JPH01297933A - Wide-area electronic mail system - Google Patents

Abstract

PURPOSE:To update address information with a small amount of communication data by providing a host computer with a request command sending means and a sending-back means for all-address information or error detection data and address variation information. CONSTITUTION:Host computers A, B, C and D of respective area subsystems are equipped with information control tables A1, B2, C3 and D4 for the respective subsystems and when there is variation in terminal address, the host computer A, for example, updates the address control table A1 and check sum value 5 in its subsystem with an A-check sum 5 and further updates address tables B6, C7 and D8. Then the host computer A sends back the all-address information or the address variation information of the address table B6, and the error detection data on the A-check sum 5, thus updating the all address control tables with the least amount of communication data.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention connects host computers connected to subsystems to each other via a wide area network, and enables e-mail communication between the subsystems connected to the host computers. This relates to a wide-area e-mail system that performs.

[Prior Art] An example of the configuration of a general wide area network system is shown in FIG.

In Figure 5, four regional subsystems A30, B
2O, C50, and D60 are hosts A31.20, C50, and D60, respectively. B41
, C51, and D61 to the wide area network 30.

Each regional subsystem 30 to 60 is composed of a LAN (local area network), and includes multiple terminals 732 to 60.
65 are each connected to the host AND (31-61).

Each terminal is capable of e-mail communication with terminals of other local subsystems connected to other hosts via the host and wide area network 30. For example, terminal 34 in subsystem A30 is terminal 6 in subsystem D60.
If an attempt is made to send an e-mail to 5, the e-mail transmission will be executed in the following steps.

First, the terminal 34 requests the host A31 for a list of e-mail registered users in the subsystem D60. Then, obtain the known e-mail registered user list using the method described below.

Then, from the obtained e-mail registered user list, terminal 6
5 and sends an e-mail generated with that address value as a header to host A31.

The host A31 recognizes from the header that the e-mail is addressed to the terminal of the subsystem D60, adds a header addressed to the host D61 to the e-mail, and sends it to the wide area network. The packet received by the host D61 is sent to the terminal 6 via the LAN in the subsystem D60.
Delivered on 5th.

In order for the above electronic mace system to operate smoothly,
It is clear that the addresses of e-mail registered users managed by each host are not limited to its own subsystem, but extend to the entire subsystem connected to the wide area network.

That is, the host A31 needs to manage the addresses of all terminals not only in the subsystem A30 but also in the subsystems B40%C50%D60.

The managed addresses must be constantly up-to-date in response to changes in the system.

Conventionally, the following two methods have been commonly used to update this address table.

The first method is that the administrator of the entire system obtains the latest address table from the administrator of each subsystem, and then
This is a method in which the information is compiled on a recording medium such as a magnetic tape and mailed periodically, for example, once every month, to the host administrator of each subsystem, so that the administrator of each host updates the information individually.

The second method is to perform communication between hosts using a wide area network. This method can be used, for example, when host A3
In this method, each host updates the address table regarding the subsystem A30 by having the administrator of the subsystem A30 transmit the latest address table of the subsystem A30 to all remaining hosts B41%C51 and D61.

[Problems to be Solved by the Invention] However, the first and second conventional methods described above have the following drawbacks.

The disadvantage of the first method is that because of the long update cycle, new users who wish to register are forced to wait for a long time until the address table is updated, and cannot receive e-mail service during that time.

Further, if the update cycle is shortened, an excessive burden will be placed on the administrator of the entire system and the administrator of each system.

The disadvantage of the second method is that as the system scale increases, the number of communications required for updating increases, resulting in an increase in communication charges.

For example, if distributed hosts are installed in 100 locations,
9900 for all hosts to finish updating all systems
This requires multiple host-to-host communications.

[Means for Solving the Problems] The present invention has been made for the purpose of solving the above-mentioned problems, and includes the following configuration as one means for solving the above-mentioned problems.

That is, a transmission means for transmitting at least two types of address information request commands to other host computers to a host computer, and receiving the first address information request command from the other host computer, allows the subsystem of the own host to be transmitted. A first return means for returning all address information to the address information request command sending host, and error detection data and address fluctuation information in the subsystem of the own host upon reception of the second address information request command from another host computer. and a second return means for returning information consisting of and to the address information request command sending host.

[Operation] With the above configuration, the address information update requesting host computer can obtain the latest address in the desired subsystem and information corresponding to the address with a minimum amount of communication data.

[Example] Hereinafter, an example according to the present invention will be described in detail with reference to the drawings. Note that the network configuration of this embodiment is the same as that shown in FIG. 4, so a description of the network configuration will be omitted.

In this embodiment, a network management table is provided which is stored in a memory built into the host computer of each regional subsystem connected to the wide area network 30, and destination terminals within a desired subsystem are determined according to the management table. An address value is determined and a transmission packet is generated according to the address value. An example of the structure of this network management table is shown in FIG.

In FIG. 1, 1 is an address management table A (A-a
ddr), 2 is an address management table B (B-a
ddr), 3 is an address management table C (C-add
r), 4 is an address management table D (D-addr) that manages terminal addresses in the subsystem D60.
)' is.

Note that among the terminal addresses within each of these subsystems, the address management table within the own subsystem is specially managed compared to other address management tables. For example, in the case of host A31, special management is performed regarding address management table A1.

That is, the management data shown in FIGS. 5 to 8 is separately provided.

5 is the checksum value (A-checksum) of the address table of subsystem A30, and 6 is the address management table A1 of subsystem A30 that host A31 sent to host B41 last time, and the current subsystem A3.
Address table B shows the difference from address management table A1 of 0, and 7 is address management table A of subsystem A30 that host A31 sent to host C51 last time.
1 and the address management table Al of the current subsystem A30.
This is an address table D showing the difference between the address management table A of A.30 and the address management table A of the current subsystem A30.

Then, the terminal address of subsystem A30 changes (Δ
If A) is found, the address management table A1 is updated, the checksum is recalculated, and the checksum value 5 is also updated. Furthermore, based on ΔA, address table B6, address table C7, address table D8
Update.

Furthermore, if the entire address management table A1 is transmitted to the host B41, there will be no difference between the address management table A1 in the host A31 and the address management table AI in the host A31. Therefore, address table B6 is reset. The same applies to host C1 and host D.

Although the address management state within the host A31 has been described above, the same applies to the other hosts B41, host c51, and host D61.

That is, it has address management table AI, address management table B2, address management table c3, address management table D4, checksum values of each address management table, and address table AND table.

Now, address information is exchanged between the hosts that perform the above address management by transmitting and receiving command frames and response frames.

The format of the transmission frame used in this embodiment will be explained with reference to FIGS. 2 and 3.

The transmission frame 100 in FIG. 2(A) is a “command AL
This is the transmission frame format of the LJ packet.

110 is a control frame indicating that this packet is a "command ALLJ packet," 120 is a source frame indicating the source host that sent the packet, and in FIG. 2(A), host A31 is the source. “#A” indicating the destination is stored. 121 is a destination frame indicating the destination (destination) host.

The destination host (host B in this example) that received this command ALLJ packet returns a response packet 150 containing address table A to the source host (host A in this example).

In this return packet 150, 151 indicates that this packet is a packet containing an address table.
Address table data is stored in the AckALLJ control frame and data frame 155. In this embodiment shown in FIG. 2(B), the source frame 120 is the host B41, and "#B" is stored therein.

Also, in the destination frame 121, "
#A” is stored.

Therefore, by transmitting a similar "command ALLJ packet" to a host that wishes to update the address management table, the host A31 can obtain all the necessary address information in the partner subsystem.

In this embodiment, in addition to the above, FIG.
Communication using packets shown in Figure (B) is also performed.

FIG. 3(A) shows a request for sending the address table and checksum value of the desired host subsystem.
This is a command DIFFJ packet 140, and the control frame 111 stores a command DIFFJ indicating that this frame is a command DIFFJ packet.

Further, 120 and 121 are a transmission source frame and a transmission destination frame similar to those in FIG. 2(A). In this case, "#A" is stored in the source frame 120 and "#B" is stored in the destination frame 121.

The destination host, for example, host B41, which received this "command DIFFj packet 140, returns a response packet 160 containing address table data shown in FIG. 3(B) to the source host, for example, host A31.

The response packet 160 sends rAck to the control frame 152.
DIFFJ is stored, and host B4 is stored in the data frame.
Address table data 165 of own host of No. 1 and checksum of own host (B-checksum)
166 are included.

Control including update processing of the address management table of the host device of this embodiment having the above configuration will be described below with reference to the flowchart of FIG. 4.

The following is from the terminal of subsystem A30 to subsystem B.
An example will be explained in which transmission processing is performed to 40 terminals.

In this case, the transmission request terminal specifies the desired destination terminal and sends the transmission data to the host A31. This means that a packet to be transmitted to the host A31 has been generated, and if there is no packet reception at step S1 in FIG. 4, the process advances to step S2, and since there is a packet to be transmitted, the process advances to step S5. Note that if there is no packet to be transmitted, the process returns to step S1 and monitors whether there is a packet reception or packet transmission request.

In step S5, the host A31 determines whether to update the address management table in the destination subsystem. This can be determined arbitrarily, such as when a predetermined amount of time has passed since the address management table was first updated, or may be determined by a specified input from the operator.

If the address management table of the destination subsystem is not to be updated, the process proceeds to step S6, where the host A31 generates a known transmission packet by referring to the address management table 2 of the subsystem B40 held in the host A31, and updates the address management table of the destination subsystem. Send to terminal.

On the other hand, if the address management table is to be updated in step S5, the process proceeds to step S7, where a command DIFFJ packet 140 is generated and transmitted from host A31 to host B41.

In response, host B41 sends rAckDIFFJ
Packet 160 is returned. For this reason, the following step S
The host A31 that received this in step S9 updates its own address management table B2 with the address table 155 sent in step S9, and then updates the address management table B2 of its own host with the address table 155 sent in step S9.
In step O, the checksum of the updated address management table B2 is calculated, and in step Sll, it is checked whether the calculated value matches the value of the B checksum frame 166 sent. If the checksums match, it means that the address management table B has been correctly updated, and the process advances to step S6.

In this way, it is normally possible to obtain the address information of other desired subsystems by sending the command DrFFJ packet. According to this method, the amount of communication data is small, and the data is sent at the same time. The correctness of the updated information can also be confirmed by the "checksum" that comes.

On the other hand, if the checksums do not match in step Sll, the address management table B has not been updated normally, and the process proceeds to step S15 and subsequent steps.
LLJ packet 100 will be used.

That is, in step S15, a "command ALLJ packet 100" as shown in FIG. The rAckALLJ frame 150 containing the address management table is returned.Therefore, in step S16, this rAckALLJ frame 150 is sent back.
J packet 150 is received, and the data in the address management table 155 in step S17 is used to create the corresponding address management table (address management table B
) to update. Then, the process advances to step S6.

On the other hand, in the case of bucket reception in step S1, the process advances to step S20, and it is checked whether or not the received command DIFFJ packet 140 is received.
) is generated and the command DIFFJ packet 140 is sent to the destination.

If the received packet is not the "command DIFFJ packet 140", the process advances to step S25, and the "command ALLJ packet 140" is not received.
Check whether packet 100 has been received. "Command AL
In the case of receiving the LJ packet 100, the process advances to step S26, and the rAckALLJ packet 1 shown in FIG. 2(B) is received.
50 is generated and "send command ALLJ packet 100 to destination.

In step S25, if the command ALLJ packet 100 has not been received, the process corresponding to the received packet is executed in step S27.

As described above, according to this embodiment, all address management table data in a desired subsystem is included, and although communication data increases, it is possible to reliably obtain all address information.

[Other Embodiments] In the above description, "checksum" is merely an example of error detection data (check code), and any other arbitrary error detection data (check code) may be used.

It is good to be able to detect errors reliably, for example, C
Of course, other error detection data such as RC (cyclic redundancy check) data may be used.

As explained above, according to this embodiment, it is not only possible to update address information at low communication costs, but also to increase the frequency of updates in terms of cost, and to quickly respond to changes in e-mail user registration. becomes possible.

[Effects of the Invention] As described above, according to the present invention, the entire address management table can be updated with a small amount of communication data. This makes it possible to update the address management table of a system desired by the network at low communication costs, and it is also possible to increase the frequency of updating the address management table. It is also possible to quickly respond to changes in e-mail user registration.

[Brief explanation of the drawing]

FIG. 1 is a diagram showing the address management state in the memory of host A according to an embodiment of the present invention, and FIG. 2(A) to FIG. 3(
B) is a diagram showing the transmission frame structure used in this embodiment, and FIG. 2(A) is a diagram showing a command ALLJ packet, FIG. A) is a diagram showing the command DIFFJ packet, Figure 3 (B) is a diagram showing the rAckD I FFJ packet, Figure 4 is a flowchart showing the address management table update process of this embodiment, and Figure 5 is the wide area electronic It is a system configuration diagram of the mail system. In the figure, 20... wide area network, 30, 40° 50
．． 60...Subsystem, 31,41,51゜61・
・Host, 32~35.42~45.52~55゜6
2-65...Terminal. Patent applicant Canon Co., Ltd.

Claims (1)

[Claims] A wide area e-mail system in which host computers connected to subsystems are connected to each other via a wide area network, and e-mail communication is performed between the subsystems connected to the host computers, in which the host computers communicate with other host computers. transmitting means for transmitting at least two types of address information request commands to the host computer; and transmitting means for transmitting all address information in the subsystem of the own host to the address information request command sending host by receiving the first address information request command from another host computer. A first return means that sends back information, and information consisting of error detection data and address fluctuation information in the subsystem of the own host upon reception of a second address information request command from another host computer. A wide area e-mail system comprising: a second return means for sending a command back to a command sending host.