JP2003283547A - Message relay node - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a message relay node in which the transfer of a message created at an application level is controlled on a network. <P>SOLUTION: The message relay node is provided with a receiving part for receiving a message created at the application level to be transferred on the network, a decision part for deciding whether or not the message received by the receiving part is to be transferred based upon decision information inserted in the message, and a transfer processing part for transferring the message when the decision part decides the message to be transferred and for stopping the transfer of the message when the decision part decides the message not to be transferred. <P>COPYRIGHT: (C)2004,JPO

Description

Detailed Description of the Invention

[0001]

TECHNICAL FIELD The present invention relates to a network system such as the Internet or an intranet,
The present invention relates to technology for controlling transfer of messages generated in an application layer.

[0002]

2. Description of the Related Art In a packet transfer type network represented by the Internet or an intranet, data to be transferred is divided into a plurality of packets, and each packet reaches a destination via a relay node. In such a network, a packet may fall into a loop state when there is a mistake in setting a relay node and / or a flaw in an algorithm for determining a relay destination. The loop state is a state in which a packet continues to travel between the same nodes without reaching the destination node and remains in the network forever. In recent years, voice transfer has been realized by a packet transfer type network. Voice transfer is real-time
(Real-time property) is an important application. In the application service where real-time property is important, the packet must reach the destination within a predetermined time. However, if the number of relay nodes that relay the packet is too large, the packet may not arrive at the destination within a predetermined time due to an increase in delay.
Packets that do not arrive within the predetermined time are useless.

In order to deal with the above-mentioned problem, a packet has a "time to live" (TTL: time) in an IP network.
A method of setting (to live) is known. FIG. 13 shows T
It is explanatory drawing of the prior art example which used TL. In FIG.
The client transfers the data to the server via the IP network. When transferring a plurality of packets into which the data has been divided to the IP network, the IP router incorporates the data indicating the TTL into the IP header for each packet. The data indicating the TTL is the maximum allowable number of relay nodes through which the packet can pass (the maximum allowable number of transit nodes) or the upper limit time (valid upper limit time) at which the survival of the packet is allowed. When the maximum allowable number is set for a packet, the maximum allowable number is decremented by 1 each time the packet passes through a relay node (router) in the network, and when the maximum allowable number becomes 0, the packet Is discarded at the relay node that received the packet. On the other hand, when the valid upper limit time is set in the packet, each relay node that receives the packet determines whether or not the current time exceeds the valid upper limit time. ,
The packet is dropped. Thus, when the relay node receives a packet, the TT embedded in the packet is received.
The data indicating L is inspected, and when the time to live of the packet is exhausted, the packet is discarded without being transferred to the next node. In this way, packets that are no longer needed due to a loop or when a predetermined time has passed
It can be removed from the P network. by this,
The transfer processing of unnecessary packets is suppressed. Therefore, IP
The burden on relay nodes in the network is reduced and resources are effectively used.

[0004]

The method of setting the time to live in the IP header of a packet can eliminate problems at the IP level (packet level) by discarding unnecessary packets.

In recent years, the performance of packet processing devices has improved,
Various application services have become possible, and application services with different properties have come to coexist on the same network. Different application services use different length messages, and different assumed lifetimes for the messages. For example, in a corporate network, in an HTTP protocol-based system for exchanging document data, it is assumed that a document to be exchanged is temporarily stored in a relay node (proxy node) in the middle of a transfer route and then transmitted. Therefore, the relay node may hold the message for a certain period of time. This message is divided into a plurality of packets and transferred in the network. When the object represented by this message is, for example, a slip of bid data in an auction, a bid has a fixed deadline, so this message also has an expiration date. If the message expires,
It is meaningless to forward the message to further relay nodes. In addition, it consumes extra resources in the network. As described above, in order to properly operate the application service, the length of the message life time required for properly operating the application service differs for each application service. On the other hand, the packet survival time is set uniformly for the entire IP network or for each IP connection.

Therefore, in the case where a plurality of application services are realized between a certain communication terminal through one IP network, when the survival time is set in the packet,
The message of each application service is transferred as a packet in which a uniform lifetime is set regardless of the category of application service. Therefore, if the packet survival time does not match the message survival time that is appropriate for each application service, the following problem may occur. That is, the application service that can be operated with the message life time longer than the packet life time is operated under severe conditions. On the other hand, in an application service that requires operating with a message life time shorter than the packet life time, the packet life time does not substantially function, and the message receiver must receive an unnecessary message.

An object of the present invention is to provide a message relay node capable of controlling message transfer on an application level.

[0008]

The present invention adopts the following configurations in order to achieve the above-mentioned object.

That is, the present invention is directed to a message relay node.
(14) which is a reception means (21) for receiving a message generated at the application level and transferred through the network, and the message is transferred based on the judgment information inserted in the message received by the reception means. Determining means (22, 23) for determining whether or not, and transferring the message when the determining means determines to transfer the message, and transferring the message when the determining means determines not to transfer the message Processing means to stop
(21) and are included.

In the present invention, a message is data transferred from a source to a destination through a network and is generated at an application level. The message received by the message relay node includes determination information inserted at the application level. The application level corresponds to the application layer of the OSI reference model.

According to the present invention, the message relay node receives the message generated at the application level, determines whether or not to transfer the message based on the determination information inserted in the message, and Transfer processing is performed according to the determination result. Therefore, message transfer control in the network can be performed based on the content of the message.

[0012]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below with reference to the drawings. The configurations in the embodiments are exemplifications, and the present invention is not limited to the scope of the configurations in the embodiments.

<Outline of the Present Invention> First, an outline of the present invention will be described. FIG. 1A is a diagram showing an example of a network system (message transfer system) to which a message relay node according to the present invention is applied, and FIG.
It is explanatory drawing of the message transferred from the client 11 to the server 12 in the network system shown to (A).

In FIG. 1A, a client 11 for sending a message and a server 1 corresponding to the destination of the message are shown.
2 and a message router 13 that transfers messages,
A network system including an IP network is shown. The IP network includes a plurality of relay nodes 14 that relay messages (the relay node 14A, 1 in FIG.
4B and 14C are illustrated, and when referring to an arbitrary relay node, “relay node 14” is included). The relay node 14 corresponds to the message relay node according to the present invention.

The message M (message in the present invention) shown in FIG. 1B is sent from the client 11 to the server 12.
Is a chunk of data to send to. The message M is generated by the function that controls the application layer (application level) in the OSI reference model. Further, when the message M is transferred on the IP network, the message is divided into a plurality of blocks, a packet in which a header including information necessary for transfer is added to each divided block is generated, and each packet is It is transferred over the IP network.

The client 11 has a function of generating a message and a function of dividing the message into a plurality of packets and sending them to the server 12. The plurality of packets transmitted from the client 11 are transferred to the message router 13 arranged on the route.

The message router 12 is installed at the entrance of the IP network. The message router 12
When a plurality of packets are received from the client 11, the message M is assembled from these plurality of packets.
Then, the message router 12 inserts the determination information into this message M. After that, the message router 12
Divides the message M in which the determination information is inserted into a plurality of packets and transfers the packets to the IP network.

The determination information is information for determining, when each relay node 14 located on the transfer route of the message M receives the message M, whether or not to transfer this message to the next node (next hop). Is. For example, as the determination information, survival information indicating the survivable range of the message M (hereinafter referred to as “survival time information”) can be applied. The survival time information is, for example, the number of relay nodes 14 through which the message M can pass on the IP network (the maximum allowable number of transit nodes). Further, the upper limit time (effective upper limit time) at which the message M can survive can be adopted as the survival time information. Since this lifetime information indicates the lifetime of the message M generated in the application layer, in this specification, "message TTL (message T
Also called "ime To Live)". FIG. 1A shows the message TT.
An example in which the number “5” of relay nodes 14 through which the message M can pass is set as L is set as a default value of the maximum allowable number as the message TTL.

Each relay node 14 receives a plurality of packets obtained by dividing the message M from the preceding node (the message router 13 or another relay node) (corresponding to receiving means). Then, the relay node 14 assembles the message M from the plurality of packets. Next, the relay node 14 extracts the determination information from the message M. Next, the relay node 14 determines whether or not to transfer the message M to the next node based on the determination information (corresponding to determination means). When the relay node 14 determines to transfer the message M to the next node, the relay node 14 updates the judgment information as necessary and then transfers the message M to the next node (corresponding to a transfer processing unit).

On the other hand, when the relay node 14 determines not to transfer the message M to the next node, it stops the transfer processing of the message M (corresponding to the transfer processing means). At this time, the relay node 14 can also execute a predetermined error process (abnormal condition process). The error process is, for example, a process of discarding (discarding) the message M. Furthermore, the relay node 14 may execute a process of outputting an error notification. The content of the error notification is, for example,
The fact that an error occurred in message transfer, the message was discarded, etc. can be applied. The destination of the error notification is the sender of the message M (client 11), the destination
(Server 12), or both may be used.

In the example shown in FIG. 1A, the relay node 14A
Receives the message M including the message TTL (maximum allowable number “5”) as the determination information from the message router 13. In the relay node 14A, the value of the maximum allowable number is "5", and the passable number of relay nodes remains.
Since it is not 0 or less, it is determined to transfer the message M to the next node. At this time, the relay node 14A
The value of the maximum allowable number is decremented by 1 and updated to "4", then the message M is divided into a plurality of packets, and the relay node 14B
Transfer to. The relay node 14B executes the same process as the relay node 14A and reduces the maximum allowable number by one.
(Updating to “3”), the message M is transmitted to the relay node 14
Transfer to C. In this way, each relay node 14 determines whether or not the value of the maximum allowable number inserted in the message M is greater than 0, and if the value is greater than 0, decrements the maximum allowable number by 1, and then Transfer the message M to the next node.

Therefore, in the case of the message M, if the value of the maximum allowable number is in a range larger than 0, the server 1 corresponding to the destination
Arriving at 2. On the other hand, although not shown in FIG. 1A, when a relay node 14 receives the message M having the maximum allowable number “0”, the transfer of the message M is stopped at the relay node 14. Then, the relay node 14 executes error handling.

FIG. 2 shows an example in which the valid upper limit time is applied to the survival time information instead of the maximum allowable number of transit nodes in the network system shown in FIG. In FIG. 2, the message router 13 inserts the valid upper limit time “1135 (indicates 11:35)” in the message M from the client 11. Upon receiving the message M, each relay node 14 extracts the effective upper limit time, compares the effective upper limit time with the current time, and determines whether the current time is before the effective upper limit time (the current time is the effective upper limit time). If the current time is before the valid upper limit time, the message M is transferred to the next node. When the time to live information is the valid upper limit time, the time to live information is not updated when the message M is transferred. As the current time, an appropriate time regarding the processing of the message M in the relay node 14 can be adopted. For example, the passing time of the message M can be adopted. As the passage time, for example, the time when the message M is received (the reception date and time or the time of the message M) can be applied.

In the example shown in FIG. 2, the transit time at each of the relay nodes 14A and 14B is "1132 (11:32).
Minutes) ”and“ 1134 (11:34) ”, which is before the effective upper limit time“ 1135 (11:35) ”.
Each relay node 14A and 14B is forwarding the message M to the next node. On the other hand, the transit time at the relay node 14C is "1136 (11:36)", which exceeds the valid upper limit time "1135 (11:35)", so the relay node 14C sends the message M Stop the transfer,
The specified error processing is being executed.

In the example shown in FIGS. 1A and 2, the message router 13 inserts the judgment information into the message M.
(Added). Alternatively, the transmission source (client 11) of the message M may generate the message M including the determination information, or may insert the determination information into the generated message M.

According to the present invention, in the client program for transmitting the message M or the message transfer device (message router) on the way, the judgment information (lifetime information) is inserted somewhere in the message data to be transferred, and Mark it so that you can see that is the judgment information (survival time information).

The relay node 14 which has received the message M
The (relay router) analyzes the message M, extracts the survival time information by using the mark as a clue, and inspects and judges it to determine the subsequent processing. At this time, if the time to live information does not violate the specified value, the message is transferred to the next node, and if the time to live information violates the specified value, the message is not transferred and is discarded. Abnormal result
Some error handling processing (error processing) such as outputting (error notification) is executed.

When the maximum allowable number of transit nodes is used as the time-to-live information (message TTL), the maximum allowable number is inserted in the message M and the maximum allowable number is decremented by 1 when passing through the relay node 14. And the relay node 1
When 4 receives the maximum allowable number “0” of the message M, the relay node 14 does not transfer the message M to the next node.

When the valid upper limit time of the message is used as the survival time information (message TTL), the valid upper limit time is inserted in the message M, and each relay node 14
Compares the current time (for example, the reception time of the message M) with the valid upper limit time, and if the current time has passed the valid upper limit time, does not transfer the message M to the next node.

According to the present invention, application level message transfer can be controlled based on the above-described operation. Therefore, it is possible to remove from the network the message M that is unnecessary or falls into a loop state at the application level.

The application level message transfer control according to the present invention can be carried out independently of the packet level transfer control by the TTL of the packet. That is, the message TTL can be applied preferentially to the packet TTL. In this way, message transfer (packet transfer) can be controlled with the message lifetime that is appropriate under the operation of the application service.

<First Embodiment> Next, a first embodiment of the message relay node according to the present invention will be described. Figure 3
1 is a diagram showing a network system to which a first embodiment (relay node 17) of a message relay node according to the present invention is applied. In FIG. 3, the network system is
It includes a client 15 that is the source of the message, a destination node 16 of the message, and an IP network that forwards the message from the client 15 to the destination node 16. The IP network includes a plurality of relay nodes 17 (FIG. 3 illustrates only one relay node 17).

FIG. 4 is a functional block diagram of the client 15 and the relay node 16 shown in FIG. In FIG. 4, the client 15 is a computer having a message communication function using an IP network. For example, the client 15 can be a computer used as a communication terminal such as a personal computer, a workstation, or a mobile computer.

Although not shown, the client 15 is a CP.
U, main memory, auxiliary memory, communication control device, input device (for example, keyboard, mouse), output device (for example, display). The CPU of the client 15 loads various programs (OS and application programs) stored in the auxiliary storage into the main storage and executes them.
As a result, the client 15 has the message creation unit 18, the message TTL insertion unit 19, and the message creation unit 18 as shown in FIG.
And functions as a device including the communication unit 20.

The message creating section 18 is a function realized by executing an application program (client program), and provides the user of the client 15 with an environment for creating a message (a mass of data to be sent to the destination node 16). , Creates a message to be transferred according to the operation of the client 15 by the user.
In this example, the message creating unit 18 uses the XML (Extansi
ble Markup Language) to provide users with a message creation environment. Using the creation environment, the user can create the transfer target data described in XML as shown in FIG. 5A, for example. Data to be transferred as shown in FIG. 5A constitutes a message body (body). The message creating unit 18 gives the created data to the application TTL inserting unit 19.

The message TTL creating unit 19 receives the data (message body) received from the message creating unit 18.
Insert message lifetime information (message TTL) as the determination information. Specifically, a message as shown in FIG. 5 (B) including the data shown in FIG. 5 (A) is created.

In the message shown in FIG. 5B, "<b
"ody>" is the start tag of the message body, and "</ body
> ”Is the end tag of the message body. Therefore, the data sandwiched between the start tag and the end tag of the message body indicates the data to be transferred to the destination node 16. Therefore, in the message creating unit 18, The created data is described between the start tag “<body>” and the end tag “</ body>”.

On the other hand, "<header>" is the start tag of the header of the message, and "</ header>" is the end tag of the header. These start tags “<header>” and end tags “</
Between header> ”, additional information indicating the content of the header of the message is described.

The message TTL inserting section 19 generates data defining the message TTL and inserts it into the header part of the message. In the above message example, “<ttl> 3 </ ttl>” is described (inserted) in the message header as the message TTL. The start tag “<ttl>” and the end tag “</ ttl>” function as marks for extracting the message TTL from the message at the relay node 17. "3" described between the start tag "<ttl>" and the end tag "</ ttl>" is a string representing an integer value indicating the upper limit of the survival time, and the message is a relay node. Indicates that the maximum number of passing nodes (maximum allowable number of passing nodes) is “3”.

Message creating unit 18 and message TT
The L insertion unit 19 creates a message composed of a body and a header with the message TTL inserted in the header. Although the message TTL is inserted in the message header in this example, the message TT is inserted in the relay node 17.
If L is searchable, the message TTL may be inserted into the message body. The message creating unit 18 and the message TTL inserting unit 19 are realized by the function of controlling the application layer of the OSI reference model.

Returning to FIG. 4, the communication section 20 controls the processing relating to the communication protocol of the layer lower than the application layer necessary for sending the message to the IP network. The communication unit 20 mainly performs the following processing.
That is, the communication unit 20 receives the message from the message TTL insertion unit 19, divides the message into a plurality of blocks, generates a packet in which a header suitable for each layer is added to each block, and the generated plurality of packets ( Message) to the IP network. The plurality of packets sent to the IP network are transferred to the destination node 16 via one or more relay nodes 17 according to the destination IP address stored in the IP header.

As described above, the client 15
Has the functions of the client 11 and the message router 13 shown in FIG. In such a case, the message router 13 as shown in FIG. 1 (A) is unnecessary.

Each relay node 17 has a packet relay function.
It is configured by using a computer having a (forwarding function) and a function of determining whether or not to transfer a message to the next node. For example, it is configured by using a router or a server machine capable of executing processing related to a message (processing related to an application layer). The relay node 17 is
For example, as shown in FIG. 4, it includes an IP communication unit 21, a message analysis / transfer unit 22, and a message TTL reference / transfer unit 23. The message analysis / transfer unit 22 and the message TTL reference / transfer unit 23 manage the processing at the application level (application layer), and the IP communication unit 21 manages the processing in layers below the application layer.

The IP communication unit 21 receives the packet arriving at the relay node 17, and based on the destination IP address stored in the IP header of this packet, the IP of the next node (the other relay node 17 or the destination node 16). It manages the function of calculating the address and forwarding the packet to the next node (forwarding function). In addition, the IP communication unit 21
A message is assembled from a plurality of received packets and given to the message analysis / transfer unit 22.

The message analysis / transfer unit 22 analyzes the message (XML data) received from the IP communication unit 21, and tags (message TTL tags: "<ttl>" and "</ </ ttl>" indicating the message TTL from the header of the message. ttl> ”)
Look for. If the message TTL tag is found, the message analysis / transfer unit 22 extracts the maximum allowable number defined between the start tag “<ttl>” and the end tag “</ ttl>”, and refers to / transfers the message TTL. Give to part 23. The message TTL reference / transfer unit 23 determines whether the maximum allowable number is larger than 0, and returns the determination result to the message analysis / transfer unit 22.

The message analysis / transfer unit 22 receives the message TTL reference / transfer unit 2 that the maximum allowable number is larger than 0.
When the judgment result of 3 is received, the value of the maximum allowable number defined between the start tag "<ttl>" and the end tag "</ ttl>" for transferring the message to the next node. Decrement by 1. In this way, the value (maximum allowable number) of the message TTL is updated. After that, the message is passed to the message analysis / transfer unit 22. The IP communication unit 21 divides the message received from the message analysis / transfer unit 22 into a plurality of blocks, generates a plurality of packets addressed to the destination node 16, and sends the packets to the node corresponding to the next hop.

On the other hand, when the message analysis / transfer unit 22 receives the determination result of the message TTL reference / transfer unit 23 that the maximum allowable number is 0 or less, the message life time has expired. , Perform the following error handling. That is, the message analysis / transfer unit 22 discards the message. Furthermore, the message analysis / transfer unit 22
Creates an error message (error notification) as needed. The error message is notified to a predetermined destination via the IP communication unit 21. The predetermined destination is, for example, the client 15, the destination node 16, and the client 15
And the destination node 16.

By the way, when the IP communication unit 21 receives a plurality of packets into which a message is divided, the IP communication unit 21 temporarily stores these packets. On the other hand, message analysis
The transfer unit 23 searches for the message TTL by analyzing the message. At this time, if the message TTL is not found, the message analysis / transfer unit 32
Assuming that the time to live is not set in this message, the fact is given to the IP communication unit 21. Then IP
The communication unit 21 performs a forwarding process on the accumulated plurality of packets and sends the packets to the next node. on the other hand,
When the message in which the message TTL is set is transferred to the next node, the value (maximum allowable number) of the message TTL is updated, so that the content of the message is updated. Therefore, the IP communication unit 21 discards the accumulated plural packets, generates new plural packets based on the updated message, and sends them to the next node.

The above-mentioned IP communication unit 21 corresponds to the receiving means and the transfer processing means in the present invention, and the message analysis / transfer section 22 and the message TTL reference / transfer section 23 correspond to the judging means in the present invention.

Returning to FIG. 3, the operation of the first embodiment will be described. When transmitting a message to the destination node 16, the client 15 inserts the message TTL (maximum allowable number of transit nodes) in the header of the message. The message in which the message TTL is set is divided into a plurality of packets and transferred to the IP network. When the relay node 17 in the IP network receives a plurality of packets, it assembles a message, analyzes the message, searches the message TTL, and determines whether to transfer the message or stop the transfer (determination of “maximum allowable number ≦ 0?”). To do. afterwards,
When the relay node 17 determines to transfer the message (maximum allowable number> 0), the maximum allowable number is reduced by 1, the message is updated, and then the message is transferred to the next node. On the other hand, when the relay node 17 determines not to transfer the message (maximum allowable number ≦ 0), the relay node 17 discards the message.

As described above, in the first embodiment, when the message does not reach the destination node 16 within the range of the default value (“3” in the example of FIG. 3) of the maximum allowable number set in the message, The message is discarded in the IP network.

Therefore, for example, in a network environment in which a large number of nodes (the client 15 and the destination node 16 in the example of FIG. 3) exchange messages via the IP network, it is not necessary to perform complicated message route verification. Also IP the message that has fallen into a loop
Can be automatically removed from the network.

The value of the message TTL (maximum allowable number)
Can set an appropriate value according to the type of application service. For this reason, the message will
For application services that need to arrive at the (destination), set the maximum allowable number of transit nodes for the message to reach the destination within the specified period in the message, and set the maximum allowable number by bypassing the transfer route. It is also possible to stop the transfer of a message (useless) that has been exceeded (destination has not been reached within a predetermined period).

As described above, according to the first embodiment, message transfer is controlled within the network. As a result, it is possible to prevent unnecessary packet transfer processing from being performed on the network. Therefore, it is possible to reduce the processing load on the network and effectively use the resources.

Further, since the destination node does not have to receive an unnecessary message, it does not have to execute the process related to the message (for example, the message receiving process or the message response process).

Further, at the destination of the error message, it can be known that the message has been discarded in the network and / or that the message has not reached the destination, and appropriate processing such as resending of the message can be carried out. You can

Further, the application level (message level) survival time can be set separately from the packet level survival time. Therefore, the survival time can be set for each application service. Therefore, it is possible to accommodate the application service whose result is to be confirmed in a short time such as bidding at an auction and the other application service in the same facility (IP network in FIG. 1A). In addition, multiple different application services can be provided in one facility (network).
Therefore, there is a cost reduction effect on the equipment for realizing the applied service.

The first embodiment can be modified as follows. That is, the client 15 sets the maximum allowable number (for example, 3) and the reference value (default value = 0) as the value of the message TTL. The relay node 17 compares the reference value with the maximum allowable number, transfers the message by incrementing the reference value by 1 when the reference value does not exceed the maximum allowable number, and the reference value exceeds the maximum allowable number. Stop forwarding messages to.

<Second Embodiment> Next, a second embodiment of the message relay node according to the present invention will be described. Figure 6
It is a figure which shows 2nd Embodiment of the network system to which a message relay node is applied. The second embodiment is the first
Since this embodiment includes common points to the embodiments, mainly different points will be described.

The network system shown in FIG.
Similarly, the client 15, the destination node 16, and 1
The IP network includes the relay node 17 described above. The client 15 and the relay node 17 have the functional blocks shown in FIG. However, in the second embodiment, the valid upper limit time of the message is used as the lifetime information (message TTL) that is the determination information. Therefore, the processes in the client 15 and the relay node 17 are different as follows.

Message creation unit 1 in the second embodiment
8 and the message TTL insertion unit 19 generate a message including a body and a header as shown in FIG. 7, for example. The message shown in FIG. 7 is a message TTL tag.
The message is the same as the message illustrated in the first embodiment except that the character strings sandwiched (enclosed) between (“<ttl>” and “</ ttl>”) are different.

Between the message TTL tags, a character string indicating the valid upper limit time of the message is described as the message TTL. In the above message example, the character string "20010912T12025" indicating the date and time of the upper limit of the survival time is displayed.
5 ”is described. This character string indicates that the valid upper limit time of the message is 12:02:55 on September 12, 2001.

However, the effective upper limit time may indicate the date and time corresponding to the effective upper limit time, or may indicate only the time on that day. Further, the effective upper limit time may be expressed by an absolute (direct) time as shown in the example. For example, "message transmission time + X seconds
(X is a predetermined time) ”may be expressed as a relative (indirect) time. In this way, the message TTL insertion unit 1
9 inserts the valid upper limit time in the message instead of the maximum allowable number of transit nodes. Except for the above points, the processing by the client 15 is the same as in the first embodiment.

The IP communication unit 21 of the relay node 17 receives a plurality of packets into which the message is divided, assembles the message from these packets, analyzes the message,
It is given to the transfer unit 22.

The message analysis / transfer section 22 analyzes the message and extracts the message TTL tag ("<ttl
> ”And“ </ ttl> ”) are found, the valid upper limit time described between them is extracted and the message TTL is referenced.
It is given to the transfer unit 23. Also, the message analysis / transfer unit 2
2 acquires the time when the message is received from the IP communication unit 21 as the reception time of the message and gives it to the message TTL reference / transfer unit 23 as the current time.

The message TTL reference / transfer unit 23 compares the current time received from the message analysis / transfer unit 22 with the valid upper limit time and determines (inspects) whether the current time exceeds the valid upper limit date and time. The determination result is given to the message analysis / transfer unit 22. However, the message TTL
The reference / transfer unit 23 can be modified so as to acquire the current time when the valid upper limit time is received from the message analysis / transfer unit 22 and compare the acquired time with the valid upper limit time. As the current time, it is possible to apply an appropriate time from when the IP communication unit 21 receives a plurality of packets into which the message has been divided until when the message TTL reference / transfer unit 23 performs the comparison / determination process. .

When the message analysis / transfer unit 22 receives the determination result that the current time does not exceed the valid upper limit time from the message TTL reference / transfer unit 23, the message analysis / transfer unit 22 transfers the message to the next node. To that effect I
Notify the P communication unit 21. The IP communication unit 21 temporarily stores a plurality of packets into which the message has been divided. When the notification from the message analysis / transfer unit 22 is received, the IP communication unit 21 performs a forwarding process on the plurality of packets to store the plurality of packets (messages). Transfer to the next node. As described above, in the second embodiment, since the content of the message is not updated, new packet data based on the updated message is not generated by the relay node 17.

On the other hand, the message analysis / transfer unit 22
When the result of determination that the current time exceeds the valid upper limit time is received from the message TTL reference / transfer unit 23, the message transfer is stopped, and the first
Error processing similar to that of the embodiment is performed. For example, the message analysis / transfer unit 22 discards the message held by the message analysis / transfer unit 22 and issues an instruction to discard the message to the IP communication unit 2
Give to one. As a result, the IP communication unit 21 discards the accumulated plurality of packets (messages). Also,
The message analysis / transfer unit 22 creates an error message (error notification) as necessary and notifies it to a predetermined destination via the IP communication unit 21. Except for the above points, the processing by the relay node 17 is the same as in the first embodiment.

Returning to FIG. 6, the operation of the second embodiment will be described. When transmitting a message to the destination node 16, the client 15 inserts the message TTL (effective upper limit time) in the header of the message. Message TTL
The message set with is divided into a plurality of packets and transferred to the IP network. When the relay node 17 in the IP network receives a plurality of packets, it assembles a message, analyzes the message, and sends a message T.
TL search, message transfer, or transfer stop determination is performed. After that, when the relay node 17 determines to transfer the message (the current time is before the expiration of the valid upper limit time), the relay node 17 transfers the message to the next node. On the other hand, when the relay node 17 determines not to transfer the message (after the current time has passed the valid upper limit time), the relay node 17 discards the message.

As described above, in the second embodiment, if the message does not reach the destination node 16 before the effective upper limit time set in the message elapses, the message is discarded in the IP network. by this,
The second embodiment can obtain the same effect as the first embodiment.

In the first and second embodiments, XML is illustrated as a language for creating a message, but if the language is supported by the application service, XML is used.
Other languages (eg HTML) can be used.

It is also possible to obtain a message relay node in which the first and second embodiments are combined. For example, the message analysis / transfer unit 22 of each relay node 17 determines the content of the lifetime information defined in the tag of the message TTL, and if the maximum allowable number, the processing described in the first embodiment is executed. If it is the effective upper limit time, the processing described in the second embodiment can be executed.

<Third Embodiment> Next, a third embodiment of the message relay node according to the present invention will be described. The third embodiment will describe a specific example of the message relay node. According to the present invention, message transfer can be controlled by setting the message TTL. Therefore, IP level
It is possible to determine whether or not a message can be reached within a certain time by the application level (message level) TTL, independently of whether or not the packet can be reached within a certain time by the (packet level) TTL. As a result, for example, as shown below, a plurality of application services can be realized on the same network without contradiction.

It is considered that a voice conference and a text chat are operated as a plurality of different application services on a certain IP network. Text chat is an application service for exchanging text messages between users (clients). Dialogue using text chat is established even if the text message does not reach the other party immediately. For this reason, as an allowable transfer time of a text message (that is, a threshold value for suspending (allowing) the reachability judgment of a text message) for the text chat service to be established without any problem (to operate properly), 1 A time width of 10 minutes is set. If such a time width is set, even if the text message transfer route becomes unstable and the text message transfer is temporarily delayed, if the text message reaches the other party within the time width. , Text chat service is established without any problems. For example, it can be assumed that a terminal accommodated in the IP network carries out a text chat with a mobile terminal capable of receiving text via the IP network and the mobile network. In this case, even if the connection of the mobile network is temporarily disconnected, if the time of restoration is within the time range, the relay node in the IP network retransmits the text message to the mobile terminal. , The text chat service can be established without any problem.

On the other hand, in a voice conference, which is an application service different from text chat, voices are mutually transmitted and received between users. Therefore, high real-time property (real-time property) is required. Therefore, the time width for determining whether or not the message (voice) of the audio conference can be reached is set to a minute time (for example, several seconds). As described above, the time difference is very large between the text chat and the voice conference. In such a case, if the reachability determination is performed only at the IP level (the TTL of the packet), the results obtained by using this reachability determination are applied to two different application services, respectively. This is because the two application services share the IP network and the TTL of the packet is uniquely set without distinguishing the two application services.

At this time, if the TTL of the packet is set in accordance with the voice conference, the survival time of the text message becomes shorter than the time width allowable in the text chat service. Therefore, even if the time range is allowable in the text chat service, the text message is discarded when the time to live of the packet for transferring the text message is exhausted. As described above, there is a possibility that the application service may be interrupted by discarding the message even if the application service has a proper time range.

Therefore, the text message is encapsulated by XML and transferred, and a criterion different from the IP level reachability judgment (for example, a criterion of 10 minutes longer than the time width of the voice conference) is introduced. As a result, even in the same IP network, it becomes possible for a plurality of services to operate after appropriately determining reachability.

FIG. 6 is a diagram showing a third embodiment of a network system to which a message relay node is applied, and operating a plurality of application services having different allowable time widths of reachability determination on the same IP network. Is shown.

FIG. 6 shows a network system that operates a plurality of different application services. The physical configuration of the network system is as follows. The network system has an IP network 27 accommodating a client 25 and a mobile network 28 accommodating a client (portable terminal) 26. The IP network 26 and the mobile network 28 are interworked via a gateway (not shown), and the IP communication via the IP network 27 and the mobile network 28 can be performed between the client 25 and the client 26. . When the IP communication is performed between the client 25 and the client 26, the IP packet transmitted from the client 25 passes through the relay nodes (servers) 29 and 30 in the IP network 27 and further through the mobile network 28. It arrives at the client 26 via.

The clients 25 and 26 are the IP network 27 and the mobile network 2.
8 is used to implement a text chat service and a voice conference service. Specifically, the text message is the client 25-relay node 29-relay node 30-
The text chat service is implemented by being exchanged with each other through the path of the mobile network 28-client 26. On the other hand, the voice conference is carried out by exchanging voice messages with each other through the route.

Since the transferred message is a voice, the voice conference is required to have higher real-time property than the text chat. For this reason, the allowable time for properly operating the voice conference service, that is, the time for holding the decision as to whether or not the message can reach the destination is shorter than that in the text chat. For example, the allowable time is set to several seconds such as 3 seconds. On the other hand, the allowable time of text chat is longer than that of voice conference. For example, the allowable time is set to 1 to 10 minutes.

The allowable time for each service is, for example,
It is set as follows. The allowable time for audio conference is
It is set using packet level TTL. For example,
The allowable time “3 seconds” is set in the IP header of the packet. On the other hand, the allowable time for text chat is the first
And the message TTL as described in the second embodiment.

The text message transmitted / received in the text chat is, for example, data described in XML as shown in FIG. In FIG. 9, in the header of the message, the source tag (<to> and </ to>) and the destination tag (<fro
m> and </ from>) and the message TTL tag (<ttl> and </
ttl>) and are inserted. The character string "ABC" enclosed in the sender tag indicates the sender name of the text message,
The character string "DEF" enclosed by the destination tag indicates the recipient (addressee) name of the text message. And the message T
The character string "10" enclosed by the TL tag is the valid upper limit time of the text message, and in this example, "the allowable time from the transmission of the text message to the arrival of the text message (determination of availability) Indicates that the threshold value for holding is 10 minutes. Actually, for example, the time corresponding to the elapsed time of 10 minutes is described. Meanwhile, in the body of the message, text tags (<text> and <text> / text>) is provided, and it is a character string enclosed by text tags ("Hel" in Fig. 9).
lo! ”) is the text data that the sender wants to convey to the recipient.

As described above, the transfer control of the voice message in the voice conference service is performed according to the TTL of the packet, and the transfer control of the text message in the text chat is performed according to the message TTL.

FIG. 10 is an explanatory diagram of message transfer control in the network system shown in FIG. In FIG. 10, when each of the voice conference and text chat application services is performed between the client 25 and the client 26, a service session between the client 25 and the client 26 via the relay nodes 29 and 30. Is set, and a message (voice message or text message) according to the application service is transferred on this session.

In a voice conference, for example, when a voice message is transferred from the client 25 to the client 26, the following transfer control is carried out. That is,
The client 25 generates a voice message to be transferred, generates a plurality of packets into which the voice message is divided, and transmits the packet to the IP network 27. A predetermined TTL is set in each packet. If the voice message is transferred normally, the voice message is relay node 2
9, reaching the client 26 through the relay node 30 and the mobile network 28.

On the other hand, the IP level connection in the mobile network 28 may be temporarily disconnected due to the movement of the client 26 or the like. In this case, the IP-level unreachability warning from the mobile network 28 to the relay node of the IP network 27 (network abnormality (fault) information)
Will continue to be notified. On the other hand, when the IP level connection in the mobile network 28 is restored, the notification of the network abnormality information is stopped.

The voice conference is required to have high real time.
Therefore, the relay node (for example, the relay node 30 shown in FIG. 10) receives the voice message when the network abnormality information is received and the recovery notification is not received (while the abnormality is occurring in the mobile network 28). If so, the transfer of the voice message is stopped.

During the voice conference, for example, the arrival point of the voice message (the client 2
Since the position (6) is frequently changed, the transfer route of the voice message may be changed and the packet may be forwarded. In this case, the TT of the voice message packet
L may exceed the upper limit (time over). When the relay node (for example, the relay node 30 shown in FIG. 10) receives a voice message whose TTL of the packet has timed out, the relay node stops the transfer.

As described above, the relay node on the service session of the voice conference receives the voice message when the abnormality occurs in the mobile network 28,
When a voice message whose TTL of the packet is timed out is received, it is determined that the voice message does not reach the destination (client 26) within the allowable time of the voice conference service, the transfer of the voice message is stopped, and the voice message is stopped. Discard the message.

On the other hand, in text chat, for example, when a text message is transferred from the client 25 to the client 26, the following transfer control is carried out. That is, the client 25 generates a text message (see FIG. 9) including the message TTL,
This text message generates a plurality of divided packets and transmits them to the IP network 27. Also in this case, each packet has an IP level TT corresponding to the voice conference.
L is set. If the text message is successfully transferred, the text message reaches the client 26 through the relay node 29, the relay node 30, and the mobile network 28.

On the other hand, even during the text chat, the relay node (the relay node 30 in FIG. 10) on the service session of the text chat receives the network abnormality information and receives the recovery notice as in the voice conference. If not (while the mobile network 28 is abnormal), a text message may be received.
In this case, the relay node suspends the transfer of the text message (stays the text message) and also sets the message TTL set in the text message.
Waits for the recovery notification to be transmitted from the mobile network 28 until the value exceeds the upper limit (time becomes over). The relay node then forwards the text message to the next node if it receives a recovery notice before the message TTL times out. On the other hand, the relay node discards the text message when the message TTL times out without receiving the recovery notification.

Also, during the execution of the text chat, the packet may be forwarded and the relay node may receive the text message in which the TTL of the packet is over. In this case, the relay node transfers the text message to the next node, regardless of the state of the TTL of the packet, if the message TTL has not timed out. As described above, the relay node controls the message TTL in the transfer control of the text message.
Use only.

The client 25 and the client 26 for realizing the transfer control of the text message described above.
The configuration of FIG. 4 is similar to that described in the second embodiment (see FIG.
6 and 7) can be applied. On the other hand, as the configurations of the client 25 and the client 26 for realizing the above-described voice message transfer control, the configuration of a communication terminal for realizing a conventional voice conference can be used.

FIG. 11 is a functional block diagram of a relay node (relay nodes 29, 30) for realizing the transfer control of the voice message and the text message shown in FIG. In FIG. 11, the relay node is the IP communication unit 31.
, Message analysis / transfer unit 32, and message TTL
The reference / determination unit 33 is provided.

FIG. 12 is a flow chart showing the operation of the relay node in controlling message transfer. The relay node is, for example, from the client 25 to the client 2
For voice messages forwarded to 6, the following works. That is, when the IP communication unit 31 receives a plurality of packets into which the voice message is divided, the IP communication unit 31 temporarily stores these packets. Then, the IP communication unit 31 assembles a voice message from a plurality of packets and gives it to the message analysis / transfer unit 32 (S301). Also, the TTL set in the header of each packet is checked, and the check result (TTL check result: time has not expired
(“No violation”) or time-out (“violation”)) is given to the message analysis / transfer unit 32 (S301). Furthermore, I
The P communication unit 31 is adapted to receive network abnormality information from the mobile network 28. IP communication unit 31
Determines the IP connection of the packet from the IP destination address of each accumulated packet and whether or not the network abnormality information for this IP connection is received (that is, I
It is determined whether or not an abnormality has occurred in the P connection, and the determination result (connection state determination result: "abnormal" or "normal") is given to the message analysis / transfer unit 32 (S301).

The message analysis / transfer unit 32 receives the message assembled by the IP communication unit 31, the TTL check result, and the connection state determination result. When the message analysis / transfer unit 32 determines the message category (S302) and recognizes that this message is a voice message (S302; voice), it determines whether the TTL check result is “violation” or “no violation”. (S303).

When the message analysis / transfer unit 32 determines that the TTL check result is "violation"(S303;
(Violation), to stop (stop) the transfer of the voice message, the voice message is discarded and a voice message discard instruction is given to the IP communication unit 31. Then I
The P communication unit 31 discards the plurality of packets corresponding to the accumulated voice message.

On the other hand, the message analysis / transfer section 32 uses the T
If it is determined that the TL check result is "no violation" (S
303; no violation), the abnormality determination result is "abnormal" or "normal"
It is determined (S304). Message analysis / transfer unit 32
Determines that the abnormality determination result is “abnormal” (S3
04: Abnormal), the same operation is performed when it is determined that the TTL check result is "violation" (S303: violation). As a result, the voice message and the packets corresponding to the voice message are discarded, and the transfer of the voice message is stopped.

On the other hand, the message analysis / transfer unit 32
If the connection status determination result is "normal"(S304; normal), the voice message transfer instruction is I
It is given to the P communication unit 31. Then, the IP communication unit 31 executes a forwarding process (indexing and forwarding of the next node) for a plurality of packets corresponding to the accumulated voice messages.

When the relay node of the voice message in the IP network 27 executes the above operation for the voice message, the transfer control of the voice message shown in FIG. 10 is realized.

On the other hand, the relay node operates as follows for a text message. That is, the IP communication unit 31
When a plurality of packets into which a text message is divided is received, the same operation as in the case of a voice message (S3
01 operation), and the assembled message and TT
Message analysis of L check result and abnormality determination result
It is given to the transfer unit 32.

The message analysis / transfer unit 32 determines the message received from the IP communication unit 31 (S30
2) If it is determined that this is a text message (S
302; text), further analyzing the text message, taking out the message TTL inserted in the header of the text message (S305), the message TTL
It is given to the reference / transfer unit 23.

The message TTL reference / transfer unit 23 determines whether or not the value of the message TTL is a time-out (whether violation or no violation), and analyzes the determination result (“violation” or “no violation”) by message analysis / It returns to the transfer unit 32.

The message analysis / transfer section 32 judges whether the judgment result of the message TTL reference / transfer section 23 is "violation" or "non-violation" (S306), and if it is "violation" (S).
306; violation), the transfer of the text message is stopped, and the text message is discarded, and a packet discard instruction is given to the IP communication unit 31 to discard the corresponding plurality of packets.

On the other hand, the message analysis / transfer unit 32
When the determination result of the message TTL reference / transfer unit 23 is "no violation"(S306; no violation), it is determined whether the connection state determination result from the IP communication unit 31 is "normal" or "abnormal". Yes (S307). At this time, when the connection status determination result is “normal”, the message analysis / transfer unit 32 determines that the text message transfer is to be continued, and the message transfer instruction is I.
It is given to the P communication unit 31. Then, the IP communication unit 31 executes a forwarding process (indexing and transfer of the next node) for a plurality of packets corresponding to the accumulated voice message, and the text message is transferred to the next node.

On the other hand, when the connection state determination result is "abnormal", the message analysis / transfer unit 32
Holds the transfer of the text message (S308).
That is, the text message is retained in the relay node.

By the way, the IP communication unit 31 receives the I / O of this message until it receives a discard instruction or a transfer instruction for the message passed to the message analysis / transfer unit 32.
The connection status determination result for the P connection is continuously passed to the message analysis / transfer unit 32. Therefore, IP
When the abnormal state of the connection is recovered, the IP communication unit 31 passes the “normal” connection state determination result to the message analysis / transfer unit 32.

On the other hand, the message analysis / transfer unit 32, if the text message is suspended, sends the message TT.
L is given to the message TTL reference / transfer unit 33 to acquire the determination result, and the process of determining whether the determination result is “violation” or “non-violation” is repeatedly executed. That is, the message analysis / transfer unit 32 repeatedly executes the loop processing of S306 to S308.

After that, the message analysis / transfer unit 32
If a “normal” connection status determination result is received from the IP communication unit 31 before the “violation” determination result (message TTL time-out) is received from the message TTL reference / transfer unit 32 (S307; normal), hold As for continuing the transfer of the text message that was being done,
The transfer instruction of the text message is given to the IP communication unit 31. As a result, the transfer processing of the text message is restarted, and the IP communication unit 31 executes the forwarding processing for the plurality of packets (stored) corresponding to the text message. This transfers the text message to the next node.

On the other hand, the message analysis / transfer unit 32
Before receiving the “normal” connection status determination result from the IP communication unit 31, the message TTL reference / transfer unit 3
When the judgment result of "violation" is received from 2 (S30
6), the IP communication unit 31 is instructed to discard the text message by suspending the transfer of the held text message. Then, the IP communication unit 31
Discards a plurality of packets corresponding to the accumulated text message.

The IP communication unit 31 described above corresponds to the receiving means, the transfer processing means and the detecting means in the present invention, and the message analysis / transfer portion 32 and the message TTL.
The reference / transfer unit 33 corresponds to the determination means in the present invention.

According to the third embodiment, the same effect as that of the second embodiment can be obtained. Furthermore, in the third embodiment, for a text message, if an abnormality occurs in the IP connection, the transfer processing is suspended by the relay node, and the IP connection is in an abnormal state before the message TTL times out. When it is recovered from, the transfer process is restarted. Therefore, the text message is transferred to the destination (the client 26) independently (regardless of) of the TTL of the packet (that is, the allowable time of the voice conference) within the time range allowed by the text chat service. be able to. In this way, the message can be transferred within the range of the permissible time appropriate for each application service. Therefore, it is possible to properly operate a plurality of application services that are operated using the same equipment.

In the third embodiment, the transfer control of the voice message of the voice conference is executed according to the TTL of the packet, but it is executed according to the message TTL set separately from the text chat. Is also good.

In the operation of the relay node (relay server) described with reference to FIGS. 11 and 12, the IP communication unit 3
The check process and the determination process performed in 1 may be performed in the message analysis / transfer unit 32 and / or the message TTL reference / transfer unit 33. Further, a part of the processing performed by the message analysis / transfer unit 32 (for example, the determination processing of S306 and S307) may be performed by the message TTL reference / transfer unit 33.

<Others> The present invention can be specified as follows. (Supplementary Note 1) Receiving means for receiving a message generated at the application level and transferred through the network, and determining whether or not to transfer the message based on the determination information inserted in the message received by the receiving means. And a transfer processing unit that transfers the message when the determination unit determines that the message is to be transferred, and stops the transfer of the message when the determination unit determines that the message is not transferred. Message relay node. (1) (Appendix 2) When the transfer of the message is stopped,
The message relay node according to appendix 1, further comprising a discarding unit that discards this message. (Supplementary Note 3) When the transfer of the message is stopped,
The message relay node according to appendix 1, further comprising a notification unit that outputs a notification indicating a message transfer error. (Supplementary Note 4) The determination information is survival information indicating whether the message is survivable, and the determination means determines to transfer the message when the survival information indicates a survivable range, The message relay node according to appendix 1, which determines not to transfer the message when the information indicates a non-viable range. (2) (Supplementary note 5) The survival information indicates the number of passages of the message through the relay node in the network, and the determination means indicates that the number of passages of the message does not exceed the upper limit. 5. The message relay node according to appendix 4, wherein the message relay node is determined to be transferred, and the message relay node is determined not to be transferred if the number of packets that can pass through the relay node exceeds the upper limit. (3) (Supplementary note 6) The survival information indicates an upper limit time at which the message can survive, and the determining unit determines to transfer the message if the current time has not passed the upper limit time, and the current time is 5. The message relay node according to appendix 4, which determines not to transfer the message if the upper limit time has elapsed. (4) (Supplementary note 7) The message is divided into a plurality of packets and transferred through the network, the survival time of each packet is set, and the transfer processing means determines that the survival time of the packet has ended. Also, the message relay node according to appendix 4, which transfers the message if the survival information of the message indicates a survivable range. (Supplementary Note 8) The device further comprises detection means for detecting a failure in the connection to which the message is transferred, and the transfer processing means, even when the determination means determines to transfer the message, causes the failure in the connection by the detection means. Is detected, the transfer processing of the message is suspended, and thereafter, when the connection recovers from the failure before the upper limit time when the message can survive, the transfer processing of the message is suspended. The message relay node described in appendix 6. (5) (Supplementary note 9) The transfer processing means, while suspending the transfer processing of the message, suspends the transfer of the message when the upper limit time at which the message can survive has passed. 8. The message relay node described in 8.

[0117]

According to the present invention, it is possible to control the transfer of the message generated at the application level.

[Brief description of drawings]

FIG. 1A is a schematic explanatory view of the present invention.
(B) is an explanatory view of a message applied to the present invention.

FIG. 2 is a schematic explanatory diagram of the present invention.

FIG. 3 is a diagram showing a network system to which a first embodiment of a message relay node according to the present invention is applied.

FIG. 4 is a functional block diagram of a client and a relay node shown in FIG.

FIG. 5A is an XML forming a message body.
It is a figure which shows the example of data, and FIG.5 (B) is a figure which shows the example of the XML data which comprises a message.

FIG. 6 is a diagram showing a network system to which a second embodiment of a message relay node according to the present invention is applied.

FIG. 7 is a diagram showing an example of XML data forming a message used in the second embodiment.

FIG. 8 shows a physical configuration of a network system to which a third embodiment of a message relay node according to the present invention is applied, and a logical configuration of a text chat service and a voice conference service realized by this network system. FIG.

FIG. 9 is a diagram showing an example of XML data forming a text message used in a text chat service.

10 is an explanatory diagram of transfer control of a voice message and a text message in the network system shown in FIG.

FIG. 11 is a functional block diagram of a third embodiment of a message relay node.

12 is a flowchart showing the operation of the message relay node shown in FIG.

FIG. 13 is an explanatory diagram of a conventional technique.

[Explanation of symbols]

M message 11,15 client 12 server 13 message router 14,17,29,30 relay node (message relay node) 16 destination node 18 message creation unit 19 message TTL insertion unit 20 communication unit 21,31 IP communication unit 22,32 message Analysis / transfer unit 23,33 Message TTL reference / transfer unit

   ─────────────────────────────────────────────────── ─── Continued front page    (72) Inventor Koichi Takeda             4-1, Kamiodanaka, Nakahara-ku, Kawasaki-shi, Kanagawa             No. 1 within Fujitsu Limited F term (reference) 5K030 HA08 HB11 HB15 JA05 JA11                       LB05 LC18

Claims (5)

[Claims] 1. Receiving means for receiving a message generated at an application level and transferred through a network, and whether or not to transfer the message based on judgment information inserted in the message received by the receiving means. Determination means for determining, and transfer processing means for transferring the message when the determination means determines to transfer the message, and stopping transfer of the message when the determination means determines not to transfer the message. A message relay node that contains. 2. The determination information is survival information indicating whether or not the message is survivable, and the determination means determines to transfer the message when the survival information indicates a survivable range, The message relay node according to claim 1, wherein the message relay node determines not to transfer the message when the survival information indicates a non-survivable range. 3. The survival information indicates the number of the message that can pass through the relay node in the network, and the determination unit indicates that the number of the message that can pass through the relay node does not exceed its upper limit. 3. The message relay node according to claim 2, wherein it is determined that the message is to be forwarded, and it is determined that the message is not to be forwarded when the number that can pass through the relay node exceeds its upper limit. 4. The survival information indicates an upper limit time when the message can survive, the determining unit determines to transfer the message if the current time has not passed the upper limit time, and the current time is the upper limit. The message relay node according to claim 2, wherein if the time has passed, it is determined not to transfer the message. 5. The device further comprises detection means for detecting a failure of a connection to which the message is transferred, and the transfer processing means uses the connection means to detect the connection even if the determination means determines to transfer the message. When a failure is detected, the transfer processing of the message is suspended, and thereafter, when the connection recovers from the failure before the upper limit time when the message can survive, the transfer processing of the message The message relay node according to claim 4, wherein