JPH05216908A - On-line system and its operation - Google Patents

Abstract

(57) [Summary] [Object] To provide an online system that realizes high system reliability and an operation method thereof, for an online system that depends on a host computer. [Structure] The branch processor distributes the processing according to the status of the host (step 2). If the host is normal, the host ledger and the business partner ledger of the customer are updated (step 3). If the host is abnormal, This is achieved by updating the business partner ledger of the business partner (step 4), and at the time of recovering the host, matching the ledger of the host with the respective business partner ledger (step 6). [Effect] Even if the host does not function for some reason, it is possible to continue operation without stopping the system, and the system can be made highly reliable.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an online system and its operation method, and more particularly, to an online system suitable for a system such as a bank system which has a ledger in a host and constantly refers to and updates the ledger. A system and its operating method.

[0002]

2. Description of the Related Art Various methods have been proposed to realize high reliability of online systems. For example,
The technique disclosed in Japanese Patent Laid-Open No. 63-158662 is a technique called so-called hot standby, and when a failure occurs in a host computer, it switches to a standby computer. Further, the technique disclosed in Japanese Patent Laid-Open No. 2-306362 aims to improve the reliability of the computer itself by using the computer as a multiprocessor.

In a bank system or the like, as a countermeasure against a line system failure, a method of connecting between the host computer and the branch office processor by a public line in addition to a leased line and switching at the time of failure, or a method of duplicating the line Is generally taken.

[0004]

However, regardless of which of the above-mentioned conventional techniques is adopted, it is impossible to construct a 100% safe system by any means. In particular, in a system such as a banking system that depends on the host computer for centralized management of the ledger, it is fatal that the host computer does not work. I need to leave.

An object of the present invention is to provide an online system and its operation which realizes high reliability of the system by avoiding a risk when the host computer fails to function due to some factor in the online system which depends on the host computer. It is to provide a method.

[0006]

To achieve the above object, according to one aspect of the present invention, a host computer,
In an online system having a plurality of terminal systems connected to a host computer via a network and processing transactions from the terminal systems by the host computer, the host computer records transaction processing results from all the terminal systems. A master file is provided, the host computer has a function of updating transaction processing results from all terminal systems with respect to the master file, and each terminal system has a terminal device for inputting / outputting transactions, A subfile for recording the processing result of a transaction related to the terminal system, a process for performing at least a communication process for the terminal device, a communication process for the host computer, and an update process for the subfile. The subfile has a function of displaying a processing result of a transaction processed at the time of the failure of the host computer and a failure processing display indicating that the processing has been processed at the time of the failure of the host computer. At times, there is provided an online system characterized by further having a function of sending a processing result of a transaction having a failure processing indication to a host computer.

The processor may further include a failure occurrence storage means for storing the occurrence of a failure when the failure is detected in the host computer until the failure recovery. Further, the processor can have a host status judging means for judging whether the current status of the host computer is normal, a failure, or a failure recovery by referring to the failure occurrence storage means.

The processor can have a communication processing function for communicating with another terminal system.

The processor may have a timer function for monitoring whether or not the response from the host computer is within a certain time after sending the message to the host computer.

As a terminal system suitable for constructing such an online system, a terminal device for inputting / outputting a transaction, a subfile for recording a processing result of a transaction related to the terminal system, and a terminal device. To the host computer, and at least a processor for performing the update process to the subfile, and the subfile is the processing result of the transaction processed when the host computer fails, and the subfile indicates that it was processed when the failure occurred. The terminal system is further provided with a function of sending a processing result of a transaction with a failure processing display to the host computer when the host computer recovers from the failure. It That.

According to another aspect of the present invention, an online system comprising a host computer and a plurality of terminal systems connected to the host computer via a network, wherein transactions from the terminal system are processed by the host computer. In, the host computer is provided with a master file for recording the processing results of transactions from all terminal systems, and each terminal system is provided with a subfile for recording the processing results of transactions related to the terminal system. For a transaction, if the host computer is normal, the master file is updated and the subfile of the terminal system related to the transaction is updated. Terminal updates only the sub file system that, at the time of fault recovery of the host computer, the method operating on-line systems and performing matching between the sub-file and the master file is provided.

With respect to the transaction processing result updated in the subfile when the host computer fails, a flag indicating that the transaction processing result is updated when the host computer fails can be set for each record.

At the time of failure recovery of the host computer, failure recovery can be transmitted to each terminal system, the master file can be updated with the transaction processing result sent from each terminal system, and the subfile and the master file can be matched. In this case, each terminal system can send the transaction processing result only for the record in which the flag is set.

Further, when the failure of the host computer is detected, a flag indicating the failure of the host computer can be set in the terminal system, and the mode of processing for the generated transaction can be distributed in the state of this flag.

The online system and its operating method of the present invention can be preferably applied to an online system used for bank transactions. In this case, the terminal system is arranged for each bank sales office, and the ledger that records the transaction results of all transaction accounts is the master file, and the sales office ledger that records the transaction results of the transaction accounts in each terminal system is the sub file. do it. Also, in the terminal system,
A transaction terminal for inputting / outputting transactions can be connected.

The terminal system sends a transaction message relating to the inputted transaction to the host computer if the host computer is normal, and if the host computer is in a failure state, the transaction is related to the terminal system of another sales office. If so, send it to the terminal system, and if the transaction is related to the terminal system of the self-employed store,
It is possible to distribute the processing so as to update the sub file of the terminal system of the self-employed store with the transaction data included in the transaction message.

When the terminal system detects the failure of the host computer, the terminal system sets a flag indicating the failure of the host computer, and the above processing can be distributed according to the setting state of the flag. In addition, the terminal system that has detected the failure of the host computer, when sending the transaction message to the terminal system of the other sales office, adds the information for notifying the failure of the host computer to the message and sends the message to the other terminal system. Can be contacted about the disorder.

The host computer updates the ledger based on the transaction message received from each terminal system and sends a response message to the terminal system which is the sender of the transaction message. The transaction message can be transmitted to the terminal system of the sales office that operates.

[0019]

Next, the operation of the present invention will be described based on an example in which the present invention is applied to a bank online system.

The host computer and each terminal system arranged in each of a plurality of business stores are connected by a network. In order to process transactions from the terminal system by the host computer, the host computer is provided with a ledger of all accounts as a master file, and the terminal system is provided with a ledger of the business office account as a subfile.

If the host computer is normal, the terminal system sends the transaction message received from the transaction terminal to the host computer, and updates the business ledger if the transaction is a transaction of a self-employed business account. Upon receiving the transaction message, the host computer updates the ledger, sends a response message to the sender, and, if the transaction is not the transaction message sender, sends the transaction message to the corresponding terminal system. The terminal system which receives the transaction message from the host computer updates the branch ledger and sends a response message to the host computer.

If the host computer is faulty,
The terminal system transmits the transaction message to the corresponding terminal system only when the transaction message received from the transaction terminal is not the transaction for the self-employed store account. The terminal system that has received the transaction message from the other terminal system updates the branch ledger and transmits a response message to the transmission source terminal system.

Upon recovery from the failure, the host computer sends a failure recovery notification to each terminal system. Upon receiving the failure recovery notification, the terminal system transmits to the host computer the record of the sales office ledger that was processed when the host computer failed. The host computer updates the ledger based on the record of the branch ledger sent from each terminal system.

As a result, even if a failure occurs in the host computer, the entire system does not stop, so the system can be made highly reliable.

[0025]

Embodiments of the present invention will be described in detail below with reference to the drawings.

FIG. 2 shows an example of the system configuration of an embodiment of the online system of the present invention. Unless otherwise stated, the following description is based on the configuration of this drawing.

As shown in FIG. 2, the online system of this embodiment is constructed by connecting a center H and a plurality of business offices (three shops are shown in FIG. 2) A, B and C through a network 50. This is an example applied to a bank online system.

The center H is provided with a host computer (hereinafter, referred to as a host) 20 and a ledger 22 which is a master file for recording information on transactions that occur at all business stores. The ledger 22 is connected to the host 20 and
Thus, the transaction information update process is performed.

A terminal system is arranged at each of the business offices A, B and C. This terminal system includes a sales office processor 10, a sales office ledger 12 that is a subfile that records only information on transactions related to the sales office, and one or more transaction terminals 11. Examples of the transaction terminal 11 include an ATM (Automatic Teller Machine), a CD (Automatic Teller Machine), and a counter terminal.

The branch processor 10 is, for example, as shown in FIG.
It is configured as shown in. That is, communication control for communication between the overall control unit 31 that controls the entire terminal system and the transaction terminal 11, and communication between the host 20 or the terminal system of another business office via the network. It has a unit 32, a file control unit 33 for controlling reading and writing of various files including the above-mentioned sales office ledger 12, and a flag register 36 for constituting a means for storing that a failure has occurred in the host 20. .. In this embodiment, as the files, in addition to the sales office ledger 12, a journal file 34 and a customer information file 35.
And are further connected.

Although not shown, the overall control unit 31, the communication control unit 32, and the file control unit 33 are all configured to have a central processing unit (CPU), a memory, etc., and the procedure shown in the flow chart described later. Then, each process is executed. For example, the overall control unit 31 uses the host 2
It functions as a means for determining the state of 0, a means for distributing processing, a means for notifying a failure to another terminal system, and the like.

FIG. 1 is a flow chart showing the movement of the entire system. A transaction input (transaction) occurs at the transaction terminal of any of the sales offices (step 1).
The processing is sorted according to the state of the host (step 2). If the host is normal, the host ledger and the business partner ledger of the customer are updated with respect to the generated transaction (step 3). If the host is abnormal, the business partner ledger of the supplier is updated for the transaction that occurred (step 4). In the case of step 3 and step 4, the transaction result is output to the transaction terminal where the transaction was input (step 5). When the host is restored, the host ledger and each business store ledger are matched (step 6).

FIG. 3 shows the relationship between the ledger and the sales office ledger.

In the ledger 22 and the branch ledger 12, the account number and various items corresponding thereto are recorded. The business ledger 12 has only data relating to the account number of the business, and the host ledger 22 has data relating to the account numbers of all the business.

Here, the feature of the present invention is that the sales office ledger 12 is used.
In addition, it has an item called a failure handling flag. The failure processing flag is for displaying a failure processing indicator that marks (turns ON) the record of the account number processed at the time of the failure of the host computer and indicates that the record has been processed at the time of the failure. This allows
When the ledger 22 and the sales office ledger 12 are matched when the failure of the host computer is recovered, it is possible to identify whether or not the record transmitted from the sales office processor is the record processed at the time of the failure. To set the error handling flag, enter the error handling flag item of the corresponding record.
This is performed by setting a code indicating the flag setting, for example, "1".

The processing of the branch office processor will be described in detail with reference to FIGS. 4 to 8, and the processing of the host will be described in detail with reference to FIGS. 9 and 10.

FIG. 4 is a flow chart showing the overall processing of the branch office processor.

First, the communication control unit 32 receives a message (step 101). If a time-out occurs here, the overall control unit 31 performs a time-out process (step 108). Timeout is
For example, this occurs when a response message from the host 20 or the like cannot be received within a fixed time. The details of the timeout process will be described later with reference to FIG. 7.

Next, the overall control unit 31 sorts the processing according to the type of the received message (response message, transaction message, other message) (step 1).
02). If the response message is from the host 20 or another sales office, the response message is transmitted to the transaction terminal 11 (step 106). The transmission is performed by the communication control unit 32. If it is another message, other message processing is performed (step 107). Details of other message processing will be described later with reference to FIG.

If it is a transaction message, the process is further sorted by the sender (step 103). If the sender is the transaction terminal 11, the message from the transaction terminal is processed (step 104). If the sender is another business office or host, message processing from the other business office or host is performed (step 105). Message processing from the transaction terminal will be described later with reference to FIG. 5, and message processing from other sales offices and hosts will be described later with reference to FIG. The branch processor 10 repeatedly executes the processing shown in FIG. 4 described above.

The transaction message used here has a format as shown in FIG. 18, for example. That is, the destination address 18 indicating the destination
1, a transmission source address 182 indicating a transmission source, a message type 183 indicating a message type necessary for message distribution, a failure contact flag 184 for notifying the occurrence of a failure, and transaction data 185 that is the content of the message. Is described. Here, the message type includes, for example, a response message and a transaction message. If the host 20 is normal, the failure contact flag may be ignored, or a format without the failure contact flag 184 may be used.

FIG. 5 is a flow chart of message processing from the transaction terminal in step 104 in FIG.

First, the overall control section 31 first sets the fault flag to O.
Processes are sorted according to N and OFF (step 20)
1). The failure flag is a flag representing whether or not a failure has occurred in the host computer, which is a feature of the present invention. The failure flag is set in the flag register 36. In the present embodiment, ON indicates that a failure has occurred, and OFF indicates that no failure has occurred. Although it will appear later, there are three ways to detect a host failure: response timeout from the host, failure notification from the host, and host failure notification from another branch processor.
I envision a way. When a failure is detected by these, the overall control unit 31 turns on the flag register. When the message indicating that the failure is recovered is received, the overall control unit 31 turns off the flag register 36.
And Although the flag register 36 is used here, the flag register 36 is not limited to this. Any means may be used as long as the flag can be set.

If the fault flag is OFF, the host 20
A transaction message to (step 202). The communication control unit 32 sets a timer for detecting a timeout (step 203). This timer may be either a hardware timer or a software timer. Next, the overall control unit 31 distributes the processing depending on the transaction partner of the transaction (step 204). If the customer is another sales office, do nothing. If the customer is a self-employed store, the corresponding record in the store ledger 12 is updated (step 20).
5).

In step 201, the failure flag is turned on.
If so, the process is further distributed depending on the transaction partner (step 206). If the customer is another business office, a business transaction message with the failure contact flag turned ON is transmitted to the business office (step 207). The failure contact flag is a flag that is a feature of the present invention, and is a flag for notifying the other branch office processor whether or not a failure has occurred in the host computer, and is set in the transaction message. Specifically, FIG.
It is set in the item of the fault contact flag 184 in the format as shown in FIG. In the present embodiment, ON indicates that a failure has occurred, and OFF indicates that no failure has occurred.

If the customer is another business office in step 206, the failure processing flag of the corresponding record in the business office ledger 12 is turned on (step 208), and the business office ledger 1
The corresponding record of No. 2 is updated (step 205).

FIG. 6 is a flow chart of message processing from the other sales office and host in step 105 in FIG.

First, the overall control unit 31 determines the state of the host 20 by the fault contact flag 184 (see FIG. 18) in the transaction message and distributes the processing (step 3).
01). If the failure contact flag is ON, the failure flag in the flag register is turned ON as in the above (step 302). If the failure contact flag is OFF, step 302 is skipped. Then, the corresponding record in the branch ledger 12 is updated (step 303).

Next, the processing is sorted according to the failure flag (step 304). If the failure flag is ON, FIG.
The failure processing flag of the corresponding record of the sales office ledger 12 shown in is turned on (step 305). Fault flag is OFF
If so, step 305 is skipped. Then, the communication control unit 3 sends a response message to the sender of the transaction message.
2 to transmit (step 306).

FIG. 7 is a flow chart of the time-out processing in the processing step 108 of the sales point processor in FIG.

The overall control section 31 first turns on the failure flag of the flag register (step 401). next,
The processing is distributed depending on the business partner of the message already transmitted to the host 20 (step 402). If the customer is a self-employed store, the failure processing flag of the record in the store ledger 12 is turned on (step 403). If the customer is another business office, a transaction message with the failure contact flag turned ON is created for the business office and transmitted from the communication control unit 32 (step 404).

FIG. 8 is a flow chart of processing other than the transaction and response message in step 107 in FIG.

The overall control unit 31 first sorts the processing depending on whether the received message is a failure notification from the host (step 501). If it is not a failure notification from the host, the processing is sorted according to whether it is a recovery notification from the host (step 502). If it is not a recovery notification, the message is processed (step 505). If it is a recovery notification, a record in which the failure processing flag of the branch ledger 12 is ON is transmitted to the host 20 via the communication control unit 32 (step 503). All the failure processing flags of the sales office ledger 12 are turned off (step 504).

If the message received in step 501 is a failure notification from the host, the overall control section 31 turns on the failure flag (step 506). Next, the process is sorted according to whether or not there is a response message to the host that is not responding (step 50).
7). If not, end. If there is, the processing is further distributed depending on the transaction partner of the transaction message (step 50).
8). If the customer is a self-employed shop, the failure processing flag of the corresponding record in the shop ledger 12 is turned on (step 5).
09). If the customer is another business office, the communication control unit 32 sends a business transaction message in which the failure contact flag is turned ON to the business office.
(Step 510).

Next, the processing on the host side will be described.
FIG. 9 is a normal processing flow of the host.

The host 20 first receives a transaction message (step 601). Then, the corresponding record in the ledger 22 is updated (step 602), and a response message is sent to the sender (step 603). Next, the process is sorted depending on whether the customer is the sender (step 60).
4). If the customer is the sender, processing is over. If it is not the transmission source, the transaction message is transmitted to the sales office (step 605). The response message from the business point is received (step 606), and the process is ended.

FIG. 10 is a processing flow of the host at the time of failure recovery.

The host 20 first performs recovery processing for the host itself (step 701). Next, a failure recovery notification is transmitted to each branch office processor 10 (step 702). The record of the sales office ledger 12 processed only by the sales office at the time of failure is received from each sales office processor 10 (step 7).
03). For the received record, the ledger 22 is updated to match with the branch ledger 12 (step 70).
4). After that, the process returns to the normal process (FIG. 9).

The data flow will be described with reference to FIGS. 11 to 15 using the overall system configuration. Also, the flow of the steps of the flow described so far will be described together. In these figures, broken lines indicate the flow of data,
Circled numbers indicate flow order.

FIG. 11 shows the flow of data when the host is normal and the customer is a self-employed store. Sales office A
A transaction message is sent from the transaction terminal 11 of the above to the sales office processor 10 (). Branch processor 10 to host 2
The transaction message passes to 0 (). The branch ledger 12 is updated (). The ledger 22 of the host is updated (). The response message returns from the host 20 to the branch office processor 10 (). The response message is the transaction terminal 11
Cross over (). When this is expressed in correspondence with the steps of the flow shown in FIGS. 4, 5, and 9, 101 is 102 → 103 → (104) → 201 → 202 → 203.
→ 204, 205, 601 → 602 → 603 →
604 becomes 101 → 102 → 106.

FIG. 12 shows the flow of data when the host is normal and the customer is another sales office. Sales office A
A transaction message is sent from the transaction terminal 11 of the above to the sales office processor 10 (). Branch processor 10 to host 2
The transaction message passes to 0 (). The ledger 22 of the host is updated (). The response message returns from the host 20 to the branch office processor 10 (). The response message is passed to the transaction terminal 11 (). A transaction message is passed from the host 20 to the sales office processor 10 of the sales office (sales office C) (). The branch ledger 12 is updated (). The response message returns from the branch processor 10 to the host 20 (). When this is expressed in correspondence with the steps of the flows shown in FIGS. 4, 5, 9 and 6, it is 101,
Is 102 → 103 → (104) → 201 → 202,
Is 601 → 602, is 603, is 101 → 102 →
106 is 604 → 605, is 101 → 102 → 1
03 → (105) → 301 → 303 → 304
6 → 606.

At this time, if a failure occurs in the host 20 at the time of, a time-out occurs at step 101 of. Then, as shown in FIG. 4 and FIG.
The sequence of steps is 01 → (108) → 401 → 402 → 404. The flow is the same as that described below with reference to FIG.

FIG. 13 shows that the host 20 has a failure and
It shows the flow of data when the customer is a self-employed store.
A transaction message goes from the transaction terminal 11 of the branch A to the branch processor 10 (). The branch ledger 12 is updated (). The response message goes to the transaction terminal 11 (). When this is expressed in correspondence with the steps of the flow shown in FIGS. 4 and 5, it is 101, 102 → 10
3 → (104) → 201 → 206 → 208 → 209,
Is 210.

FIG. 14 shows that the host 20 has a failure and
This shows the flow of data when the business partner is another sales office.
A transaction message goes from the transaction terminal 11 of the branch A to the branch processor 10 (). The transaction message is passed to the sales point processor 10 of the sales office (sales office C) ().
The branch ledger 12 is updated (). The response message returns from the branch processor 10 (branch C) to the branch processor 10 (branch A) (). The response message is passed to the transaction terminal 11 (). When this is expressed in correspondence with the steps of the flow shown in FIGS. 4, 5 and 6, it is 1
01, 102 → 103 → (104) → 201 → 20
6 → 207, 101 → 102 → 103 → (105)
→ 301 → 302 → 303 → 304 → 305, is 30
6 is 101 → 102 → 106.

FIG. 15 shows the flow of data at the time of failure recovery of the host 20. A failure recovery notification is sent from the host 20 to each branch office processor 10 (). A transaction message is sent from each branch processor 10 to the host 20 (). The host 20 receives the unprocessed record at the time of failure, and the ledger 22 is updated (). When this is expressed in correspondence with the steps of the flow shown in FIGS. 10, 4 and 8, is 701 → 702, is 101 → 102 → (10
7) → 501 → 502 → 503 → 504, is 703 →
It becomes 704.

Next, a case where a branch office processor communicates with another branch office processor in the event of a host failure will be described.

In this case, the sales office processor, which is about to send the transaction message to the host, refers to the terminal address table shown in FIG. 17 and communicates with the sales office processor of the customer. For example, when the host 20 fails, the branch office processor 10 of the branch office A sets the destination address 181 in the format shown in FIG. 18 of the transaction message to the terminal address B found from the terminal address table of FIG. The terminal address A, which is the terminal address of the self-employed store, is set in the original address 182. In addition, a code indicating that the transaction data is set in the message type 183 for distribution of processing in the branch office processor 10. Furthermore, the failure contact flag 18
4 is set to ON, the transaction data 185 is set, and the transaction data 185 is transmitted. As described above, the receiving side processes the sent data according to the flow shown in FIG.

The above is an example in which the present invention is applied to a bank online system, but the present invention is not limited to this. Any system may be used as long as the processing result is sent to another terminal via the host. For example, it can be applied to a system in which transactions are performed with each other via a host.

[0069]

As described above, according to the present invention, even if the host does not function for some reason, it is possible to continue the operation without stopping the system, and it is possible to improve the reliability of the system.

[Brief description of drawings]

FIG. 1 is a system outline flowchart showing the operation of the entire system of an embodiment of an online system of the present invention.

FIG. 2 is a block diagram showing a system configuration of an embodiment of the present invention.

FIG. 3 is an explanatory diagram showing a relationship between a host ledger and a sales office ledger, which are components of an embodiment of the present invention, and a failure processing flag.

FIG. 4 is a flowchart showing the overall processing of a branch office processor which is a component of an embodiment of the present invention.

5 is a detailed flowchart of message processing from the transaction terminal in FIG.

FIG. 6 is a detailed flowchart of message processing from another business office and host in FIG.

FIG. 7 is a detailed flowchart of the timeout process in FIG.

FIG. 8 is a detailed flowchart of processing other than the transaction and response message in FIG.

FIG. 9 is a flowchart showing normal processing of a host which is a constituent element of an embodiment of the present invention.

FIG. 10 is a flow chart showing a process at the time of failure recovery of a host which is a component of one embodiment of the present invention.

FIG. 11 is an explanatory diagram showing a data flow when the host, which is a component of one embodiment of the present invention, is normal and the customer is a self-employed store.

FIG. 12 is an explanatory diagram showing a data flow when the host, which is a component of one embodiment of the present invention, is normal and the customer is another sales office.

FIG. 13 is an explanatory diagram showing a data flow when the host, which is a component of one embodiment of the present invention, has a fault and the customer is a self-employed store.

FIG. 14 is an explanatory diagram showing a data flow when the host, which is a component of one embodiment of the present invention, is normal and the customer is another sales office.

FIG. 15 is an explanatory diagram showing a data flow at the time of failure recovery of a host which is a component of an embodiment of the present invention.

FIG. 16 is a block diagram showing an example of a configuration of a branch office processor.

FIG. 17 is an explanatory diagram showing an example of a terminal address table used when communication is performed between branch office processors.

FIG. 18 is an explanatory diagram showing an example of a format of a transaction message.

[Explanation of symbols]

10 ... Branch processor, 11 ... Transaction terminal, 12 ... Branch ledger, 20 ... Host computer (host), 22 ...
ledger.

─────────────────────────────────────────────────── ─── Continuation of front page (51) Int.Cl. ⁵ Identification code Internal reference number FI Technical display location G06F 15/00 320 A 7459-5L 15/16 470 S 9190-5L G07D 9/00 451 D 8513- 3E G07F 19/00 (72) Inventor Masamitsu Tenma 1099 Ozenji, Aso-ku, Kawasaki-shi, Kanagawa Stock Company Hitachi Systems Development Laboratory

Claims (18)

[Claims] 1. An online system comprising a host computer and a plurality of terminal systems connected to the host computer via a network, wherein a transaction from the terminal system is processed by the host computer, wherein the host computer includes all terminal systems. A master file for recording transaction processing results from the host computer is provided, and the host computer has a function of updating transaction processing results from all terminal systems with respect to the master file. A terminal device for outputting, a subfile for recording the processing result of a transaction related to the terminal system, a communication process for the terminal device, a communication process for the host computer, and a subfile. The subfile has a processor for performing at least update processing, and the subfile has a function of displaying a processing result of a transaction processed at the time of the failure of the host computer as a failure processing display indicating that the processing is performed at the time of the failure. An online system characterized by further having a function of sending a processing result of a transaction having a failure processing indication to the host computer when the failure of the host computer is recovered. 2. The processor according to claim 1, when the processor detects that the host computer has a failure,
An online system further comprising failure occurrence storage means for storing occurrence of a failure until failure recovery. 3. The host state determining means according to claim 2, wherein the processor refers to the fault occurrence storage means to determine whether the current state of the host computer is normal, faulty, or at the time of fault recovery. An online system with. 4. The online system according to claim 3, wherein the processor has a communication processing function for communicating with another terminal system. 5. The processor according to claim 4, when the processor detects a failure of the host computer, when sending a message to another terminal system, the processor adds information for notifying the failure of the host computer to the message, An online system that has the function of reporting host failures to other terminal systems. 6. The online system according to claim 2, 3, 4 or 5, wherein the processor has a timer function for monitoring whether or not a response from the host computer is within a predetermined time after sending a message to the host computer. 7. The master file according to claim 2, 3, 4 or 5, wherein the master file is a ledger for recording the transaction results of all transaction accounts, and the sub-file is a sales business for recording the transaction results of the transaction accounts in each terminal system. An online system used for bank transactions that is a store ledger. 8. An online system comprising a host computer and a plurality of terminal systems connected to the host computer via a network, wherein a transaction from the terminal system is processed by the host computer, wherein the host computer includes all terminal systems. There is a master file that records the transaction processing results from the terminal system, and a subfile that records the transaction processing results related to the terminal system is provided in each terminal system. If there is, update the master file and
Update the subfile of the terminal system related to the transaction, update only the subfile of the terminal system related to the transaction when the host computer fails, and match the subfile with the master file when the failure of the host computer is recovered. A method of operating an online system characterized by performing the following. 9. An online system according to claim 8, wherein a flag indicating that the transaction processing result updated in the subfile at the time of host computer failure is updated at the time of host computer failure is set for each record. How to operate the system. 10. The fault recovery system according to claim 9, wherein the fault recovery is transmitted to each terminal system when the host computer recovers from the fault,
An operation method of an online system, characterized in that a master file is updated according to a transaction processing result sent from each terminal system, and a sub file and a master file are matched. 11. The method of operating an online system according to claim 10, wherein when the failure of the host computer is recovered, the transaction processing result is sent from each terminal system only for the record in which the flag is set. 12. A method according to claim 8, wherein when a failure of the host computer is detected, a flag indicating a failure of the host computer is set in the terminal system, and the mode of processing for the generated transaction is distributed in the state of this flag. A method of operating an online system characterized by. 13. The terminal system according to claim 8, wherein the terminal system is arranged for each business office of a bank, the master file is a ledger for recording transaction results of all transaction accounts, and the subfile is
An operating method of an online system used for banking transactions, which is a business ledger for recording transaction results of a transaction account in each terminal system. 14. The terminal system according to claim 13, which has detected a failure of the host computer, sends a transaction message to a terminal system of another sales office, in the message,
Give information to notify the failure of the host computer,
A method of operating an online system that informs other terminal systems of host failures. 15. The terminal system according to claim 14,
If the host computer is normal, the transaction message related to the input transaction is sent to the host computer, and if the host computer is in a failure state, if the transaction is a transaction related to the terminal system of another branch office, the terminal system. If the transaction is related to the terminal system of the self-employed store, the process should be distributed so that the sub file of the terminal system of the self-employed store is updated with the transaction data included in the transaction message. Characteristic online system operation method. 16. When a failure of the host computer is detected according to claim 15, a flag indicating the failure of the host computer is set in the terminal system, and the processing is distributed according to the setting state of the flag. A method of operating an online system characterized by. 17. The host computer according to claim 13, 14, 15 or 16, updating the ledger based on the transaction message received from each terminal system, and sending a response message to the terminal system which transmitted the transaction message. An operating method of an online system, wherein the transaction message is transmitted and the transaction message is transmitted to a terminal system of a corresponding sales office when the transaction is not a transaction message transmission source. 18. A terminal system for constructing an online system of a type which is connected to a host computer via a network and transmits a generated transaction to the host computer for processing, and a terminal device for inputting / outputting a transaction. ,
The subfile includes a subfile that records the processing result of a transaction related to the terminal system, a processor that performs at least a communication process for the terminal device, a communication process for the host computer, and an update process for the subfile. The processor has the function of displaying the processing result of the transaction processed at the time of failure in the processing of the failure processing indicating that it was processed at the time of the failure. To a host computer.