JPH08249281A - Online processing system - Google Patents

Abstract

PURPOSE: To improve the reliability of a system by providing alternative computers not for every computers system but a smaller number of alternative computers than the number of the computer system so as to attain the backup of the computer systems at the time or a fault and stop. CONSTITUTION: Each online system consists of a computer system and a terminal system and each computer system is connected to a server 6 through a network 3. The updating result of a file is reflected to a master file 7 and contents in the file are unitarily managed by the master. The monitor part 9 of the server 6 monitors the state of the computer system, and if being in the state of a fault and stop, the part 9 makes the alternative computer 8 take a turn in the processing of the computer system. A line switching part reconnects the terminal system from the computer system to the alternative computer 8.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION The present invention relates to an online processing system in which each online system is composed of a plurality of terminal systems connected to a computer system via a network, and the computer systems are connected to each other via a network. The present invention relates to an online processing system that continues processing by switching to a replacement computer provided in a center when one or more computer systems are in a failure stop state.

[0002]

2. Description of the Related Art FIG. 11 shows the configuration of a conventional online processing system. The online processing system shown in the figure
It consists of multiple online systems. That is, the first online system 70a includes a computer system 72, a plurality of terminal systems 71 connected via a network 73, and a file 74. The second and third online systems 70b and 70c are also configured similarly to the first online system 70a.

Each computer system 72 is connected to a relay computer 75 via a network 73. When such an online processing system is applied to a bank system, the relay computer 75 becomes the Zengin Center.
Each online system processes the transaction from the terminal system in each computer system and updates the file. On the other hand, for example, when a transaction from the terminal system 71 in the first online system 70a updates the file 74 held by the computer system 72 in the third online system 70c, the transaction from the terminal system 71 is transmitted via the network. 7
3. The file 74 is updated by being processed by the computer system 72 of the third online system 70c via the relay computer 75. In such an online system, recovery measures for a failure of each computer system have been individually performed for each computer system.

By the way, in a bank system or the like to which the online system as described above is applied, the processing is decentralized, the files are distributed and managed between the host computer and the branch office processor, and the file is processed on the branch office processor side. The system is configured so that it can. Therefore, it is necessary to improve the reliability of the branch office processor and improve the failure recovery method of the branch office processor.

Various methods have heretofore been proposed in order to realize the high reliability of the online system. For example, a plurality of computer systems each have an input message storage area, input messages processed online on other computer systems are stored in the storage area, and mutual failure monitoring of the computer systems is monitored, and one computer system fails. When stopped, the other computer system processes the input message stored in the input message storage area, and then switches the terminal device connected to the failed computer system to the other computer system. There is an online processing system switching method that employs a hot standby method for connection (Japanese Patent Laid-Open No. 63-15).
See 8662).

Another example is an operation method of an online system such as a bank system. That is, in this online system, 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 the ledger of the branch office account as a subfile. If the host is normal when a transaction input occurs at the transaction terminal of the sales office, the host ledger and the sales office ledger of the supplier are updated.If the host is abnormal, the sales office ledger of the supplier is updated. By updating and reconciling the host ledger and each sales office ledger when the host recovers,
Even if the host computer fails, the online processing is continued by the file of the computer on the terminal system side (see JP-A-5-216908).

[0007]

However, the above-mentioned first problem
The technique requires a waiting computer for each host computer, so when this is applied to an online processing system in which computer systems are connected via a network, a replacement computer is provided for each computer system. There is a problem that it must be prepared and requires resources such as a lot of hardware.

Further, the above-mentioned second technique does not take into consideration the measures to be taken when the computer on the terminal system side fails. Furthermore, when a failure occurs in the host computer and a transaction of one terminal system computer is a transaction for a file of another terminal system computer, the terminal system computers mutually communicate in order to process such transaction. It must be connected and configured to communicate, for which CCITT Recommendation X. VC (Virtual Call) defined in 25 is a connection method between terminals that establishes a virtual line with a communication partner prior to communication and releases it when communication ends. There is a problem that requires.

An object of the present invention is to provide a backup computer of a computer system at the time of failure stop by providing a backup computer of a number smaller than the number of computer systems in the center without providing a backup computer for each computer system. It is to provide an online processing system with improved system reliability.

[0010]

In order to achieve the above object, according to the present invention, a computer system, a plurality of terminal systems connected to the computer system via a network, and a transaction related to the terminal system are provided. A file for recording a processing result, in an online processing system in which a plurality of computer systems are mutually connected via a network, a server connecting the computer systems via the network, and connected to the server,
A master file that manages the contents of each of the files, and the server and the master file, which are connected to the master file and take over the operating environment of the computer system in the faulty stopped state, and alternate processing.
The number of replacement computers is less than the number of computer systems, the server comprises means for monitoring the operating state of each computer system and means for storing the operating environment of each computer system, and the replacement computer Instructing switching of the terminal system connected to the computer system in the failure stop state to the replacement computer in response to the detection of the failure stop state of the one or more computer systems by the monitoring means. A means for instructing to switch the connection of the terminal system connected to the substitute computer to the computer system in which the failure is recovered, in response to the detection of the failure recovery in the computer system in which the failure has stopped It is characterized by having.

According to the present invention, the server transfers the update information in the master file processed by the substitute computer to the computer system in which the failure has been restored, in response to the recovery of the computer system in which the failure has stopped. It is characterized by having means for transferring.

[0012]

Each online system comprises a computer system and a terminal system, and each computer system is connected to a central transaction server via a network. The transaction that occurred in the terminal system, after the normal processing, when the file is registered and updated, the update result is the central transaction
It is reflected in the master file provided in the server, and the contents of each file are centrally managed by the master file. When the transaction generated in the terminal system is for a file of another computer system, the transaction is transferred to the other computer system via the central transaction server to provide the service. The monitoring unit of the central transaction server monitors the state of the computer system, and when a failure stop state of the computer system is detected, the processing of the computer system in the failure stop is replaced by the substitute computer, and the failure stop is detected. Replace the terminal system connected to the new computer system with the replacement computer so that the processing from the terminal system can be continued.

[0013]

DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of the present invention will be specifically described below with reference to the drawings. 1 and 2 show the system configuration of an embodiment of the present invention. 1 and 2 are part of a divided system configuration, respectively, and are a general view by connecting FIG. 1 and FIG. 2 with “a” and “b” respectively shown in the figure. The system of the present invention includes multiple online systems (eg,
System A and B). Online system A
The point that a plurality of terminal systems 1a, a computer system 2a, and a file 4a are provided is similar to the conventional example described in FIG. The online system B has the same configuration as A. In the figure, a two-system online system is shown for simplification of description, but this embodiment is applied to an n-system online system.

When the online processing system of the present invention is applied to a bank system, the online system A
Is the system of Bank A and the online system B is B
Bank system, each terminal system is AT
The terminal system of the system A includes M, CD, PC, controller, and the like. For example, each terminal system of the system A is provided in the Tokyo store, the Osaka store, or the like.

The computer systems 2a and 2b are connected to a central transaction server 6 (hereinafter referred to as a server) via a network (dedicated line) 3 (the network of the present embodiment is a dedicated line, and therefore, the above-mentioned description is omitted). It does not require the VC function.

The server 6 is connected to a master file 7 for recording the results of transactions from the terminal systems 1a and 1b, and a substitution computer 8 for altering the processing when the computer systems 2a and 2b fail. 8 and the master file 7 are directly connected. Further, the server 6 is managed by a monitoring unit 9 that monitors the operating states of the computer systems 2a and 2b, a state management table 10 that is managed by the monitoring unit, a control unit 11 that controls the entire system, and a control unit. , A table 12 that stores operating environment recording addresses, an operating environment storage unit 13 that stores in advance the operating environment (operating system, business program, etc.) of the computer system, and receives update data from the computer system, or It is composed of a master file updating unit 14 for transmitting update data.
Further, the computer systems 2a and 2b are provided with file update units 5a and 5b, respectively.
a and 5b transfer the updated data to the server 6.

The replacement computer 8 is provided with a line switching instruction section 15. This line switching instruction unit 15
Is a network (public line such as ISDN) 3a, 3
Line switching unit 16 of online system A via b
a and the line switching unit 16b of the online system B, the terminal system 1a is connected to the line switching unit 16b.
a is connected to the computer system 2a via the network 3, and the terminal system 1b is connected to the computer system 2b via the line switching unit 16b and the network 3. Then, the line switching instructing unit 13 changes the connection between the terminal systems 1a and 1b connected to the computer systems 2a and 2b in which the failure has occurred to the replacement computer 8, and when the computer system in which the failure has stopped is restored, Line switching units 16a, 1 so that they are connected to
Instruct 6b.

FIG. 5 shows the structure of the status management table managed by the monitoring unit. In this table, the identifier 30 of the computer system 2, the computer address 31 for communication, and the computer system 2 are periodically inquired, and whether the computer system is normal or abnormally stopped by response monitoring, State management information 32 representing the monitoring state is recorded.

That is, the monitoring section 11 periodically makes an inquiry (polling) to the computer system 2 from the initial state, and when there is a response from the computer system 2, records "normal" as the state management information 32, If there is no response during status or "normal" status, status management information 3
When "failure" is recorded as 2 and the process is switched to the substitute computer due to "failure", "abnormal" is recorded as the state management information 32, and the computer system is periodically inquired in the "abnormal" state ( Polling), and when there is a response from the computer system, "recovery" is recorded as the state management information 32, and when the process is returned to the computer system recovered from the substitute computer by "recovery", the state management information “Normal” is recorded and managed as 32.

FIG. 9 shows the state transition of the state management information 32. In addition, the state management information 32 of the replacement computer has two states, that is, the initial state and the “replacement of the faulty computer system” which is a state where another computer system is being replaced. The state transition is as shown in FIG.

FIG. 6 is a master file 7 recording the processing results of transactions from the terminal systems 1a and 1b.
Contents and each file 4 of computer system 2a, 2b
It is a figure which shows the content of a and 4b. Files 4a, 4b
The account number 40 traded from the terminal systems 1a and 1b and various items 41 subordinate to the account number are stored in. In the master file 7, the account number 40 and
In order to record various items 41 depending on the account number, a processing flag 42 that records that the replacement computer has performed an update process when the computer system is "abnormal", and in which computer system the account number 40 exists. The computer address 31 of is stored.

FIG. 7 shows the configuration of a table which is managed by the control unit and stores operating environment recording addresses. In this table, the computer address 31 of the computer system 2
And an operating environment recording address 50 for referring to an operating environment storage unit that stores the operating environment of the computer system.

FIG. 8 shows an address table of the line switching unit, which is managed by the alternate computer, and in the table, the computer address 31 of the computer system 2 and
Each line switching unit 9a, 9 corresponding to the computer address
The address 60b is written.

The operation of the present invention will be described below with reference to the drawings. FIG. 3 is a flow chart for explaining the processing operation of the system of the present invention when the computer system 2 (a, b) is normal.

Computer system 2 via network 3
When a transaction is input (occurrence of transaction) in the terminal system 1a connected to (for example, 2a) (step 10
1), the terminal system 1 that generated the transaction
The a transfers the transaction to the computer system 2a connected via the network 3.
The computer system 2a looks at the computer address sent together with the transaction and determines whether or not it is processing of the file 4a of its own computer (step 10).
2).

If it is the file processing of the own computer, the transaction is processed in the own computer system 2a, and if the file 4a of the own computer is updated, the file is updated (step 103). As a result of the determination in step 102,
For file processing of another computer (for example, the computer system 2b), the computer system 2a sends the transaction to the server 6.

The control unit 11 of the server 6 looks at the computer address sent together with the transaction and transfers the transaction to the computer system 2b having the transaction file of the transaction (step 10).
4). Next, the transaction is processed in the computer system 2b, and if the file 4b of the computer is updated, the file is updated (step 105).

Finally, only when the file 4 of the computer system is updated, the computer system 2
The file update result is transferred to the server 6 and the master file 7 is updated. For example, when the computer system 2a updates the file 4a in step 103, the system A
The file update unit 5a of 5 transfers the update data to the server 6. The master file update unit 14 of the server 6 receives the update data, writes the data in the master file 7, and notifies the file update unit 5a of the system A of the completion of the write, thereby ending the update processing (step 106).

By the above processing, the contents of the files 4 of the plurality of computer systems 2 are recorded in the master file 7 of the server 6 in a unified manner.

FIG. 4 is a flow chart for explaining the processing operation of the system of the present invention when a certain computer system is stopped due to a failure and when the failure is recovered. When the monitoring unit 9 of the server 6 is in the monitoring time zone (step 3
01), the computer systems 2a and 2b connected via the network 3 are periodically inquired (polled), and whether the computer system is in a normal state or a failure stop state is monitored by response monitoring (step 302). ).
For example, when the computer system 2a does not respond to the inquiry from the monitoring unit 9, the monitoring unit 9 detects a failure stop of the computer system 2a.

When the monitoring unit 9 detects a failure stop of the computer system 2a, the status management information 32 of the computer system 2a (A) shown in FIG. 5 is rewritten to "failure", and the control unit 11 causes the computer in which the failure has occurred. The replacement computer 8 that replaces the system 2a is online connected to the server 6 (step 303). Further, the control unit 11 refers to the table of FIG. 5, and the status management information 32 of the computer system 2a (A) is “fault”, so that the other computer system 2
The transaction for the file 4a of the failure computer system 2a is stopped from b.

Subsequently, the control unit 11 reads out the operating environment (operating system, business program, etc.) of the failure computer system 2a from the operating environment storage unit 13,
The replacement computer 8 is loaded, and the replacement computer 8 is started (step 304). That is, when the system 2a whose computer address is the address A is stopped due to a failure, the control unit 11 reads the operating environment recording address (0140) of the computer address A from the table of FIG. The operating environment of the computer system 2a is read from the designated operating environment storage unit 13 and loaded into the substitute computer 8.

At the time when the substitute computer 8 becomes operable, the monitoring unit 9 rewrites the state management information 32 of the fault computer system 2a to "abnormal" and changes the state management information 32 of the substitute computer 8 to "fault computer". System 2a (A) replacement ”
Rewrite Therefore, the control unit 11 refers to the state management table of FIG. 5, and the state management information 32 of the faulty computer system 2a is "abnormal", so that the file 4a of the faulty computer system 2a is stored in another computer system 2b. The transaction will be processed by the replacement computer 8 which is the "replacement of the faulty computer system".

Next, in step 305, the line switching instructing section 15 of the substitution computer 8 makes the line switching section 1
For 6 (for example, composed of changeover switch etc.),
Instruct to switch the line. That is, the replacement computer 8
Refers to the address table of the line switching unit (FIG. 8) and reads the address of the line switching unit 9a of the computer address A in which the failure has stopped. In this example, assume that 34521001 (Osaka store 001) and 34522002 (Kobe store 001) are read as the addresses of the line switching unit 9a. Each of these stores corresponds to each terminal system 1a in FIG.

When the substitution computer 8 gives the read address to the line switching instructing section 15, the line switching instructing section 15 sends the address 34521001 + switching instruction and the address 34522002 + switching instruction to the network 3a. As a result, each line switching unit 16a is selected, and the connection of the terminal system 1a is switched from the computer system 2a to the replacement computer 8 by the switching command, and thus each terminal system 1a connected to the fault computer system 2a. Is connected to the substitution computer 8.

Then, the transaction from the terminal system 1a is processed by the substitution computer 8, and the update processing is performed by the master file 7 in place of the file 4a of the faulty computer system 2a. Then, as described with reference to FIG. 6, when the computer system 2a is "abnormal", the processing flag 42 that records that the replacement computer 8 has performed the update processing is turned on. If the transaction from the terminal system 1a is a file process of another computer system, the transaction is transferred to the other computer system (for example, 2b) via the server 6 and processed as described above.

As described above, the processing when the fault of the fault computer system 2a is recovered while the substitution computer 8 is performing the substitute process of the fault computer system 2a will be described. In step 302, the monitoring unit 9 sets the computer system 2
When the failure recovery status of a is detected, the status management information 32 of the failure computer system 2a of FIG. 5 is rewritten to "recovery". Then, the control unit 11 is previously provided with the computer system 2a.
Automatically recovers when the
Alternatively, it is specified whether or not to wait for the restoration until a manual instruction is given, and the control unit 11 confirms the restoration instruction (step 306). Note that the reason for not performing all automatic restoration here is to prevent the online processing of the computer system from stopping during the time before the automatic switching.

When the automatic or manual recovery instruction is confirmed, the control unit 11 transfers the failure recovery computer (2a) address (A) to the master file updating unit 14. The master file update unit 14 refers to the master file 7 of FIG.
The file data of the failure recovery computer (2a) address (A) in the master file 7 is locked, and access to the file data of the failure recovery computer 2a is prohibited.

Further, the master file updating unit 14 sets various item data 41 in the file data of the failure recovery computer 2a in the master file 7 in which the processing flag 42 is set to the ON state, that is, the updated item data 41. , To the file updating unit 5a of the failure recovery computer 2a via the network 3. As a result, the computer system 2a
The data in the master file 7 updated during the failure period is copied to the file 4a of the failure recovery computer 2a.
Then, the master file updating unit 14 resets the processing flag 42 in the master file 7 that has been copied to the file 4a to the off state (step 307).

Next, the control unit 11 disconnects the substitution computer 8 from the server 6 (step 308). When the disconnection process ends, the control unit 11 causes the master file update unit 1
4 is instructed to unlock the file data of the failure recovery computer 2a in the master file 7. By this cancellation, the failure recovery computer 2a can execute the original online processing, and the monitoring unit 9 rewrites the state management information 32 of the computer system 2a to "normal".

Further, the monitoring unit 9 causes the replacement computer 8 to be separated from the server 6 by the disconnection of the replacement computer 8 from the server 6.
The state management information 32 of is rewritten to the “initial state”. Then, the line switching instructing unit 15 sends the address of the line switching unit 16 and the switching command corresponding to the computer address 31 of the failure recovery computer to the network 3a as described in FIG. In the same manner as described above, the line switching unit 16a transfers the terminal system 1a connected to the network 3a side to the network 3 side, that is, the computer system 2a in the failure recovery state from the replacement computer 8.
The connection of the computer system 2a is switched to, and the recovery of the online processing of the computer system 2a is completed (step 309). If it is in the monitoring time zone, the process returns to step 302 again to continue monitoring the computer system.

In the system configuration shown in FIG. 1, the line connecting the server 6 and the computer system 2 is composed of one line. However, in consideration of line failure and communication equipment (modem, communication control device) failure. Therefore, it is desirable to monitor the line by providing a plurality of routes. In other words, if some routes of multiple lines are normal and the other routes are abnormal, it is not a computer system error but a communication route error.If all routes are abnormal, first determine that the computer system is abnormal. As a result, erroneous determination of network failure and computer system abnormality can be avoided. Also, the server 6
It is desirable that the communication device (modem, communication control unit) on the side be completely duplicated. By completely duplicating the communication device on the local side, even if one communication device on the local side fails and communication with the computer system is disabled, it is erroneously determined to be a fault on the computer system. Is prevented.

Further, in the above-mentioned embodiment, the case where the number of replacement computers is one has been described, but the present embodiment is not limited to this, and in consideration of the failure of a plurality of computer systems, the number of resources is small. In order to be able to cope with a failure, for example, for n computer systems, m (where n> m) replacement computers are provided on the center side, and The processing can be configured to be processed by a plurality of alternation computers. Further, the substitute computer can be configured to perform other tasks when it is offline, or can have a hot standby configuration in which it is online connected to the server, and various changes can be made.

Further, in the above-mentioned embodiment, the example of the replacement computer of the computer system in a plurality of bank systems is explained. However, in one bank system, files are distributed and managed between the host computer and the business office processor, and the business office is operated. It can also be applied to the replacement of a branch office processor in a system that allows processing on the processor side. Further, the processor may be a workstation or a personal computer, and the network may be a LAN.

[0045]

As described above, according to the present invention, a common backup computer for backup is provided on the center side for a computer system that constitutes a plurality of online systems, so that it requires less resources. It is possible to cope with the failure stop of the computer system and improve the reliability of the online processing system. Also,
The files of each computer system are centrally managed in the master file, and the files of the computer system in the failure stop are replaced by the master file, and the update data in the master file is copied at the time of failure recovery. Online processing can be continued without stopping.

[Brief description of drawings]

FIG. 1 shows a part of a system configuration of an embodiment of the present invention.

FIG. 2 shows the remaining part of the system configuration of the embodiment of the present invention.

FIG. 3 is a flow chart for explaining the processing operation of the system of the present invention when the computer system is normal.

FIG. 4 is a flowchart for explaining the processing operation of the system of the present invention when a computer system is stopped due to a failure and when the failure is recovered.

FIG. 5 shows a configuration of a status management table managed by a monitoring unit.

FIG. 6 is a diagram showing the contents of a master file and the contents of each file of a computer system.

FIG. 7 shows a configuration of a table storing operating environment recording addresses.

FIG. 8 shows an address table of a line switching unit.

FIG. 9 shows a state transition of state management information.

FIG. 10 shows a state transition of state management information in the substitution computer.

FIG. 11 shows a configuration of a conventional online processing system.

[Explanation of symbols]

 1a, 1b Terminal system 2a, 2b Computer system 3, 3a Network 4a, 4b File 5a, 5b File updating unit 6 Central transaction server 7 Master file 8 Replacement computer 9 Monitoring unit 10 State management table 11 Control unit 12 Operating environment Recording address table 13 Operating environment storage unit 14 Master file updating unit 15 Line switching instruction unit 16 Line switching unit

Claims (2)

[Claims] 1. A computer system comprising: a computer system; a plurality of terminal systems connected to the computer system via a network; and a file for recording processing results of transactions related to the terminal system. In an online processing system connected to each other via a network, a server connecting between the computer systems via the network, and connected to the server,
A master file that manages the contents of each of the files, and the server and the master file, which are connected to the master file and take over the operating environment of the computer system in the faulty stopped state, and alternate processing.
The number of replacement computers is less than the number of computer systems, the server comprises means for monitoring the operating state of each computer system and means for storing the operating environment of each computer system, and the replacement computer Instructing switching of the terminal system connected to the computer system in the failure stop state to the replacement computer in response to the detection of the failure stop state of the one or more computer systems by the monitoring means. A means for instructing to switch the connection of the terminal system connected to the substitute computer to the computer system in which the failure is recovered, in response to the detection of the failure recovery in the computer system in which the failure has stopped An online processing system characterized by having. 2. The server has means for transferring the update information in the master file processed by the substitute computer to the computer system in which the failure has been recovered, in response to the recovery of the computer system in which the failure has stopped. The online processing system according to claim 1, further comprising: