the multicast relay apparatus 1000 receives an IGMP leave request message. When the multicast relay apparatus 1000 receives an IGMP leave request message from a multicast receiver (step S6200), the control packet transmission / reception processing unit 1120 analyzes the IGMP leave request message and acquires a multicast group address and a reception port (step S6201). ).

  The entry generation unit 1130 searches the entry management table T1131 using the acquired multicast group address as a key (step S6202), and determines whether there is an entry that matches the multicast group address (step S6203). .

  If there is no entry that matches the acquired multicast group address (step S6203: No), the entry generation unit 1130 ends the process (step S6204). On the other hand, when there is an entry that matches the acquired multicast group address (step S6203: Yes), the entry generation unit 1130 determines whether or not the acquired port is registered (step S6205). If the port is not registered (step S6205: NO), the entry generating unit 1130 ends the process (step S6206).

  On the other hand, when the port is registered (step S6205: Yes), the entry generation unit 1130 determines whether the port is the last port (step S6207). When the port is the last port (step S6207: Yes), the entry itself of the multicast group address is deleted from the entry management table T1131 (step S6208). If the port is not the last port (step S6207: No), the port is deleted from the port information field 11313 of the entry (step S6209).

  When the entry management table 1131 is updated as described above, the entry generation unit 1130 notifies the updated entry to the multicast packet relay processing unit 1300 (step S6210).

  The reception processing of the IGMP join request / leave request message in the multicast relay apparatus 1000 described above will be described with a specific example.

  FIG. 9 is an explanatory diagram showing a flow of processing when the multicast relay apparatus 1000 receives an IGMP join request message from the multicast receiver 40. At this time, the multicast group address of the multicast group that the multicast receiver 40 requests to join is “G1”.

  When the multicast receiver 40 transmits the IGMP join request message, the multicast relay apparatus 1000 receives the IGMP join request message from the port 1004 to which the multicast receiver 40 is connected (arrow F200). Since the VLAN 101 is registered in the VLAN registration slip T1310 of the multicast packet relay processing unit 1300, the IGMP join request message received from the VLAN 101 is notified to the control packet transmission / reception processing unit 1120 in the IGMP snooping processing unit 1100 ( Arrow F200).

  The control packet transmission / reception processing unit 1120 analyzes the notified IGMP join request message and acquires a multicast group address and a reception port. The control packet transmission / reception processing unit 1120 notifies the entry generation unit 1130 of the acquired multicast group address, the acquired reception port, and the VLAN name of the VLAN that received the IGMP join request message (arrow F303).

  The entry generation unit 1130 registers the notified multicast group address, reception port, and VLAN name in the entry management table T1131. The entry generating unit 1130 notifies the multicast packet relay processing unit 1300 of the VLAN name, the multicast group address, and the reception port registered in the entry management table T1131 (arrow F304), and the notified content is the multicast packet relay control table T1320. Register with.

  Through the above processing, when the multicast packet relay processing unit 1300 receives a multicast packet from the VLAN registered in the multicast packet relay control table T1320, the multicast packet relay control table T1320 selects a multicast packet that matches the registered multicast group address. Relay to only the port registered in the entry (arrow F101).

  FIG. 10 is an explanatory diagram showing how the entry management table T1131 is updated. FIG. 10A shows the entry management table T1131 when an IGMP join request message is received from the port 1004 to which the multicast receiver 40 is connected. As shown in FIG. 10A, “VLAN101” is registered in the VLAN name field 11311 of the entry management table T1131, “G1” is registered in the multicast group address field 11312, and the port information field 11313 is registered. Is registered with “1004”, which is a port to which the multicast receiver 40 is connected.

  FIGS. 10B to 10D show the entry management table T1131 when the multicast relay apparatus 1000 further receives an IGMP join request message or an IGMP leave request message. Details will be described later.

  Next, processing when multicast relay apparatus 1000 receives an IGMP join request message for the same multicast group address “G1” from another port will be described.

  FIG. 11 is an explanatory diagram showing the flow of processing when the multicast relay apparatus 1000 receives an IGMP join request message from the multicast receiver 30. When receiving the IGMP join request message from the port 1003 to which the multicast receiver 30 is connected, the multicast relay apparatus 1000 updates the entry management table T1131 based on the received IGMP join request message as described above. Do.

  Specifically, the control packet transmission / reception processing unit 1120 analyzes the received IGMP join request message, and acquires a multicast group address and a reception port (arrow F202). The control packet transmission / reception processing unit 1120 notifies the entry generation unit 1130 of the acquired multicast group address, the acquired reception port number, and the VLAN name of the VLAN that received the IGMP join request message (arrow F305).

  The entry generation unit 1130 searches the entry management table T1131 using the acquired multicast group address as a key, and determines whether the same multicast group address is registered in the VLAN. In the example shown in FIG. 11, since the multicast group address is registered in the entry management table T1131 when the IGMP join request message from the port 1004 has already been received, the entry generation unit 1130 uses the acquired reception port. A certain “1003” is added to the port information field 11313 in the multicast group address.

  The entry generation unit 1130 notifies the multicast packet relay processing unit 1300 of the updated entry management table T1131 entry and registers it in the multicast packet relay control table 1320 (arrow F306).

  When the entry management table T1131 is updated as described above, the multicast packet addressed to the multicast group is relayed to the ports 1003 and 1004 (arrow F101). The updated entry management table T1131 is shown in FIG. In the port information field 11313, “1003” which is the acquired reception port is added.

  Next, update processing of the entry management table 1131 when the multicast relay apparatus 1000 receives an IGMP leave request message from a multicast receiver will be described.

  FIG. 12 is an explanatory diagram showing the flow of processing when the multicast relay apparatus 1000 receives an IGMP leave request message from the multicast receiver 30. In the state before the multicast relay apparatus 1000 receives the IGMP leave request message, the multicast receivers 30 and 40 exist at the ports 1003 and 1004, so the multicast packet is relayed to the ports 1003 and 1004.

  When receiving the IGMP leave request message from the port 1003 to which the multicast receiver 30 is connected, the multicast relay apparatus 1000 sends the IGMP leave request message to the control packet transmission / reception processing unit as in the case of receiving the IGMP join request message. 1120 is notified (arrow F203).

  The control packet transmission / reception processing unit 1120 analyzes the received IGMP leave request message, and acquires a multicast group address, a reception port, and a VLAN name. The control packet transmission / reception processing unit 1120 notifies the entry generation unit 1130 of the acquired multicast group address, the acquired reception port, and the acquired VLAN name (arrow F307).

  The entry generation unit 1130 searches the entry management table T1131, and determines whether or not the multicast group address has been registered in the received VLAN entry. If it has been registered, it is obtained from the port information field of the entry. Delete the receiving port. Then, the entry generation unit 1130 notifies the updated entry management table T1131 entry to the multicast packet relay processing unit 1300 and registers it in the multicast packet relay control table 1320 (arrow F308).

  Through the above processing, the multicast packet addressed to the multicast group is relayed only to the port 1004 (arrow F101). The updated entry management table T1131 is shown in FIG. In the port information field 11313, the acquired reception port “1003” is deleted, and only “1004” is obtained.

  Next, processing when an IGMP leave request message is received from the last multicast receiver will be described.

  FIG. 13 is an explanatory diagram showing a flow of processing when the multicast relay apparatus 1000 receives an IGMP leave request message from the multicast receiver 40. In a state before receiving the IGMP leave request message, since the multicast receiver 40 exists at the port 1004, the multicast packet is relayed to the port 1004.

  When receiving the IGMP leave request message from the port 1004 to which the multicast receiver 40 is connected, the multicast relay apparatus 1000 notifies the control packet transmission / reception processing unit 1120 of the IGMP leave request message as described above ( Arrow F204).

  The control packet transmission / reception processing unit 1120 analyzes the received IGMP leave request message, and acquires a multicast group address, a reception port, and a VLAN name. The control packet transmission / reception processing unit 1120 notifies the entry generation unit 1130 of the acquired multicast group address, the acquired reception port number, and the acquired VLAN name (arrow F309).

  The entry generation unit 1130 searches the entry management table T1131, and determines whether or not the multicast group address has been registered in the received VLAN. If it is registered and the port is the last port, the entry is deleted. Then, the entry generation unit 1130 notifies the multicast packet relay processing unit 1300 of the deleted entry, and deletes the entry from the multicast packet relay control table T1320 (arrow F310).

  Through the above processing, the multicast packet addressed to the multicast group is not registered in the multicast packet relay control table T1320, and therefore is not relayed anywhere (arrow F101). The updated entry management table T1131 is shown in FIG. Since there are no more multicast recipients, the entry itself has been deleted.

A4. Relay process of IGMP control packet in multicast relay apparatus 1000:
When the multicast relay apparatus 1000 receives the IGMP join request / leave request message, the multicast relay apparatus 1000 updates the entry management table T1131 and adds additional information to the IGMP join request / leave request message to the multicast router 2000. Relay. Further, when receiving the IGMP inquiry message, the multicast relay apparatus 1000 relays (floods) the inquiry message to the port in the VLAN. Below, the relay process of the IGMP control packet in the multicast relay apparatus 1000 is demonstrated.

  FIG. 14 is a flowchart showing control packet relay processing in the multicast relay apparatus 1000. When receiving the IGMP control packet from the port in the VLAN (step S6300), the multicast packet relay processing unit 1300 refers to the “type” included in the IGMP control packet (step S6301) and determines the type of the packet (step S6301). S6302).

  If the received control packet type is an IGMP inquiry message (step S6302: Yes), the control packet transmission / reception processing unit 1120 searches the VLAN management table T1111 for the VLAN that has received the IGMP inquiry message, and excludes the reception port. All port numbers in the VLAN are acquired (step S6304). The control packet transmission / reception processor 1120 relays the IGMP inquiry message to the port corresponding to the acquired port number (step S6305).

  On the other hand, when the type of the received control packet is an IGMP join request message or an IGMP leave request message (step S6302: No), the control packet transmission / reception processing unit 1120 refers to the VLAN management table T1111 (step S6306), It is determined whether or not the VLAN that has received the IGMP join request message or the IGMP leave request message is a replication entry target (step S6307).

  If the VLAN is a replication entry target (step S6307: Yes), the control packet transmission / reception processing unit 1120 determines whether the control packet is an IGMP join request message (step S6308). When the control packet is an IGMP join request message (step S6308: Yes), the control packet transmission / reception processing unit 1120 determines whether the VLAN_IP address included in the additional information of the IGMP join request message matches the IP address of the reception VLAN. It is determined whether or not (step S6309). The VLAN_IP address of the additional information will be described later.

  If the IP addresses match (step S6309: Yes), the packet is distributed from the multicast router 2000 at the time of failure recovery, which will be described later, and is an IGMP join request message valid only in this multicast relay device. The process ends without relaying the message anywhere (step S6310).

  On the other hand, when the IGMP join request message does not match the IP address (step S6309: No) or the IGMP leave request message (step S6308: No), the control packet transmission / reception processing unit 1120 analyzes and receives the packet. A port number is acquired (step S6311). The additional information processing unit 1121 sets the acquired reception port and the IP address of the VLAN as additional information in the packet (step S6312).

  On the other hand, when the VLAN is not a replication entry target (step S6307: No), the additional information processing unit 1121 does not set additional information in the packet. Then, the control packet transmission / reception processing unit 1120 searches the VLAN management table 1111 for the VLAN that has received the IGMP join request message or the IGMP leave request message, and acquires the multicast router port number in the VLAN (step S6303). The control packet transmission / reception processing unit 1120 transmits the IGMP join request message or the IGMP leave request message to the multicast router port 1001 (step S6305).

  FIG. 15 is an explanatory diagram showing a format of an IGMP join request / leave request message to which additional information is added. The additional information processing unit 1121 adds the VLAN_IP address, the number of ports, and the port number to the header of a normal IGMP join request / leave request message as additional information. The additional information processing unit 1121 sets the IP address set for the replication entry target VLAN in the VLAN_IP address area, and sets the port information subscribing to the multicast group in the port number and port number areas. When there are a plurality of ports, the number of ports can be set in the port number area, and the port number can be set repeatedly by the number of the ports. Note that “Type” represents the “type” of the IGMP control packet.

  FIG. 16 is an explanatory diagram showing the flow of the relay process of the IGMP join request message in the multicast relay apparatus 1000. When receiving the IGMP join request message from the port 1004, the multicast packet relay processing unit 1300 notifies the control packet transmission / reception processing unit 1120 of the IGMP join request message (arrow F217). The control packet transmission / reception processing unit 1120 searches the VLAN management table 1111 for the VLAN that has received the IGMP join request message, acquires the multicast router port number of the VLAN, and stores the duplicate entry information D1116 included in the VLAN management table T1111. A search is performed to determine whether the VLAN that has received the IGMP join request message is a copy entry target (arrow F315).

  When the VLAN is the target of the replication entry, the additional information processing unit 1121 in the control packet transmission / reception processing unit 1120 displays “IP_addr1” that is the IP address of the VLAN of the multicast relay device 1000 and the multicast relay device 1000. Is added to the IGMP join request message. The control packet transmission / reception processing unit 1120 relays the IGMP join request message to which the additional information is added to the acquired multicast router port 1001 (arrow F220).

  In this embodiment, the IP address set in the VLAN of the relevant device is added to discriminate the replication entry target device (multicast relay device 1000), but the replication entry target device can be discriminated. In this case, it is not limited to the IP address set in the VLAN of the device, but may be a VLAN name or the like.

  FIG. 17 is an explanatory diagram showing the flow of the relay process of the IGMP leave request message in the multicast relay apparatus 1000. When receiving the IGMP leave request message from the port 1003, the multicast packet relay processing unit 1300 notifies the control packet transmission / reception processing unit 1120 of the IGMP leave request message in the same manner as when receiving the IGMP join request message (arrow F203). The control packet transmission / reception processing unit 1120 searches the VLAN management table 1111 for the VLAN that has received the IGMP join request message, acquires the multicast router port number of the VLAN, and stores the duplicate entry information D1116 included in the VLAN management table T1111. A search is performed to determine whether the VLAN that has received the IGMP leave request message is a duplicate entry target (arrow F312).

  When the VLAN is the target of the replication entry, the additional information processing unit 1121 in the control packet transmission / reception processing unit 1120 displays “IP_addr1” that is the IP address of the VLAN of the multicast relay device 1000 and the multicast relay device 1000. Is added to the IGMP leave request message. The control packet transmission / reception processing unit 1120 relays the IGMP leave request message to which the additional information is added to the acquired multicast router port 1001 (arrow F206).

  FIG. 18 is an explanatory diagram showing the flow of the relay process of the IGMP inquiry message in the multicast relay apparatus 1000. As described above, the multicast router 2000 of this embodiment is an IGMP querier. For this reason, when receiving the IGMP leave request message, the multicast router 2000 transmits an IGMP inquiry message to the multicast group. The multicast packet relay processing unit 1300 notifies the control packet transmission / reception processing unit 1120 of the IGMP inquiry message received from the multicast router 2000, similarly to the IGMP join request / leaving request message (arrow F207).

  The control packet transmission / reception processing unit 1120 searches the VLAN management table 1111 for the VLAN that has received the IGMP inquiry message, and acquires all port numbers in the VLAN other than the port that has received the IGMP inquiry message (arrow F313). Then, the control packet transmission / reception processing unit 1120 relays (floods) the IGMP inquiry message to the acquired port (arrow F208).

  At this time, if a multicast receiver belonging to the multicast group exists at the port 1004 in the VLAN of the multicast relay apparatus 1000, an IGMP join request message is returned from the multicast receiver 40 (arrow F209). The IGMP join request message returned from the multicast receiver 40 is relayed to the multicast router port 1001 by the same method as described above. Thereby, since the multicast router 2000 can recognize that the multicast receiver still exists, the multicast relay from the upstream is not stopped.

  On the other hand, when there is no multicast receiver belonging to the multicast group at the port in the VLAN of the multicast relay apparatus 1000, since the IGMP join request message is not responded by any multicast receiver, the multicast router 2000 receives the multicast reception. It is determined that there is no user, and multicast relay from the upstream is stopped.

A5. Updating the replication entry table T2430 in the multicast router 2000:
As described above, the multicast relay apparatus 1000 adds additional information to the join request / leave request message and relays the join request / leave request message to the multicast router 2000. The multicast router 2000 updates the copy entry table T2430 based on the join request / leave request message to which the additional information is added. Hereinafter, update processing of the replication entry table T2430 in the multicast router 2000 will be described.

  FIG. 19 is a flowchart showing the update process of the duplicate entry table T2430 when the multicast router 2000 receives the IGMP join request message. When the multicast packet relay processing unit 2300 in the multicast router 2000 receives the IGMP join request message from the port in the VLAN (step S6400), the IGMP join request message is notified to the control packet transmission / reception processing unit 2420. The control packet transmission / reception processing unit 2420 searches the replication entry management table T2410 and determines whether or not the received VLAN is a replication entry target (step S6401).

  As a result of the search, if it is not a duplicate entry target (step S6402: No), the control packet transmission / reception processing unit 2420 ends the process (step S6403). On the other hand, if it is a replication entry target (step S6402: Yes), the control packet transmission / reception processing unit 2420 analyzes the IGMP join request message and copies the replication target device (this embodiment) included in the multicast group address and additional information. Then, the port number of the multicast relay apparatus 1000) is acquired (step S6404).

  The control packet transmission / reception processing unit 2420 searches the replication entry table T2430 using the acquired multicast group address and received VLAN information as keys, and determines whether or not the multicast group address has been registered in the replication entry table T2430 (step S2430). S6405).

  If the multicast group address is not registered in the duplicate entry table T2430 as a result of the search (step S6406: No), the control packet transmission / reception processing unit 2420 newly creates an entry for the multicast group address in the duplicate entry table T2430. (Step S6407), the port number acquired from the additional information is added to the port information field of the entry (Step S6410).

  On the other hand, if the multicast group address has already been registered in the replication entry table T2430 (step S6406: Yes), the control packet transmission / reception processing unit 2420 displays the IGMP information in the port information field corresponding to the multicast group address in the replication entry table T2430. It is determined whether the port acquired from the additional information of the subscription request message is registered (step S6408).

  As a result of the confirmation, if registered (step S6409: Yes), the control packet transmission / reception processing unit 2420 ends the process (step S6411). If not registered (step S6409: No), the control packet transmission / reception processing unit 2420 adds the port number acquired from the additional information of the IGMP join request message to the port information field of the multicast group in the duplicate entry table T2430 ( Step S6410).

  FIG. 20 is a flowchart showing the update process of the duplicate entry table T2430 when the multicast router 2000 receives the IGMP leave request message. When the multicast packet relay processing unit 2300 in the multicast router 2000 receives an IGMP leave request message from a port in the VLAN (step S6500), the IGMP leave request message is notified to the control packet transmission / reception processing unit 2420. The control packet transmission / reception processing unit 2420 searches the replication entry management table 2410 and determines whether or not the received VLAN is a replication entry target (step S6501).

  As a result of the search, if it is not a duplicate entry target (step S6502: No), the control packet transmission / reception processing unit 2420 ends the process (step S6503). On the other hand, if it is a replication entry target (step S6502: Yes), the control packet transmission / reception processing unit 2420 analyzes the IGMP leave request message and copies the replication target device (this embodiment) included in the multicast group address and additional information. Then, the port number of the multicast relay apparatus 1000) is acquired (step S6504).

  The control packet transmission / reception processing unit 2420 searches the replication entry table T2430 using the acquired multicast group address and received VLAN information as keys, and determines whether or not the multicast group address has been registered in the replication entry table T2430 (step S2430). S6505).

  As a result of the search, if the multicast group address is not registered in the replication entry table T2430 (step S6506: No), the control packet transmission / reception processing unit 2420 ends the process (step S6507).

  On the other hand, when the multicast group address has already been registered in the replication entry table T2430 (step S6506: Yes), the control packet transmission / reception processing unit 2420 displays the IGMP information in the port information field corresponding to the multicast group address in the replication entry table T2430. It is determined whether the port number acquired from the additional information of the withdrawal request message is registered (step S6508).

  If it is determined that the port number is not registered (step S6509: NO), the control packet transmission / reception processing unit 2420 ends the process (step S6510). On the other hand, if the port number has been registered (step S6509: Yes), the control packet transmission / reception processing unit 2420 determines whether the port is the last port.

  If the port is the last port (step S6511: Yes), the entry itself is deleted from the duplicate entry table T2430 (step S6513). If the port is not the last port (step S6511: No), the port is deleted from the port information field of the entry (step S6512).

  FIG. 21 is an explanatory diagram showing a flow of update processing of the replication entry table T2430 when the multicast router 2000 receives the IGMP join request message. As described above, when receiving the IGMP join request message from the VLAN registered in the VLAN registration table T2310, the multicast packet relay processing unit 2300 notifies the duplicate entry processing unit 2400 of the IGMP join request message (arrow F220). ).

  The control packet transmission / reception processing unit 2420 in the replication entry processing unit 2400 analyzes the notified IGMP join request message and acquires the VLAN_IP address included in the additional information. The control packet transmission / reception processing unit 2420 determines whether or not the received VLAN and the acquired VLAN_IP address are registered in the replication entry management table T2410 (arrow F318).

  When it is determined that the received VLAN and the acquired VLAN_IP address are registered, the control packet transmission / reception processing unit 2420 transmits the VLAN name that received the IGMP join request message, the IP address of the VLAN, and the notified IGMP. The multicast group address acquired from the join request message and the port information included in the additional information are registered in the replication entry table T2430 (arrow F319).

A6. State before failure:
FIG. 22A is an explanatory diagram illustrating a situation of multicast communication before a failure occurs in the multicast relay apparatus 1000. FIG. 22B is an explanatory diagram showing a state of the entry management table T1131 before a failure occurs in the multicast relay apparatus 1000.

  Four ports are connected to the VLAN 101 of the multicast relay apparatus 1000, and multicast receivers 20, 30, and 40 are connected to the ports 1002, 1003, and 1004, respectively. In the entry management table T1131 shown in FIG. 22B, the multicast receiver 20 has joined the multicast groups G1, G3, G4, and the multicast receiver 30 has joined the multicast groups G2, G3. 40 subscribes to multicast groups G2 and G4.

  Since the multicast relay apparatus 1000 performs relay processing according to the multicast packet relay control table T1320 having the same contents as the entry management table T1131, the multicast packet transmitted from the multicast sender 10 is registered in the port information field of the entry management table T1131. It is relayed only to the ports that are connected.

  That is, the multicast packet of multicast group G1 is relayed to port 1002 (arrow F102), the multicast packet of multicast group G2 is relayed to ports 1003 and 1004 (arrow F103), and the multicast packet of multicast group G3 is relayed to ports 1002 and 1003. Relayed (arrow F104), the multicast packet of multicast group G4 is relayed to ports 1002 and 1004 (arrow F105).

A7. Status immediately after recovery from failure:
FIG. 23A is an explanatory diagram showing the state of multicast communication immediately after the multicast relay apparatus 1000 recovers from a failure. FIG. 23B is an explanatory diagram showing the state of the entry management table T1131 immediately after the multicast relay apparatus 1000 recovers from the failure.

  Immediately after the multicast relay apparatus 1000 recovers from the failure, the entry management table T1131 is initialized, and no entry is registered in the entry management table T1131. As described above, the multicast packet relay control table T1320 of the multicast relay apparatus 1000 is in the same state as the entry management table T1131, and no IGMP snooping entry exists. Accordingly, all multicast packets received by the VLAN are not relayed to any port (arrow F102, arrow F103, arrow F104, arrow F105).

A8. Recovery method:
FIG. 24 is a flowchart showing processing for requesting distribution of a duplicate entry when the multicast relay apparatus 1000 recovers from a failure. When the IGMP snooping processing unit 1100 recovers from the failure, the IGMP snooping processing unit 1100 acquires the VLAN information D1210 and the IGMP snooping information D1220 from the device configuration definition unit 1200 (step S6600).

  The control packet transmission / reception processing unit 1120 refers to the duplicate entry information D1116 included in the acquired IGMP snooping information D1220, and determines whether or not a duplicate entry target VLAN exists (step S6601). If there is no replication entry target VLAN (step S6602: No), the control packet transmission / reception processing unit 1120 ends the process (step S6603).

  On the other hand, if there is a VLAN for replication entry (step S6602: Yes), the control packet transmission / reception processing unit 1120 acquires the IP address and multicast router port set for the VLAN from the VLAN management table T1111. The control packet transmission / reception processing unit 1120 is notified (step S6604).

  The control packet transmission / reception processing unit 1120 generates an IGMP inquiry message in which the acquired IP address is set as a source address, and transmits the IGMP inquiry message to the acquired multicast router port 1001 (step S6605).

  FIG. 25 is a flowchart showing the replication entry distribution processing in the multicast router 2000. The multicast packet relay processing unit 2300 in the multicast router 2000 receives the IGMP inquiry message transmitted from the multicast relay apparatus 1000 from the port in the VLAN (step S6700). The control packet transmission / reception processing unit 2420 acquires an IP address from the source address included in the IGMP inquiry message, and determines whether or not it has been registered in the replication entry management table T2410 (step S6701).

  As a result of the determination, if it is not registered (step S6702: No), the control packet transmission / reception processing unit 2420 ends the process (step S6703). On the other hand, if registered (step S6702: Yes), the control packet transmission / reception processing unit 2420 acquires port information from the duplicate entry management table T2410 (step S6704).

  The control packet transmission / reception processing unit 2420 searches the replication entry table T2430 and determines whether or not an entry corresponding to the acquired IP address exists in the replication entry table T2430 (step S6705). As a result of the search, if there is no entry corresponding to the acquired IP address (step S6706: No), the control packet transmission / reception processing unit 2420 ends the process (step S6707).

  On the other hand, if there is an entry corresponding to the acquired IP address (step S6706: Yes), the control packet transmission / reception processing unit 2420 generates an IGMP join request message and uses the VLAN registered in the entry as additional information. IP address and port information are added (step S6708). The control packet transmission / reception processing unit 2420 transmits the generated IGMP join request message to the port acquired from the duplicate entry management table T2410 (step S6709). The format of the IGMP subscription request message generated here is the same as the format shown in FIG.

  FIG. 26 is a flowchart showing a process in which the multicast relay apparatus 1000 receives the IGMP join request message distributed from the multicast router 2000 and recreates the entry management table T1131 before the failure occurs. When the multicast packet relay processing unit 1300 receives the IGMP join request message from the multicast router 2000 (step S6100), the IGMP packet request message is notified to the control packet transmission / reception processing unit 1120.

  The control packet transmission / reception processing unit 1120 refers to the VLAN management table T1111 and determines whether the VLAN is a replication entry target (step S6109).

  As a result of the determination, if it is a duplicate entry target (step S6110: Yes), the control packet transmission / reception processing unit 1120 determines whether the VLAN_IP address included in the additional information of the IGMP join request message matches the VLAN address. Is determined (step S6111).

  If the IP addresses do not match (step S6111: No), it is not a response to the distribution request for the duplicate entry, and therefore the control packet transmission / reception processing unit 1120 is the same as the processing when receiving the IGMP join request message from the multicast receiver. The process proceeds to (step S6101 and subsequent processes). The processing after step S6101 is the same processing as the step with the same number shown in FIG.

  On the other hand, if the IP addresses match (step S6111: YES), the control packet transmission / reception processing unit 1120 determines that the IGMP subscription request message is a response to the replication entry distribution request, and adds the IGMP subscription request message. From the information, the multicast group address and port information before the occurrence of the failure are acquired (step S6112). The acquired multicast group address and port information before the occurrence of the failure are notified to the entry generation unit 1130.

  Based on the acquired multicast group address and port information before the failure occurrence, the entry generation unit 1130 performs the processing from step S6102 and updates the entry management table T1131.

  FIG. 27 is an explanatory diagram showing a flow of processing for requesting distribution of a duplicate entry when the multicast relay apparatus 1000 recovers from a failure. Immediately after recovery from the failure, the entries in the entry management table T1131 and the multicast packet relay control table T1320 are initialized, and nothing is registered. Therefore, multicast packet relay processing unit 1300 does not relay the multicast packet received from port 1001 anywhere (arrow F107).

  When recovering from the failure, the IGMP snooping processing unit 1100 reads the VLAN information D1210 and the IGMP snooping information D1220 from the device configuration definition unit 1200 (arrow F320, arrow F321), and registers them in the VLAN management table T1111.

  The control packet transmission / reception processing unit 1120 searches the duplicate entry information field 1116 of the VLAN management table T1111 using “YES” as a key (arrow F322a). That is, the control packet transmission / reception processor 1120 confirms the existence of a VLAN for which entry duplication processing is set to be valid.

  As a result of the search, the control packet transmission / reception processor 1120 displays “1001” in the multicast router port field 1115 and “1001” in the IP address field 1113 registered in the VLAN 101 of the entry for which “YES” is set in the duplicate entry information field 1116. IP_addr1 ”is acquired (arrow F322b).

  The control packet transmission / reception processing unit 1120 generates an IGMP inquiry message, and sets “IP_addr1” that is the acquired IP address as the source address of the IGMP inquiry message. The control packet transmission / reception processing unit 1120 transmits the generated IGMP inquiry message to the acquired multicast router port 1001 of the VLAN. This IGMP inquiry message is transmitted as a duplicate entry distribution request to the multicast router 2000 as an adjacent device (arrow F221).

  FIG. 28 is an explanatory diagram showing the flow of the replication entry distribution process in the multicast router 2000. When receiving an IGMP inquiry message from the port 2002 to which the multicast relay apparatus 1000 is connected, the multicast packet relay processing unit 2300 in the multicast router 2000 sends the IGMP inquiry message to the control packet transmission / reception processing unit 2420 in the duplicate entry processing unit 2400. Notify (arrow F221).

  The control packet transmission / reception processing unit 2420 analyzes the notified IGMP inquiry message and acquires an IP address from the source address included in the IGMP inquiry message. The control packet transmission / reception processing unit 2420 searches the replication entry management table T2410 using the acquired IP address and the received VLAN name as keys (arrow F323a). If there is a matching entry as a result of the search, the IGMP inquiry message is transmitted as an IGMP inquiry transmitted from the duplication target device (in this embodiment, the multicast relay device 1000) registered in the duplication entry management table T2410. Means a message.

  When the IGMP inquiry message is an IGMP inquiry message transmitted from the multicast relay apparatus 1000, the control packet transmission / reception processing unit 2420 is a port to which the multicast relay apparatus 1000 is connected from the duplicate entry management table T2410. 2002 "(arrow F323b).

  The control packet transmission / reception processing unit 2420 searches the replication entry table T2430 using the acquired IP address as a key (arrow F324a). As a result of the search, the control packet transmission / reception processing unit 2420 acquires “G1” in the multicast group address field of the matched entry (arrow F324b).

  The control packet transmission / reception processing unit 2420 generates an IGMP join request message in which the acquired information is added as additional information. Specifically, the control packet transmission / reception processing unit 2420 adds the IP address of the replication target VLAN and the port information of the replication target device (multicast relay device 1000) as additional information of the IGMP join request message. The control packet transmission / reception processing unit 2420 transmits the generated IGMP join request message to the port 2002 connected to the multicast relay apparatus 1000 (arrow F222).

  FIG. 29 is an explanatory diagram showing a flow of processing in which the multicast relay apparatus 1000 receives the IGMP join request message distributed from the multicast router 2000 and recreates the entry management table T1131 before the failure occurs. When receiving the IGMP join request message to which the additional information transmitted from the multicast router 2000 is added, the multicast packet relay processing unit 1300 in the multicast relay apparatus 1000 notifies the control packet transmission / reception processing unit 1120 (arrow F222).

  The control packet transmission / reception processing unit 1120 determines whether or not the VLAN that has received the IGMP join request message is a replication entry target from the replication entry information field 1112 of the VLAN management table T1111 (arrow F325).

  When the VLAN is a replication entry target, the control packet transmission / reception processing unit 1120 analyzes the additional information of the IGMP join request message, acquires the VLAN_IP address included in the additional information, and compares it with the IP address of the VLAN. .

  When the IP address of the VLAN and the VLAN_IP address included in the additional information match, the control packet transmission / reception processor 1120 determines that the received IGMP join request message is a response to the distribution request, and the IGMP join request message The port information included in the additional information is analyzed, and the port information of the multicast relay apparatus 1000 before the failure occurs is acquired.

  The control packet transmission / reception processing unit 1120 notifies the entry generation unit 1130 of the acquired multicast group address and port information before the failure occurs (arrow F326). The entry generation unit 1130 registers the notified multicast group address and port information in the entry management table T1131 and also in the multicast packet relay control table T1320 (arrow F327).

  As a result of the above processing, the entry management table T1131 and the multicast packet relay control table T1320 before the occurrence of the failure are reproduced, and multicast relay is resumed (arrow F107).

  A plurality of ports can be set as additional information in the IGMP join request message transmitted by the multicast router 2000. Therefore, even when there are a plurality of multicast receivers in one multicast group, a plurality of port information can be aggregated in one IGMP join request message and notified to the multicast relay apparatus 1000.

  FIG. 30 is an explanatory diagram showing a state after the entry is reproduced. In the example shown in FIG. 30A, the VLAN 101 is set in the multicast relay apparatus 1000, and the multicast receivers 20, 30, 40, 50, 60 are connected to the VLAN 101. Each multicast receiver subscribes to the same multicast group G1.

  When the multicast relay apparatus 1000 recovers from the failure, it can set a plurality of port information (five port numbers) based on the additional information of the IGMP join request message distributed from the multicast router 2000. The format of the IGMP join request message distributed by multicast router 2000 in this case is shown in FIG.

  As described above, in the first embodiment, since the entry used in the IGMP snooping function is reproduced based on the additional information of the IGMP join request message distributed from the multicast router 2000, the entry is quickly reproduced with a small number of packets. It is possible. That is, one multicast group entry can be recreated by one packet IGMP join request message, and the time until multicast communication is restored can be greatly shortened.

  Further, as the number of ports set in the VLAN is larger and the number of multicast receivers that are joined to the same multicast group is larger, the entry can be reproduced with fewer IGMP join request messages.

A9. How to restore the entry management table in the comparative example:
FIG. 31 is an explanatory diagram showing a method for restoring the entry management table of the multicast relay apparatus 1000 in the comparative example. The flow of the control packet in the entry management table recovery process is shown in FIG. In this comparative example, the replication entry table is not created in the multicast router 2000. When the multicast relay apparatus 1000 recovers from the failure, an IGMP inquiry message is transmitted from the multicast router 2000 (arrow F210), and the IGMP inquiry message is flooded in the VLAN 101 of the multicast relay apparatus 1000.

  Each multicast receiver connected to the multicast relay apparatus 1000 responds to the IGMP inquiry message with an IGMP join request message. Specifically, the multicast receiver 20 transmits an IGMP join request message to the multicast groups G1, G3, G4 (arrow F211, arrow F212, arrow F213). The multicast receiver 30 transmits an IGMP join request message to the multicast groups G2 and G3 (arrow F214 and arrow F215). The multicast receiver 40 transmits an IGMP join request message to the multicast groups G2 and G4 (arrow F216, arrow F217).

  FIG. 31B is an explanatory diagram showing an entry management table remanufactured based on the IGMP join request message transmitted from each multicast receiver. The entry generation unit 1130 in the IGMP snooping processing unit 1100 creates the entry management table 1131 based on the received IGMP join request message and registers it in the multicast relay control table. Through the above processing, multicast communication is resumed.

  In the remanufacturing method of the entry management table shown in this comparative example, when there are a large number of multicast receivers and a large number of multicast groups are joined, a huge number of IGMP join request messages are sent to the multicast relay apparatus 1000. Responded. For this reason, it takes a long time for multicast communication to resume after the IGMP inquiry message is flooded in the VLAN. That is, until the multicast relay apparatus 1000 receives the IGMP inquiry message, and until the IGMP inquiry message is flooded, an entry is created in response to the response to the IGMP join request message, and is set in the multicast packet relay processing unit During this time, multicast communication stops.

  Further, since a large amount of IGMP inquiry messages are transmitted from the multicast sender and the multicast relay apparatus 1000 receives a large amount of IGMP subscription request messages in a concentrated manner, there is a possibility that the communication band may be compressed.

B. Second embodiment:
FIG. 32 is an explanatory diagram showing a network configuration in the second embodiment. In this network configuration, two multicast relay apparatuses 1000 and 3000 are connected to one multicast router 2000. The internal configuration of the multicast relay devices 1000 and 3000 and the internal configuration of the multicast router 2000 in the present embodiment are the same as the internal configuration of the multicast relay device 1000 and the internal configuration of the multicast router 2000 in the first embodiment, respectively. The multicast relay apparatuses 1000 and 3000 in the present embodiment have an IGMP snooping function as in the first embodiment.

  FIG. 33 is an explanatory diagram showing various information of the multicast relay device 1000 and the contents of the VLAN management table T1111. FIG. 33A is an explanatory diagram showing the contents of the VLAN information D1210 of the multicast relay apparatus 1000. The VLAN name field 1211 of the VLAN information D1210 is set to “VLAN101” which is the VLAN set in the multicast relay apparatus 1000, and the IP address field 1212 is the IP address set to the VLAN 101 of the multicast relay apparatus 1000 “ "IP_addr1" is set, and "1001, 1002, 1003" that are ports belonging to the VLAN 101 are set in the port information field 1213.

  FIG. 33 (B) is an explanatory diagram showing the contents of IGMP snooping information D1220 of multicast relay apparatus 1000. In the VLAN name field 1221 of the IGMP snooping information D1220, “VLAN 101” that is a VLAN that uses the IGMP snooping function is set, and in the multicast router port field 1222, “1001” that is a port to which the multicast router 2000 is connected is set. In the replication entry information field 1223, “YES” indicating that the VLAN 101 is a replication entry target is set.

  FIG. 33C is an explanatory diagram showing the contents of the VLAN management table T1111 of the multicast relay apparatus 1000. The VLAN management table T1111 stores each piece of information acquired from the VLAN information D1210 and the IGMP snooping information D1220.

  FIG. 34 is an explanatory diagram showing various information of the multicast relay device 3000 and the contents of the VLAN management table T3111. FIG. 34A is an explanatory diagram showing the contents of the VLAN information D3210 of the multicast relay apparatus 3000. The VLAN name field 3211 of the VLAN information D3210 is set to “VLAN102” which is the VLAN set in the multicast relay device 3000, and the IP address field 3212 is the IP address set to the VLAN 102 of the multicast relay device 3000. IP_addr2 ”is set, and“ 3001, 3002, 3003 ”, which are ports belonging to the VLAN 102 of the multicast relay device 3000, are set in the port information field 3213.

  FIG. 34 (B) is an explanatory diagram showing the contents of IGMP snooping information D3220 of multicast relay apparatus 3000. The VLAN name field 3221 of the IGMP snooping information D3220 is set to “VLAN102” which is a VLAN using the IGMP snooping function, and the multicast router port field 3222 is set to “3001” which is a port to which the multicast router 2000 is connected. In the replication entry information field 3223, “YES” indicating that the VLAN 102 is a replication entry target is set.

  FIG. 34C is an explanatory diagram showing the contents of the VLAN management table T3111 of the multicast relay device 3000. The VLAN management table T3111 stores each piece of information acquired from the VLAN information D3210 and the IGMP snooping information D3220.

  FIG. 35 is an explanatory diagram showing various information of the multicast router 2000 and the contents of various tables. FIG. 35A is an explanatory diagram showing the contents of the duplicate entry creation information D2120 of the multicast router 2000. In the duplicate entry creation information D2120, two VLANs 101 and 102 that are duplicate entry targets are set.

  As the setting of the VLAN 101, “VLAN 101” is set in the VLAN name field 2121, “IP_addr1” that is the IP address set in the VLAN 101 of the multicast relay apparatus 1000 is set in the IP address field 2122, and the port information field 2123 is set. Is set to “2002”, which is a port to which the multicast relay apparatus 1000 is connected.

  As the setting of the VLAN 102, “VLAN102” is set in the VLAN name field 2121, and “IP_addr2” that is the IP address set in the VLAN 102 of the multicast relay device 3000 is set in the IP address field 2122. In the field 2123, “2003” that is a port to which the multicast relay device 3000 is connected is set.

  FIG. 35B is an explanatory diagram showing the contents of the replication entry management table T2410 of the multicast router 2000. The duplicate entry management table T2410 stores the same information as the duplicate entry generation information D2120.

  FIG. 36A shows a duplicate entry when the multicast receiver 20 belonging to the VLAN 101 transmits an IGMP join request message for the multicast group G1, and the multicast receiver 50 belonging to the VLAN 102 transmits an IGMP join request message for the multicast group G2. It is explanatory drawing which shows the production | generation process of table T2430. FIG. 36B is an explanatory diagram showing a duplicate entry table T2430 updated based on the IGMP join request message transmitted from the multicast receivers 20 and 50.

  When the receiver 20 transmits an IGMP join request message for the multicast group G1 (arrow F223), the multicast relay apparatus 1000 adds additional information to the IGMP join request message in the same manner as in the first embodiment described above. The IGMP join request message to which the additional information is added is relayed to the multicast router 2000 via the multicast router port 1001 (arrow F224).

  Similarly, when the receiver 50 transmits an IGMP join request message for the multicast group G2 (arrow F225), the multicast relay device 3000 adds additional information to the IGMP join request message, and the IGMP join to which the additional information is added. The request message is relayed to the multicast router 2000 via the multicast router port 3001 (arrow F226).

  When the IGMP join request message is relayed to the multicast router 2000 (arrow F224, arrow F226), the multicast router 2000 generates a duplicate entry table T2430 by the same method as in the first embodiment described above. As shown in FIG. 36B, the multicast router 2000 generates a replication entry table T2430 based on the IGMP join request message received from each of the VLANs 101 and 102 that are replication entry targets.

  When the multicast relay apparatus 1000 or 3000 recovers from the failure, it makes a replication entry distribution request to the multicast router 2000 in the same manner as in the first embodiment. The multicast router 2000 that has received the distribution request for the replication entry searches the replication entry table T2430 for the replication entry of the device that has made the distribution request and distributes it. The multicast relay apparatus 1000 or 3000 can reproduce the entry before the failure based on the distributed copy entry.

  Thus, in the second embodiment, one multicast router 2000 duplicates the entries of a plurality of multicast relay devices. Then, when the multicast relay device recovers from the failure, the multicast router 2000 distributes the duplicate entry of the device to the multicast relay device. That is, in the second embodiment, a single multicast router 2000 can promptly recreate entries before a failure of a plurality of multicast relay devices. As a result, the plurality of multicast relay devices can quickly resume multicast communication.

C. Third embodiment:
FIG. 37 is an explanatory diagram showing a network configuration in the third embodiment. In the network configuration of this embodiment, two multicast relay devices are connected in series to one multicast router. In other words, the multicast relay device is set in multiple stages in the multicast router.

  A multicast relay device 1000 is connected to the multicast router 2000, and a multicast relay device 3000 is further connected downstream of the multicast relay device 1000. The internal configurations of the multicast relay device 3000 and the multicast router 2000 are the same as the internal configurations of the multicast relay device 1000 and the multicast router 2000 in the first embodiment, respectively. Since the multicast relay apparatus 1000 in the present embodiment is located between the multicast router 2000 and the multicast relay apparatus 3000, a function for requesting an upstream apparatus (multicast router 2000) to copy the entry of the own apparatus, and a downstream apparatus Both functions of generating a duplicate entry of (multicast relay apparatus 3000) are implemented.

  FIG. 38 is an explanatory diagram showing the internal configuration of the multicast relay apparatus 1000 in the third embodiment. The difference from the first embodiment shown in FIG. 3 is that a replication entry processing unit 1400 that generates a replication entry is further provided, and that the replication entry generation information D1230 for setting a replication entry target is included in the device configuration definition unit 1200. Other points are the same as those in the first embodiment. Also, the functions of the duplicate entry generation information D1230 and the duplicate entry processing unit 1400 are the same as the corresponding functions provided in the multicast router 2000 of the first embodiment.

  FIG. 39 is an explanatory diagram showing various information of the multicast relay device 3000 and the contents of the VLAN management table T3111. FIG. 39A is an explanatory diagram showing the contents of the VLAN information D3210 of the multicast relay apparatus 3000. The VLAN name field 3211 of the VLAN information D3210 is set to “VLAN101” which is the VLAN set in the multicast relay device 3000, and the IP address field 3212 is the IP address set to the VLAN 101 of the multicast relay device 3000. IP_addr2 ”is set, and“ 3001, 3002, 3003 ”that are ports belonging to the VLAN 101 of the multicast relay device 3000 are set in the port information field 3213.

  FIG. 39B is an explanatory diagram showing the contents of the IGMP snooping information D3220 of the multicast relay device 3000. The VLAN name field 3221 of the IGMP snooping information D3220 is set to “VLAN101” which is a VLAN using the IGMP snooping function, and the multicast router port field 3222 is “3001” which is a port in the direction to which the multicast router 2000 is connected. Is set, and “YES” indicating that the VLAN 101 is a replication entry target is set in the replication entry information field 3223.

  FIG. 39C is an explanatory diagram showing the contents of the VLAN management table T3111 of the multicast relay apparatus 3000. The VLAN management table T3111 stores each piece of information acquired from the VLAN information D3210 and the IGMP snooping information D3220.

  FIG. 40 is an explanatory diagram showing various information of the multicast relay apparatus 1000 and the contents of various tables. FIG. 40A is an explanatory diagram showing the contents of the VLAN information D1210 of the multicast relay apparatus 1000. “VLAN101”, which is the VLAN set in the multicast relay apparatus 1000, is set in the VLAN name field 1211 of the VLAN information 1210D, and the IP address set in the VLAN 101 of the multicast relay apparatus 1000 is set in the IP address field 1212. "IP_addr1" is set, and "1001, 1002, 1003" that are ports belonging to the VLAN 101 are set in the port information field 1213.

  FIG. 40B is an explanatory diagram showing the contents of the IGMP snooping information D1220 of the multicast relay apparatus 1000. In the VLAN name field 1221 of the IGMP snooping information D1220, “VLAN 101” that is a VLAN that uses IGMP snooping is set, and in the multicast router port field 1222, “1001” that is a port to which the multicast router 2000 is connected is set. In the replication entry information field 1223, “YES” indicating that the VLAN 101 is a replication entry target is set.

  FIG. 40C is an explanatory diagram showing the contents of the VLAN management table T1111 of the multicast relay apparatus 1000. The VLAN management table 1111 stores each piece of information acquired from the VLAN information 1210 and the IGMP snooping information 1220.

  FIG. 40D is an explanatory diagram showing the contents of the duplicate entry creation information D1230 of the multicast relay apparatus 1000. In the duplicate entry creation information D1230, information about the VLAN 101 of the downstream multicast relay device 3000 is set. “VLAN101” is set in the VLAN name field 1231 of the duplicate entry creation information D1230, and “IP_addr2” that is the IP address set in the VLAN 101 of the downstream multicast relay device 3000 is set in the IP address field 1232, and the port information In the field 1233, “1002” which is a port to which the downstream multicast relay device 3000 is connected is set.

  FIG. 40E is an explanatory diagram showing the contents of the replication entry management table T1410 of the multicast relay apparatus 1000. The duplicate entry management table T1410 stores information similar to the duplicate entry generation information D1230.

  FIG. 41 is an explanatory diagram showing the contents of the duplicate entry creation information D2120 and duplicate entry management table T2410 of the multicast router 2000. FIG. 41A is an explanatory diagram showing the contents of the duplicate entry creation information D2120 of the multicast router 2000. Information about the VLAN 101 of the downstream multicast relay apparatus 1000 is set in the duplicate entry generation information D2120. “VLAN101” is set in the VLAN name field 2121 of the duplicate entry creation information D2120, “IP_addr1” that is the IP address set in the VLAN 101 of the multicast relay apparatus 1000 is set in the IP address field 2122, and the port information field 2123. Is set to “2002”, which is a port to which the downstream multicast relay apparatus 1000 is connected.

  FIG. 41B is an explanatory diagram showing the contents of the replication entry management table T2410 of the multicast router 2000. The duplicate entry management table T2410 stores information similar to the duplicate entry generation information D1230.

  FIG. 42 is an explanatory diagram showing the flow of control packets in the network configuration of the third embodiment. When multicast receiver 30 transmits an IGMP join request message for multicast group G1, multicast relay apparatus 3000 receives the IGMP join request message from port 3003 (arrow F227). The multicast relay device 3000 adds additional information to the IGMP join request message and relays it to the multicast relay device 1000 (arrow F228).

  FIG. 43 is an explanatory view showing an IGMP join request message to which additional information is added. FIG. 43A is an explanatory diagram showing a format of an IGMP join request message that is transmitted from the multicast receiver 30 and then added with additional information by the multicast relay device 3000 and relayed to the multicast relay device 1000. In the additional information VLAN_IP address, “IP_addr2” which is the IP address set in the multicast relay device 3000 is set, “1” is set in the port number, and the multicast relay device 3000 has the port number. “3003” which is a port to which the receiver 30 is connected is set.

  When the multicast relay apparatus 1000 (FIG. 42) receives the IGMP join request message with additional information added from the multicast relay apparatus 3000, the multicast relay apparatus 1000 analyzes the received IGMP join request message and creates a duplicate entry. The multicast relay apparatus 1000 further overwrites the IGMP join request message with the additional information of the own apparatus and relays it to the multicast router 2000 via the multicast router port 1001 (arrow F229).

  FIG. 43B is an explanatory diagram showing a format of an IGMP join request message that is relayed from the multicast relay device 3000 and then overwritten by the multicast relay device 1000 and relayed to the multicast router 2000. “IP_addr1” that is an IP address set in the multicast relay apparatus 1000 is set in the VLAN_IP address of the additional information, “1” is set in the port number, and in the multicast relay apparatus 1000 in the port number. “1002”, which is a port to which the multicast relay device 3000 is connected, is set.

  When the multicast relay apparatus 1000 (FIG. 42) receives an IGMP join request message for the multicast group G2 from the multicast receiver 40, the multicast relay apparatus 1000 (FIG. 42) adds the additional information of the own apparatus to the IGMP join request message, Relay to the multicast router 2000 via the port 1001 (arrow F231). Since the IGMP join request message at this time is an IGMP join request message directly transmitted from the multicast receiver 40, no additional information is set in the IGMP join request message. Therefore, the multicast relay apparatus 1000 does not create a duplicate entry for this IGMP join request message.

  FIG. 43C is an explanatory diagram showing a format of an IGMP join request message that is transmitted from the multicast receiver 40 and then added with additional information by the multicast relay apparatus 1000 and relayed to the multicast router 2000. “IP_addr1” that is an IP address set in the multicast relay apparatus 1000 is set in the VLAN_IP address of the additional information, “1” is set in the port number, and in the multicast relay apparatus 1000 in the port number. “1003” which is a port to which the receiver 40 is connected is set.

  The multicast router 2000 creates a duplicate entry based on the IGMP join request message to which the additional information is added (arrow F229, 231).

  FIG. 44 is an explanatory diagram of a duplicate entry table. FIG. 44A is an explanatory diagram showing a replication entry table T1430 generated in the multicast relay apparatus 1000. The multicast relay apparatus 1000 generates a duplicate entry table T1430 based on the IGMP join request message received from the multicast relay apparatus 3000. Therefore, the duplicate entry table T1430 has contents corresponding to the IGMP join request message shown in FIG.

  FIG. 44B is an explanatory diagram showing a replication entry table T2430 generated in the multicast router 2000. The multicast router 2000 generates a duplicate entry table T2430 based on the IGMP join request message received from the multicast relay apparatus 1000. Therefore, the duplicate entry table T2430 has contents corresponding to the IGMP join request message shown in FIGS. 43 (B) and 43 (C).

  FIG. 45 is an explanatory diagram showing a processing flow when the multicast relay apparatus 1000 receives an IGMP join request message to which additional information has been added by the multicast relay apparatus 3000. In the multicast relay apparatus 1000, the VLAN set in the duplicate entry creation information D1230 is registered in the VLAN registration slip 1310. Therefore, the multicast packet relay processing unit 1300 notifies the IGMP packet request message received from the replication entry target VLAN to the control packet transmission / reception processing unit 1420 of the replication entry processing unit 1400 (arrow F228).

  In the multicast relay apparatus 1000, the VLAN set in the IGMP snooping information D1220 is registered in the VLAN registration table T1310. Therefore, the multicast packet relay processing unit 1300 also notifies the control packet transmission / reception processing unit 1120 of the IGMP snooping processing unit 1100 of the IGMP join request message received from the VLAN targeted for IGMP snooping processing (arrow F228).

  The control packet transmission / reception processing unit 1420 of the duplicate entry processing unit 1400 analyzes the notified IGMP join request message 228 and acquires the VLAN_IP address included in the additional information.

  The control packet transmission / reception processing unit 1420 of the replication entry processing unit 1400 determines whether the received VLAN and the acquired VLAN_IP address are registered in the replication entry management table T1410 (arrow F328). Generates a duplicate entry. Specifically, the replication entry is generated by the VLAN name (VLAM101) of the VLAN that received the IGMP join request message, the IP address (IP_addr2) of the VLAN, and the multicast group address (G1) obtained from the notified IGMP join request message. This is done by registering the port information (3003) included in the additional information in the duplicate entry table T1430 (arrow F329).

  When the control packet transmission / reception processing unit 1120 of the IGMP snooping processing unit 1100 receives the notification of the IGMP join request message, the control packet transmission / reception processing unit 1120 searches for the duplicate entry information D1116 in the VLAN management table T1111 and the VLAN that received the IGMP join request message It is determined whether or not it is a target (arrow F330).

  When the VLAN is a copy entry target, the additional information processing unit 1121 of the control packet transmission / reception processing unit 1120 includes the IP address (IP_addr1) of the VLAN and the reception port of the multicast relay apparatus 1000 in the IGMP join request message. 1002 is added as additional information and relayed to the multicast router port 1001 (arrow F229). At this time, if additional information has already been added, the additional information processing unit 1121 deletes the added additional information and overwrites the newly acquired additional information.

  Through the above processing, the entry of the multicast relay device 3000 is copied to the multicast relay device 1000, and the entry of the multicast relay device 1000 is copied to the multicast router 2000. When the multicast relay apparatus 3000 recovers from the failure, the entry before the failure can be reproduced by making a distribution entry distribution request to the multicast relay apparatus 1000. Further, when the multicast relay apparatus 1000 recovers from the failure, the entry before the failure can be reproduced by making a distribution entry distribution request to the multicast router 2000.

  As described above, in the third embodiment, the same effect as in the first embodiment is produced, and a duplicate entry is generated in the adjacent upstream device. Therefore, even if the multicast relay device has a multi-stage configuration, the failure can be quickly detected. You can recreate the entry before it occurs.

  In the first to third embodiments, the IGMP connection request / removal request message relay process is used to cause a device connected upstream to create a duplicate entry. Therefore, an IGMP control packet for duplicating the entry is used. There is also an advantage that the number of notification processes does not increase.

D. Fourth embodiment:
FIG. 46 is an explanatory diagram showing a network configuration adopting the redundant configuration in the fourth embodiment. In this embodiment, in a network configuration adopting a redundant configuration, it is possible to quickly restore multicast communication when an active device becomes a failure and a standby device becomes a new operating state.

  In this embodiment, the multicast relay device 4000A and the multicast relay device 4000B have a redundant configuration. In the example illustrated in FIG. 46, the multicast relay device 4000A is in an operating state, and the multicast relay device 4000B is in a standby state. In the configuration of the present embodiment, when a failure occurs in the multicast relay device 4000A, the multicast relay device 4000B switches to an operating state. In this embodiment, duplicate entries are generated in the multicast relay device 4000A and the multicast relay device 4000B.

  The multicast relay apparatuses 3000, 4000A, and 4000B use the IGMP snooping function. Since the multicast relay device 3000 notifies the upstream device of the IGMP join / leave request message, the ports 3002 and 3003 to which the upstream multicast relay device is connected are set as multicast router ports.

  The multicast relay apparatus 1000 sends multicast packets to both of the multicast relay apparatuses 4000A and 4000B so that multicast communication can be performed immediately even if one of the multicast relay apparatuses 4000A and 4000B having a redundant configuration is in an operating state. Relaying. Therefore, the multicast relay apparatus 1000 does not use the IGMP snooping function.

  A state before the IGMP join request message is transmitted from the multicast receiver 20 in the network configuration of the present embodiment will be described. A multicast packet transmitted from the multicast sender 10 is relayed to the multicast relay apparatus 1000. Since multicast relay apparatus 1000 does not use the IGMP snooping function, the multicast packet transmitted from multicast sender 10 is relayed to ports 1002 and 1003 (arrow F109).

  At this time, since multicast relay apparatus 4000B is in a standby state, it discards even if a multicast packet is received. Since multicast relay apparatus 4000A has not received the IGMP join request message from the multicast receiver, multicast relay apparatus 4000A does not relay the received multicast packet anywhere.

  Next, a state when an IGMP join request message is transmitted from the multicast receiver 20 will be described. The IGMP join request message transmitted from the multicast receiver 20 is relayed to the multicast relay device 3000. The multicast relay apparatus 3000 relays the IGMP join request message to the multicast router ports 3002 and 3003 according to the IGMP snooping function, and relays it to the multicast relay apparatuses 4000A and 4000B (arrow F232).

  At this time, since multicast relay apparatus 4000B is in a standby state, it discards it even if it receives an IGMP join request message. When the IGMP join request message is relayed to the multicast relay apparatuses 3000 and 4000A, as in the first embodiment, the multicast relay apparatuses 3000 and 4000A respectively create entry management tables based on the IGMP join request message, Multicast communication is started (arrow F110).

  FIG. 47 is an explanatory diagram showing a state when a failure has occurred in the multicast relay device 4000A in the operating state. When a failure occurs in the multicast relay device 4000A, the standby multicast relay device 4000B enters an operational state. At this time, since multicast relay apparatus 4000B is in an initial state and does not receive an IGMP join request message, it does not relay anywhere even if a multicast packet is received (arrow F111).

  FIG. 48 is an explanatory diagram showing how the replication entry table for the multicast relay device 4000A in the operating state is generated in the multicast relay device 4000B in the standby state. In this embodiment, by designating an arbitrary port, it is possible to generate a replication entry in an arbitrary apparatus, not limited to an apparatus connected upstream such as a multicast router. This duplicate entry generation process is executed before a failure occurs in the multicast relay apparatus 4000A. In the configuration of this embodiment, the multicast relay device 4000A is in the operating state and the multicast relay device 4000B is in the standby state.

  When the multicast relay device 4000A in operation receives the IGMP join request / leave request message transmitted from the multicast receiver 20 (arrow F233), additional information is added to the IGMP join request / leave request message, and an arbitrary port Relay to the designated port 4003A (arrow F234). Multicast relay apparatus 4000B creates duplicate entry table T4430 based on the additional information of the received IGMP join request / leave request message.

  When a failure occurs in the multicast relay device 4000A, the multicast relay device 4000B changes from the standby state to the operational state, and learns the replication entry table 4430 created in the multicast relay device 4000B as its own entry. In this way, multicast relay apparatus 4000B can immediately perform multicast communication before failure occurs.

  FIG. 49 is an explanatory diagram showing the internal configuration of the multicast relay apparatus 4000A. The multicast relay device 4000A has a redundant configuration with the multicast relay device 4000B, and in order to create a mutual replication entry, the replication entry information D4223, the replication entry generation information D4230, and the IGMP snooping process in the device configuration definition unit 4200 The copy entry information D4116 in the unit 4100, an additional information processing unit 4121, and a copy entry processing unit 4400 are included. Multicast relay apparatus 4000A includes an operation state determination unit 4500 that determines whether the own apparatus is in an operation state or a standby state. The configuration of multicast relay apparatus 4000B is the same as that of multicast relay apparatus 4000A.

  FIG. 50 is an explanatory diagram showing various information and contents of various tables in multicast relay apparatus 4000A. FIG. 50A is an explanatory diagram showing the VLAN information D4210A of the multicast relay apparatus 4000A. “VLAN100” is set in the VLAN name field 4211A of the VLAN information D4210A, “IP_addr1” that is the IP address set in the VLAN 100 of the multicast relay apparatus 4000A is set in the IP address field 4212A, and the port information field 4213A. “4001A, 4002A, 4003A” is set.

  FIG. 50 (B) is an explanatory diagram showing IGMP snooping information D4220A of multicast relay apparatus 4000A. Connection port information is newly provided in the duplicate entry information field 4223A of the IGMP snooping information D4220A. In this connection port information, a port to which a multicast relay device that generates a duplicate entry is connected is set. In this embodiment, “4003A”, which is a port to which the multicast relay device 4000B is connected, is set.

  For the other fields of the IGMP snooping information D4220A, “VLAN100” is set in the VLAN name field 4221A, and “YES” is set in the duplicate entry information field 4223A. Note that nothing is set in the multicast router port field 4222A because it is not necessary to set in the present embodiment.

  FIG. 50C is an explanatory diagram showing the VLAN management table T4111A of the multicast relay apparatus 4000A. The VLAN management table T4111A is generated based on the VLAN information D4210A and the IGMP snooping information D4220A.

  FIG. 50D is an explanatory diagram showing the duplicate entry generation information D4230A of the multicast relay apparatus 4000A. “VLAN100” is set in the VLAN name field 4231A of the duplicate entry creation information D4230A, “IP_addr1” that is the IP address set in the VLAN 100 of the multicast relay device 4000B is set in the IP address field 4232A, and the port information field 4233A. Is set to “4003A”, which is a port to which the multicast relay device 4000B is connected. Since the multicast relay device 4000A and the multicast relay device 4000B have a redundant configuration, the IP addresses set in the VLANs 100 of the multicast relay devices 4000A and 4000B are the same.

  FIG. 50E is an explanatory diagram showing a replication entry management table 4410A of the multicast relay apparatus 4000A. The contents of the duplicate entry management table 4410A are the same as the contents of the duplicate entry generation information D4230A described above.

  FIG. 51 is an explanatory diagram showing various information and the contents of various tables in the multicast relay apparatus 4000B. FIG. 51A is an explanatory diagram showing the VLAN information D4210B of the multicast relay apparatus 4000B. “VLAN100” is set in the VLAN name field 4211B of the VLAN information D4210B, “IP_addr1” that is the IP address set in the VLAN 100 of the multicast relay device 4000B is set in the IP address field 4212B, and the port information field 4213B is set in the port information field 4213B. “4001B, 4002B, 4003B” is set.

  FIG. 51 (B) is an explanatory diagram showing IGMP snooping information D4220B of multicast relay apparatus 4000B. Connection port information is newly provided in the replication entry information field 4223B of the IGMP snooping information D4220B. In this connection port information, a port to which a multicast relay device that generates a duplicate entry is connected is set. In this embodiment, “4003B”, which is a port to which the multicast relay apparatus 4000A is connected, is set.

  For the other fields of the IGMP snooping information D4220B, “VLAN100” is set in the VLAN name field 4221B, and “YES” is set in the duplicate entry information field 4223B. Note that nothing is set in the multicast router port field 4222B because it is not necessary to set in the present embodiment.

  FIG. 51C is an explanatory diagram of the VLAN management table T4111B of the multicast relay apparatus 4000B. The VLAN management table T4111B is generated based on the VLAN information D4210B and the IGMP snooping information D4220B.

  FIG. 51 (D) is an explanatory diagram showing copy entry generation information D4230B of multicast relay apparatus 4000B. “VLAN100” is set in the VLAN name field 4231B of the duplicate entry creation information D4230B, “IP_addr1” that is the IP address set in the VLAN 100 of the multicast relay apparatus 4000A is set in the IP address field 4232B, and the port information field 4233B. Is set to “4003B” which is a port to which the multicast relay device 4000A is connected. As described above, since multicast relay device 4000A and multicast relay device 4000B have a redundant configuration, the IP addresses set in VLANs 100 of multicast relay devices 4000A and 4000B are the same.

  FIG. 51E is an explanatory diagram showing a replication entry management table T4410B of the multicast relay apparatus 4000B. The contents of the duplicate entry management table T4410B are the same as the contents of the duplicate entry generation information D4230B described above.

  FIG. 52 is an explanatory diagram showing the flow of processing when the multicast relay device 4000A in the operating state receives an IGMP join request message. When receiving the IGMP join request message, the multicast relay apparatus 4000A adds additional information to the IGMP join request message and relays the IGMP join request message with the added information added to the multicast relay apparatus 4000B. Details will be described below.

  In the present embodiment, the VLAN set in the IGMP snooping information D4220 and the duplicate entry generation information D4230 is registered in the VLAN registration table T4310 of the multicast packet relay processing unit 4300. Therefore, when multicast relay apparatus 4000A receives the IGMP join request message from multicast relay apparatus 3000, the IGMP join request message is notified to duplicate entry processing section 4400 and IGMP snooping processing section 4100 (arrow F233).

  At this time, since no additional information is added to the IGMP join request message relayed from the multicast relay apparatus 3000, even if the IGMP join request message is notified to the duplicate entry processing unit 4400, the additional information for the multicast relay apparatus 3000 is not received. No duplicate entry table is created.

  On the other hand, when the IGMP join request message is notified to the control packet transmission / reception processing unit 4120 of the IGMP snooping processing unit 4100, the control packet transmission / reception processing unit 4120 analyzes the IGMP join request message, as in the first embodiment, It is determined whether or not it is an IGMP join request message to be copied (arrow F331). If the IGMP subscription request message is to be copied, the additional information processing unit 4121 adds additional information to the IGMP subscription request message, as in the first embodiment.

  Next, the control packet transmission / reception processing unit 4120 acquires connection port information from the replication entry information D4116 of the VLAN management table 4111, and relays an IGMP join request message with additional information added to the acquired port. In this embodiment, since the connection port information is 4003A, the control packet transmission / reception processing unit 4120 relays an IGMP join request message with additional information added to the port 4003A (arrow F234). Note that the relay process is also performed for the IGMP leave request message by the same method as the IGMP join request message.

  FIG. 53 is an explanatory diagram showing the flow of processing when the multicast relay device 4000B in a standby state receives an IGMP join request message to which additional information has been added. In the case where the multicast relay device 4000B is in the standby state, when the duplicate entry creation information D4230 is set, the duplicate entry processing unit 4400 acquires the duplicate entry creation information D4230 (arrow F332) and creates the duplicate entry management table T4410. Then, the VLAN for which the replication entry table is to be generated is registered in the VLAN registration table T4310 (arrow F333). The duplicate entry processing unit 4400 also registers the connection port information set in the duplicate entry creation information D4230 in the VLAN registration table T4310.

  With the above processing, even when the multicast relay device 4000B is in a standby state, the IGMP entry request / leave request message received from the registered port is notified to the duplicate entry processing unit 4400 (arrow F234). The duplicate entry processing unit 4400 analyzes the notified IGMP join request message by the same method as in the first embodiment (arrow F334), and creates a duplicate entry table T4430 (arrow F335).

  FIG. 54 is an explanatory diagram showing a flow of processing when a failure occurs in the multicast relay apparatus 4000A and the multicast relay apparatus 4000B enters an operating state. When the multicast relay device 4000B enters the operation state, the operation state determination unit 4500 of the multicast relay device 4000B confirms the copy entry generation information D4230 (arrow F336), and if at least one copy entry generation information entry is set. Then, the replication entry processing unit 4400 is notified that the operation state has changed (hereinafter also referred to as “operation state change notification”) (arrow F337).

  Receiving the operation state transition notification, the replication entry processing unit 4400 refers to the VLAN information D4210 (arrow F338a), and acquires the VLAN name and IP address set in the own device (arrow F338b). The control packet transmission / reception processing unit 4420 in the replication entry processing unit 4400 determines whether or not the acquired VLAN name and IP address are registered in the replication entry management table T4410 (arrow F339).

  When the VLAN information is registered in the duplicate entry management table T4410, the control packet transmission / reception processing unit 4420 searches for the duplicate entry from the created duplicate entry table T4430 (arrow F340), and joins the IGMP. Create a request message. The control packet transmission / reception processing unit 4420 adds additional information to the created IGMP join request message in the same manner as in the first embodiment, and transmits it to the control packet transmission / reception processing unit 4120 in the IGMP snooping processing unit 4100 (arrow F235). ).

  The control packet transmission / reception processing unit 4120 in the IGMP snooping processing unit 4100 analyzes the IGMP join request message to which the additional information is added (arrow F341), and uses the same method as in the first embodiment in the operation apparatus before the occurrence of the failure. Get multicast group address and port information. The control packet transmission / reception processing unit 4120 notifies the entry generation unit 4130 of the acquired port information, multicast group address, and VLAN name (arrow F342) and registers them in the entry management table T4131. The entry generation unit 4130 notifies the multicast packet relay processing unit 4300 of the information registered in the entry management table T4131, and registers the information in the multicast packet relay control table T4320 (arrow F343). In this way, multicast communication is resumed in multicast relay device 4000B that is a new operation device.

  FIG. 55 is a flowchart showing a processing flow of the operation state determination unit 4500 when the multicast relay apparatus 4000B transitions from the standby state to the operation state. The operation state determination unit 4500 in the multicast relay device 4000B monitors the operation state of the own device (step S6800). When the operation state determination unit 4500 detects that the own device has changed to the operation state during monitoring (step S6801: Yes), the operation state determination unit 4500 confirms the replication entry generation information D4230 (step S6802).

  When even one copy entry generation information D4230 is set (step S6803: Yes), the operation state determination unit 4500 notifies the copy entry processing unit 4400 of an operation state transition (step S6805). When the replication entry generation information D4230 is not set (step S6803: No), the operation state determination unit 4500 ends the process (step S6804).

  FIG. 56 is a flowchart showing a flow of processing from when multicast relay apparatus 4000B receives an operation state transition notification from the operation state determination unit to when an IGMP join request message is transmitted. The duplicate entry processing unit 4400 in the multicast relay device 4000B monitors the operation state transition notification (step S6900). When the duplicate entry processing unit 4400 detects the operational state transition notification (step S6901: Yes), the control packet transmission / reception processing unit 4420 in the duplicate entry processing unit 4400 uses the VLAN information 4210 in the device configuration definition unit 4200 to determine the VLAN of the own device. Information is acquired (step S6902). The control packet transmission / reception processing unit 4420 determines whether or not the acquired VLAN information is set in the replication entry management table T4410 (step S6903).

  When the VLAN is not set (step S6907: No), the control packet transmission / reception processing unit 4420 ends the process (step S6908). On the other hand, if the VLAN is set (step S6904: Yes), the control packet transmission / reception processing unit 4420 searches the duplicate entry table T4430 to determine whether or not an entry corresponding to the VLAN exists ( Step S6906).

  When the VLAN entry exists (step S6907: Yes), the control packet transmission / reception processing unit 4420 acquires the multicast group and port information from the VLAN entry of the replication entry table T4430, and adds the additional information to the IGMP. A subscription request message is created and transmitted to the IGMP snooping processing unit 4100 of the own device (step S6909).

  On the other hand, if the VLAN entry does not exist, the control packet transmission / reception processing unit 4420 ends the process (step S6908). Note that the entry setting process to the own apparatus after the control packet transmission / reception processing unit 4120 in the IGMP snooping processing unit 4100 of the multicast relay apparatus 4000B receives the IGMP join request message to which the additional information is added is the first embodiment. Since the processing is the same as that shown in FIG.

  As described above, in the fourth embodiment, it is possible to create a duplicate entry not only for a multicast router but also for an arbitrary device. Further, in this embodiment, a function for generating a duplicate entry for each device having a redundant configuration is implemented, and a duplicate entry is created in a multicast relay device in a standby state. Therefore, when the standby multicast relay device is switched to the operational state, the multicast relay device that is in the operational state quickly changes the entry of the old operational device based on the duplicate entry generated in the local device. Can be set to

  Even when the configuration is not redundant, a device for generating a replication entry is selected from any devices, a port to which the device is connected is specified, and a replication entry generation process is performed on the device using the port. , And when the failure is recovered, the entry before the occurrence of the failure can be quickly reproduced by requesting the apparatus to distribute the duplicate entry.

  After the multicast relay device 4000B transitions to the operation state, an IGMP inquiry message is transmitted from the IGMP querier. When the multicast receiver 20 responds to the IGMP inquiry message with the IGMP join request message, the multicast relay device 4000B receives the IGMP join request message. In this way, multicast communication is resumed.

  As in this comparative example, the multicast relay device 4000B transmits an IGMP inquiry message after transition from the standby state to the operational state, receives a response to the IGMP join request message, and resumes multicast communication. It takes time until the restart, and further, there is an effect that the communication band is compressed due to the concentrated reception of the IGMP join request message and the receiving process of the IGMP join request message becomes heavy.

E. Variations:
The present invention is not limited to the above-described examples and embodiments, and can be implemented in various modes without departing from the gist thereof. For example, the following modifications are possible.

Modification 1:
In the above embodiments, the multicast relay apparatus having the IGMP snooping function has been described as an example. However, the present invention can also be applied to a multicast relay apparatus having the MLD snooping function.

Modification 2:
In each of the embodiments described above, the additional information processing unit transmits the additional information by adding it to the IGMP join request / leave request message. Further, the additional information may be transmitted when the usage rate of the communication band is lower than a predetermined value or when a command for instructing transmission of the additional information is received.

Modification 3:
In each of the above-described embodiments, the multicast relay device requests distribution of the duplicate entry at the timing of recovery from the failure. Instead, the multicast relay device performs predetermined processing such as when the system is started or when a communication error is detected. It is also possible to request the distribution of the duplicate entry at the timing.

Modification 4:
In the second embodiment, one multicast relay device creates a duplicate entry for two multicast relay devices. Instead, one multicast relay device includes three or more multicast relay devices. A duplicate entry may be created.

Modification 5:
A multicast relay system may be constructed by combining some or all of the configurations of the multicast relay devices in the above embodiments.

Modification 6:
In the above embodiment, a part of the functions realized by software may be realized by hardware, or a part of the functions realized by hardware may be realized by software.

DESCRIPTION OF SYMBOLS 10 ... Multicast sender 20 ... Multicast receiver 30 ... Multicast receiver 40 ... Multicast receiver 50 ... Multicast receiver 1000 ... Multicast relay apparatus 1001 ... Port 1002 ... Port 1003 ... Port 1004 ... Port 1100 ... IGMP snooping processing part 1120 ... Control packet transmission / reception processing unit 1121 ... Additional information processing unit 1130 ... Entry generation unit 1200 ... Device configuration definition unit 1300 ... Multicast packet relay processing unit 1400 ... Duplicate entry processing unit 1420 ... Control packet transmission / reception processing unit 2000 ... Multicast router 2001 ... Port 2002 ... Port 2003 ... Port 2100 ... Device configuration definition unit 2200 ... Multicast protocol processing unit 2300 ... Multicast packet relay Processing unit 2400 ... Duplicate entry processing unit 2420 ... Control packet transmission / reception processing unit 3000 ... Multicast relay device 3001 ... Port 3002 ... Port 3003 ... Port 4000A ... Multicast relay device 4000B ... Multicast relay device 4003A ... Port 4003B ... Port 4100 ... IGMP snooping processing 4120 ... Control packet transmission / reception processing unit 4121 ... Additional information processing unit 4130 ... Entry generation unit 4200 ... Device configuration definition unit 4300 ... Multicast packet relay processing unit 4400 ... Duplicate entry processing unit 4420 ... Control packet transmission / reception processing unit 4500 ... Operation status determination Part T2430 ... Duplicate entry table T1131 ... Entry management table D2120 ... Duplicate entry creation information T2410 ... Duplicate entry management table T1320 ... Ma Multicast packet relay control table D2110 ... Multicast information T2320 ... Multicast packet relay control table D1116 ... Duplicate entry information D1230 ... Duplicate entry generation information T1410 ... Duplicate entry management table T1430 ... Duplicate entry table T4430 ... Duplicate entry table D4223 ... Duplicate entry information D4230 ... Duplicate entry creation information D4116 ... Duplicate entry information D4230A ... Duplicate entry creation information D4230B ... Duplicate entry creation information T4410A ... Duplicate entry management table T4410B ... Duplicate entry management table T4410 ... Duplicate entry management table T4131 ... Entry management table T4320 ... Multicast packet relay control table

Claims (10)

A first multicast relay device that relays a multicast packet to a line connected to a recipient of the multicast packet in a logically independent network, and a second multicast relay device that is connected to the first multicast relay device A multicast relay system comprising:
The first multicast relay device is:
A relay processing unit that performs relay processing of the multicast packet based on relay control information including information of a line to which a recipient of the multicast packet is connected;
A transmission unit that transmits information for creating a copy, which is information used to create a copy of the relay control information, to the second multicast relay device;
A distribution request unit that requests distribution of a copy of the relay control information to the second multicast relay apparatus at a predetermined timing;
A setting unit configured to set the relay control information based on a copy of the relay control information distributed from the second multicast relay device,
The second multicast relay device is
A copy creation unit for creating a copy of the relay control information based on the copy creation information transmitted from the first multicast relay device;
A multicast relay system comprising: a distribution unit that distributes a copy of the relay control information to the first multicast relay device in response to a request from the first multicast relay device. The multicast relay system according to claim 1, further comprising:
A plurality of the first multicast relay devices;
The second multicast relay device is connected to the plurality of multicast relay devices;
The copy creating unit of the second multicast relay device is configured to use the first multicast relay device based on the copy creation information transmitted from each of the plurality of connected first multicast relay devices. A copy of the relay control information for
The distribution unit of the second multicast relay apparatus distributes a copy of the relay control information to the first multicast relay apparatus in response to a request from the first multicast relay apparatus. Multicast relay system. The multicast relay system according to claim 1, further comprising:
A third multicast relay device connected to the second multicast relay device;
The second multicast relay device further includes:
A relay processing unit that performs relay processing of the multicast packet based on second relay control information including information of a line to which a receiver of the multicast packet is connected;
A transmitter for transmitting second copy creation information, which is information used for creating a copy of the second relay control information, to the third multicast relay device;
A distribution requesting unit that requests distribution of a copy of the second relay control information to the third multicast relay device at a predetermined timing;
A setting unit configured to set the second relay control information based on a copy of the second relay control information distributed from the third multicast relay device;
The third multicast relay device is
A copy creation unit for creating a copy of the second relay control information based on the second copy creation information transmitted from the second multicast relay device;
A multicast relay system comprising: a distribution unit that distributes a copy of the second relay control information to the second multicast relay device in response to a request from the second multicast relay device. The multicast relay system according to any one of claims 1 to 3,
The multicast relay system, wherein the transmission unit transmits the copy creation information in addition to a join request message or a withdrawal request message transmitted from the multicast receiver. The multicast relay system according to any one of claims 1 to 4, wherein
The distribution request unit transmits an inquiry message to the second multicast relay device as the distribution request,
In response to the inquiry message, the distribution unit adds a copy of the relay control information to a subscription request message and distributes it. The multicast relay device according to any one of claims 1 to 5,
A multicast relay system, wherein the relay control information distributed by the distribution unit includes a plurality of pieces of line information. A first multicast relay device that relays a multicast packet to a line connected to a recipient of the multicast packet in a logically independent network, and a second multicast relay device that is connected to the first multicast relay device A multicast relay system comprising:
The first multicast relay device is:
A relay processing unit that performs relay processing of the multicast packet based on first relay control information including information of a line to which a receiver of the multicast packet is connected;
A transmission unit that transmits information for creating a copy, which is information used to create a copy of the first relay control information, to the second multicast relay device;
The second multicast relay device is
A copy creation unit for creating a copy of the first relay control information based on the copy creation information transmitted from the first multicast relay device;
An operational state detector that detects that the second multicast relay device has transitioned from a standby state to an operational state;
When the operation state detection unit detects that the second multicast relay device has transitioned from the standby state to the operation state, the second multicast relay device in the second multicast relay device is based on the copy of the first relay control information. A setting unit for setting the relay control information of
A multicast processing system, comprising: a relay processing unit that performs relay processing of the multicast packet based on the second relay control information. Information provided in a multicast relay device for relaying a multicast packet to a line to which a multicast packet receiver is connected in a logically independent network, the information about the line to which the multicast packet receiver is connected. A method for restoring relay control information including:
The multicast relay device transmitting copy creation information, which is information used to create a copy of the relay control information, to an external multicast relay device;
The external multicast relay device creating a copy of the relay control information based on the copy creation information transmitted from the multicast relay device;
The multicast relay device requesting the external multicast relay device to distribute the copy of the relay control information at a predetermined timing;
The external multicast relay device delivering a copy of the relay control information to the multicast relay device in response to a request from the multicast relay device;
The multicast relay device includes a step of setting the relay control information based on a copy of the relay control information distributed from the external multicast relay device. A multicast relay device that relays a multicast packet to a line to which a receiver of the multicast packet is connected in a logically independent network,
A relay processing unit that performs relay processing of the multicast packet based on relay control information including information of a line to which a recipient of the multicast packet is connected;
A transmission unit for transmitting information for creating a copy, which is information used for creating a copy of the relay control information, to an external multicast relay device connected to the multicast relay device;
A distribution request unit that requests distribution of a copy of the relay control information at a predetermined timing to the external multicast relay device;
A multicast relay device comprising: a setting unit configured to set the relay control information based on a copy of the relay control information distributed from the external multicast relay device. A multicast relay device connected to an external multicast relay device that relays a multicast packet to a line connected to a receiver of the multicast packet in a logically independent network,
A copy creation unit for creating a copy of the relay control information based on the copy creation information transmitted from the external multicast relay device;
A multicast relay apparatus comprising: a distribution unit that distributes a copy of the relay control information to the external multicast relay apparatus in response to a request from the external multicast relay apparatus.

Images