JPH09212401A - Fault recovery method for decentralized data base, and data base management device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To make a master data base, which has plural data bases and a master data base having the same contents and is reliable, recover from a fault. SOLUTION: If a fault occurs to the master data base 5, backup data are copied to the master data base 5 from an external storage device 4 according to the indication from a maintenance device 3, one of the data bases corresponding to plural processors 2-1 to 2-n is selected, and data are transferred and copied to a copy data base 6. Then a comparing process part 8 compares the backup data of the master data base 5 with the data of the copy data base 6, stores their difference information in a difference information storage part 7, and corrects the backup data of the master data base 5 according to the difference information to make a recovery from the fault.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a distributed database failure recovery method and a database management apparatus in which a plurality of databases having the same content are distributed and arranged. In a system that accesses a database and executes various processes, the frequency of access to the database increases as the system scale increases, so waiting for access to the database occurs and the overall processing speed decreases. I have a problem to do. Therefore, a configuration is conceivable in which a plurality of databases having the same content are distributed and arranged so that access requests are distributed. In such a system, it is necessary to improve the efficiency of recovery when a failure occurs.

[0002]

2. Description of the Related Art FIG. 18 is an explanatory diagram of a conventional intelligent network system, and 51 is a service management system (SMS).
m) and 52 are service control points (SCP; Services)
Control Point, 53 is a database, 54 is a service switching point (SSP; ServiceS)
witching Point), 55 is a telephone subscriber.

This intelligent network system manages a service switching point 54 corresponding to a switching center which makes a call connection and a database 53 which stores subscriber data, and controls the call connection by the subscriber data. The case where the service control point 52 and the service management system 51 that updates the database 53 in response to a subscriber data change request or the like is shown.

For example, in response to a call connection request from the telephone subscriber 55, the service switching point 54 transfers the telephone number input by the calling subscriber to the service control point 52, and the service control point 52 transmits the database 53. To identify the called subscriber and transfer the called subscriber's data to the service switching point 54. The service switching point 54 makes a call based on the data of the called subscriber and, by detecting a response thereto, establishes a connection between the calling subscriber and the called subscriber.

Further, when the number of accommodated telephone subscribers increases and the service contents are diverse, the frequency of access to the database 53 increases remarkably, and it becomes difficult to operate by one service control point 52. It is conceivable to provide service control points 52 to distribute the processing.

Therefore, as shown in FIG. 19, a plurality of service control points (SCP) 52-1 to 52-3 are provided, and databases 53-1 to 53-4 storing the same subscriber data are stored in each of them. The service switching point (SSP) 54 is provided with a database 53 in response to a call connection request from a telephone subscriber (not shown).
Access is distributed to each of -1 to 53-3,
Call processing can be performed. Therefore, it is possible to prevent waiting for access to the database.

In this case, each service control point 52-
It is very difficult to perform the synchronization processing between the databases 53-1 to 53-3 due to the processing capacity of 1 to 52-3 and the variation of the installed number or the used number. Therefore, the service management system (SMS) 51, the database 53
A master database 53-4 storing the same subscriber data as that of -1 to 53-3 is provided, and the subscriber data of the master database 53-4 is updated in response to a subscriber data update request. Master database 53
-4 is downloaded to each of the databases 53-1 to 53-3 under the control of the service management system 51, and the synchronization processing between the databases 53-1 to 53-3 is performed.

Also, an external storage medium 56 such as a magnetic tape device is provided, and a master
A configuration for backing up the database 53-4 can be adopted. This backup is generally performed regularly.

[0009]

Problems to be Solved by the Invention Databases 53, 5
Although 3-1 to 53-4 are configured by a storage device such as a magnetic disk device, a failure may occur in which stored data is destroyed. For example, FIG.
If any one of the databases 53-1 to 53-3 in No. 9 fails, the master database 53
-4 is alive, the master database 53
The fault can be recovered by downloading from -4.

When a failure occurs in this master database 53-4, the backup external storage medium 5
Data from 6 can be used to restore. However, this backup data is not always the data immediately before the occurrence of a failure. Therefore, if failure recovery of the master database 53-4 is performed using the data from the external storage medium 56, the database 53-1
A case different from the content of 53-3 occurs. The present invention provides the latest data in the event of a master database failure.
In addition, the purpose is to restore the distributed databases in a synchronized state.

[0011]

DISCLOSURE OF THE INVENTION A distributed database failure recovery method according to the present invention comprises (1) a master database 5
In a failure recovery method for a distributed database having a plurality of databases 9-1 to 9-n whose update is controlled so as to be the same as the contents of the above, according to an instruction from the maintenance device 3 when a failure occurs in the master database 5. Data of, for example, the database 9-1 of the plurality of selected databases 9-1 to 9-n is transferred to the master database 5 via a transfer device (not shown), and the failure of the master database 5 is recovered. It includes the process of making.

(2) A database management apparatus 1 having a master database 5, a copy database 6 and a difference information storage unit 7, and an external unit connected to the database management apparatus 1 for backing up the master database 5. Databases 9-1 to 9 that are connected to the storage medium 4 and the database management device 1 and that are updated and controlled so as to have the same contents as the master database 5 respectively.
A plurality of processing devices 2-1 to 2-n each having -n and a maintenance device 3 connected to the database management device 1 are provided, and when a failure occurs in the master database 5, an external device is provided according to an instruction from the maintenance device 3. Backup data is copied from the storage medium 4 to the master database 5, and data of a single or a plurality of databases selected from the plurality of databases 9-1 to 9-n is copied to the copy database 6, and this copy is made. The data of the database 6 and the backup data of the master database 5 are compared by the comparison processing unit 8, the difference information of the comparison processing result is stored in the difference information storage unit 7, and the master database is based on this difference information. 5 includes the process of correcting the backup data copied in 5.

Further, (3) a step of copying the backup data from the external storage medium 4 to the faulty master database 5 based on the time information registered in the clock management table can be included.

(4) A process of transferring and copying the data of the database selected from the plurality of databases 9-1 to 9-n to the copy database 6 based on the time information registered in the clock management table. Can be included.

(5) Based on the time information registered in the clock management table, the data copied from the database to the copy database 6 and the master database 5
It may include a process of comparing with the backup data copied to.

(6) The last update date information according to the contents of the master database 5 is added to the databases 9-1 to 9-n of the processing devices 2-1 to 2-n, and the failure of the master database 5 occurs. At the time of occurrence, upon request from the maintenance device 3, the processing device name or database name including the last update date information is transferred and displayed, and single or multiple databases are specified based on the displayed last update date, and copied. It may include a process of transferring data to the database 6.

Also, (7) a step of collecting and comparing the last update date information and automatically selecting the database of the latest update date can be included.

(8) A database for transferring data to the copy database 6 is designated from the maintenance device 3 in advance,
When a failure occurs in the master database 5, the maintenance device 3
It is possible to include a process of transferring data from a predesignated database to the copy database 6 according to a transfer instruction from.

(9) When a failure occurs in the master database 5, a plurality of databases are selected and data is copied to the copy database 6 corresponding to the plurality of databases. Comparison processing is performed with the copied backup data, difference information is stored in the difference information storage unit 7 corresponding to a plurality of databases, and copied to the master database 5 based on information obtained by logical processing of the difference information corresponding to the plurality of databases. The process of modifying the backup data may be included.

(10) The last update date information of the database is added to the difference information stored in the difference information storage unit 7,
A process of designating a date or a range of dates and modifying the backup data copied to the master database 5 based on the difference information within the date or the range of dates can be included.

The database management apparatus of the present invention is (1
1) A master database 5 that controls updating so that the plurality of databases 9-1 to 9-n have the same contents.
In the database management device 1 equipped with
An input / output device to / from the external storage medium 4 for backup of the database 5, a transfer device for transferring data from a specified database when a failure occurs in the master database 5, and a transfer device for transferring data The copy database 6 for copying the copied data, the comparison processing section 8 for comparing the data copied to the copy database 6 and the backup data copied to the master database 5, and the comparison processing section 8 for comparison. A difference information storage unit 7 that stores the difference information of the result and a control unit that controls each unit based on the instruction information from the maintenance device 3 are provided.

Further, (12) it is possible to provide a clock management table for registering a device to be activated and time, and a clock management device for performing registration in the clock management table and controlling activation of the registration device according to the registration time.

(13) Multiple databases 9-1 to 9
A transfer for holding the specified information such as a database name previously specified by the maintenance device 3 from -n, and controlling the transfer device according to this specified information to transfer the data from the specified database to the copy database 6. An information section can be provided.

(14) A plurality of databases 9-1 to 9
With respect to the last update date information added to -n, the predetermined update date or the update date range is held in advance by the designated information from the maintenance device 3, and the transfer device is controlled according to this designated information to selectively connect the database. It is possible to provide a transfer information section for transferring the data of the selectively connected database to the copy database 6.

[0025]

DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 is an explanatory diagram of a first embodiment of the present invention, in which 1 is a database management device, 2-1 to 2-n.
Is a processing device, 3 is a maintenance device, 4 is an external storage medium, 5 is a master database, 6 is a copy database, 7
Is a difference information storage unit, 8 is a comparison processing unit, and 9-1 to 9-n are databases.

The processing devices 2-1 to 2-n have databases 9-1 to 9-n respectively storing the same data and correspond to the service control points (SCP) in FIG. The database management device 1 includes a master database 5 and a copy database 6, and corresponds to the service management system (SMS) shown in FIG. Further, the configuration for updating the data in the databases 9-1 to 9-n of each of the processing devices 2-1 to 2-n from the master database 5 by downloading or the like is omitted. Database 9-
When a failure of 1 to 9-n occurs, it can be restored by downloading from the master database 5.

Further, 1) to 6) show a recovery procedure when a failure occurs in the master database 5, and first,
1). The backup data is read from the external storage medium 4 for backup, and then one of the plurality of distributed databases 9-1 to 9-n, for example, 9-1 is selected and 2). The data in the database 9-1 is transferred and stored in the copy database 6. That is, the copy database 6 corresponds to a memory area for copying the contents of the database in the selected processing device in order to recover from the failure of the master database 5.

In this case, the data in the database 9-1 is the latest, but is not always completely correct.
The backup data held in the external storage medium 4 is
Since the data is transferred and held from the master database 5 every predetermined period, the data may not be the latest one but may be updated after the data in the master database 5 is collected. Therefore, 3). The comparison processing unit 8 compares the contents of the master database 5 and the contents of the copy database 6, and stores the difference information in the difference information storage unit 7.

Next, 4). The difference information stored in the difference information storage unit 7 is transferred to the maintenance device 3 and displayed, and 5). Data is selected according to the difference information, and 6). The data such as corrections are reflected in the master database 5. Therefore, the contents of the master database 5 should be restored to correct data based on the difference information between the latest data stored in the operating database and the backup data, and the failure of the master database 5 should be recovered. You can

FIG. 2 is an explanatory view of the second embodiment of the present invention. The same reference numerals as those in FIG. 1 indicate the same parts, 11 is an input / output device, 12 is a DB (database) matching device, and 13 is a DB. A (database) input / output device, 14 is a DB (database) comparison device, and 15 is a DB (database) transfer device. The database management device 1 manages and controls a master database 5, a copy database 6, a difference information storage unit 7, an input / output device 11 that transmits and receives data to and from the maintenance device 3, and each unit. A DB matching device 12, a DB input / output device 13 that transmits and receives data to and from the external storage medium 4, and a DB that compares the contents of the master database 5 and the contents of the copy database 6.
Comparison device 14 and distributed databases 9-1 to 9-1
DB transfer device 15 for transmitting / receiving data to / from 9-n
And The DB comparison device 14 is shown in FIG.
Corresponding to the comparison processing unit 8 in FIG.
Corresponds to the control unit described above.

When a failure occurs in the master database 5, the following recovery procedure-is carried out, and (1) to (24) indicate processing points in the order of processing according to the recovery procedure. It is a thing. That is,. (1) The backup data name of the external storage medium 4 is designated from the maintenance device 3. (2) The input / output device 11 analyzes the information from the maintenance device 3 and activates the DB matching device 12. (3) The DB matching device 12 activates the DB input / output device 13. (4) The DB input / output device 13 reads the designated backup data from the external storage medium 4 and copies it to the master database 5. (5) After the backup data is copied to the master database 5, the maintenance device 3 is notified of the copy end.

[0032] (6) From the maintenance device 3, any one of the databases 9-1 to 9-n of the processing devices 2-1 to 2-n to be compared with the contents of the master database 5 is designated. (7) The input / output device 11 analyzes the information from the maintenance device 3 and activates the DB matching device 12. (8) The activated DB matching device 12 is the DB transfer device 15
Start (9) The DB transfer device 15 uses the designated database,
For example, from the database 9-1 of the processing device 2-1 to the copy database 6 as the memory area of the copy destination.
Transfer data to (10) Notify the maintenance device 3 of the end of data transfer to the copy database 6.

.. (11) From the maintenance device 3 to the master database 5
, And a content comparison with the copy database 6 is instructed. (12) The input / output device 11 analyzes the information from the maintenance device 3 and activates the DB matching device 12. (13) The DB matching device 12 activates the DB comparison device 14. (14) The DB comparison device 14 compares the data read from the master database 5 with the data read from the copy database 6.

[0034] (15) The difference information as a result of the comparison process in the DB comparison device 14 is accumulated in the difference information storage unit 7. (16) When the comparison process is completed, the maintenance device 3 is notified of the presence or absence of difference information.

.. (17) The maintenance device 3 specifies reading of the difference information. (18) The input / output device 11 analyzes the information from the maintenance device 3 and activates the DB matching device 12. (19) The DB manufacturing device 12 reads the difference information from the difference information storage unit 7. (20) The read difference information is transferred to the maintenance device 3 and displayed, and necessary data is selected according to the judgment of the maintenance person.

[0036] (21) The maintenance device 3 designates data to be reflected in the master database 5. (22) The input / output device 11 analyzes the information from the maintenance device 3 and activates the DB matching device 12. (23) The DB matching device 12 corrects, adds, and reflects the designated data in the master database 5.
Control such as deletion. (24) After the reflection process in the master database 5, the maintenance device 3 is notified of the end.

According to the above-mentioned restoration procedure, even if a failure occurs in the master database 5, it becomes possible to restore the correct data as compared with the case where the backup data of the external storage medium 4 is simply used.

FIG. 3 is an explanatory diagram of the third embodiment of the present invention. The same reference numerals as those in FIG. 2 denote the same parts, 16 is a clock management device, and 17 is a clock management table. Master
When a failure occurs in the database 5, the backup data name to be copied from the external storage medium 4 to the master database 5 and its time are designated from the maintenance device 3. The input / output device 11 analyzes information from the maintenance device 3 and
The matching device 12 is activated. The DB matching device 12 activates the DB input / output device 13. This DB input / output device 13
The designated backup data name and time are registered in the time management table 17 via the time management device 16. The state shown in the figure is at 31 o'clock on December 31, 1995 at DB.
The case where the activation of the input / output device is registered is shown. That is, the date and time, the device name, the data name, etc. can be registered as necessary.

At the designated time, the clock management device 16 activates the DB input / output device 13. DB started
The input / output device 13 reads the data of the designated backup data name from the external storage medium 4 and copies it to the master database 5. Then, at the same time as the backup data copy processing or after the copy processing is completed, one of the distributed databases 9-1 to 9-n is selected, and the data is copied to the copy database for comparison processing. Become. If the time is not designated, the backup data is immediately read from the external storage medium 4 and copied to the master database 5, as in the case of the second embodiment described above.

The processing devices 2-1 to 2-n have the database 9-1 regardless of the failure of the master database 5.
~ 9-n is executed and the process is executed. Therefore, by specifying the time zone in which the process is reduced as described above, the databases 9-1 to 9-n are selected and their data are selected. Can be transferred to the database management device 1, the influence on the processing of the processing devices 2-1 to 2-n can be reduced.

FIG. 4 is an explanatory view of a fourth embodiment of the present invention, in which the same reference numerals as those in FIG. 3 indicate the same parts, and the database management device 1 includes a master database (not shown). It is a thing. Backup data copied to the master database when a failure occurs in the master database, and data copied by selecting any one of the databases 9-1 to 9-n of the processing devices 2-1 to 2-n. Shows the case of designating the time for comparison processing, and the comparison processing time, the database name, etc. are designated from the maintenance device 3.

The input / output device 11 analyzes the information from the maintenance device 3 and activates the DB matching device 12. The activated DB matching device 12 activates the DB transfer device 15. D
The B transfer device 15 registers the DB transfer device and the time when the DB transfer device is activated in the clock management table 17 via the clock management device 16. In the case shown, December 31, 1995,
A case in which the DB transfer device is registered at 17:00 is shown. The database name can be included in the registration.

The clock management device 16 includes the clock management table 1
When the time registered in 7 is reached, the registered DB transfer device 15 is activated. Thereby, the DB transfer device 15 transfers, for example, the data in the database 9-1 to the copy database 6 according to the designated database name and copies the data. Then, as described above, the comparison process of the copied data and the backup data copied to the master database is performed. Also in this case, if the time is not registered, the data is copied from the designated database to the copy database 6 immediately by the instruction from the maintenance device 3.

FIG. 5 is an explanatory diagram of the fifth embodiment of the present invention. The same reference numerals as those in FIG.
It is a monitoring device. The database management device 1 also includes a master database (not shown). Also, the databases 9-1 to 9-1 of the processing devices 2-1 to 2-n
9-n holds the last update date information of the data.
When a failure occurs in the master database, the maintenance device 3 causes the databases 9-1 to 9-of all the processing devices 2-1 to 2-n to be released.
Instruct to read the list of n.

The input / output device 11 analyzes the information from the maintenance device 3 and activates the DB matching device 12. The DB matching device 12 activates the DB monitoring device 18. DB monitoring device 1
Reference numeral 8 reads out data for displaying a list of the databases 9-1 to 9-n of all the processing devices 2-1 to 2-n connected to the database management device 1, and the DB matching device 12
And the I / O device 11 to the maintenance device 3. The maintenance person uses the databases 9-1 to 9-1 displayed on the maintenance device 3.
One or a plurality of 9-n are selected to instruct data transfer.

The input / output device 11 analyzes the information from the maintenance device 3 and activates the DB matching device 12, and the DB matching device 1
2 starts the DB transfer device 15 by designating a processing device, and D
The B transfer device 15 transfers the data from the database corresponding to the designated one or a plurality of processing devices to the copy database 6. In that case, the copy database 6 has a capacity capable of storing the number of databases corresponding to a plurality of databases or the data of a plurality of databases.

Further, when a failure occurs in the master database, the maintenance device 3 can instruct to read the last update date information of the databases 9-1 to 9-n of all the processing devices 2-1 to 2-n. According to the instruction, the DB monitoring device 18 reads the last update date information of each of the databases 9-1 to 9-n and transfers it to the maintenance device 3.

The maintenance person displays the displayed database 9-1.
Based on the last update date information corresponding to 9-n, for example, the database corresponding to the latest date is designated. The latest date database of the latest update date information shown is 9-2.
Therefore, the maintenance person designates the database 9-2 of the processing device 2-2 and instructs the data transfer to the copy database 6. Thereby, the DB transfer device 15
Copy data from database 9-2 database 6
Transfer to In this case as well, data transfer can be performed by selecting a plurality of databases.

FIG. 6 is an explanatory view of the sixth embodiment of the present invention, in which the same reference numerals as those in the above-mentioned drawings indicate the same parts, and 20
Is a DB transfer information section. Database management device 1
Has a master database (not shown). From the maintenance device 3 to the processing device 2 in the DB transfer information section 20.
-1 to 2-n or databases 9-1 to 9-n are registered. In the illustrated case, the processing device 2-2 is registered.

Therefore, even if the maintenance device 3 does not specify the processing devices 2-1 to 2-n or the databases 9-1 to 9-n when a failure occurs in the master database, the DB
The matching device 12 uses the DB transfer device 15 based on the processing device 2-2 or the database registered in the DB transfer information unit 20.
Can be activated to transfer the data in the database 9-2 of the processing device 2-2 to the copy database 6. Also in this case, it is possible to register a plurality of processing devices or databases, and it is also possible to change them from the maintenance device 3 as needed.

FIG. 7 is an explanatory view of the seventh embodiment of the present invention, in which the same symbols as those in the above-mentioned respective drawings indicate the same parts, and 21
Is a DB transfer information section. In this DB transfer information section 21,
Information for selecting a database from the maintenance device 3 based on the date of the last update is registered. Database management device 1
Includes a master database (not shown) as in the above-described embodiments.

When a failure occurs in the master database,
When the maintenance device 3 gives an instruction to transfer data from the database to the copy database 6, the input / output device 11 analyzes the information from the maintenance device 3 and activates the DB matching device 12. The DB matching device 12 reads the registration information of the DB transfer information unit 21, and as in the above-described sixth embodiment, the processing devices 2-1 to 2-n or the databases 9-1 to 9-n.
Is not registered by designating the registered information, but the registered information is selected based on the date of the last update.
To start the database 9 of all processing devices 2-1 to 2-n.
The last update date information of -1 to 9-n is read and compared, and the latest update date database is selected and notified.

The DB matching device 12 designates a processing device and activates the DB transfer device 15 based on the notification from the DB monitoring device 18. Since the latest update date in this case is the database 9-2 of the processing device 2-2, the DB matching device 12 specifies the processing device 2-2 and the DB transfer device 1
5 is started. The DB transfer device 15 copies the data from the database 9-2 of the designated processing device 2-2.
Transfer to database 6. Then, it is compared with the backup data copied to the master database (not shown).

Further, in the DB transfer information section 21, it is possible to register information for selecting the databases 9-1 to 9-n based on the date of the last update, and also register the processing device or the database of the selected range. For example, the processing device 2-
It is possible to specify 2 to 2-5 as the selection range and register information to be selected based on the date of the last update. It is also possible to register in advance so as to select a plurality of databases in order from the date of the last update without inputting each time from the maintenance device 3.

FIG. 8 is an explanatory diagram of an eighth embodiment of the present invention, in which the same reference numerals as those in the above-mentioned drawings indicate the same portions, and when a failure occurs in the master database 5, as described above,
Backup data is copied from an external storage medium (not shown) to the master database 5, data is copied from the selected database to the copy database 6, and the time at which comparison processing is performed in the DB comparison device 14 is set. , Is registered in the clock management table 17 from the maintenance device 3. In the illustrated state, the date and time at 17:00 on December 31, 1995 and the DB comparison device are registered.

When the date and time registered in the clock management table 17 arrives, the clock management device 16 activates the DB comparison device 14. The DB comparison device 14 is a master database 5
The backup data copied to the copy database 6 is compared with the data from the database copied to the copy database 6, and the difference information is stored in the difference information storage unit 7.
Upon completion of the comparison process, the difference information is transferred to the maintenance device 3 automatically or in accordance with an instruction from the maintenance device 3, and as described above, the data in the master database 5 is updated to prevent a failure of the master database 5. Restore.

In this case, if the date and time of the comparison processing is not registered in the clock management table 17, the comparison processing is started when the data is copied from the databases 9-1 to 9-n to the copy database 6. .

FIG. 9 is an explanatory diagram of the ninth embodiment of the present invention, in which the database management device 1 is provided with the master database 5 and the databases 9-1 to 9- of the processing devices 2-1 to 2-3. 3-compatible copy database 6-1
6-3 and difference information storage units 7-1 to 7-3. Further, the same reference numerals as those in the above-mentioned drawings indicate the same parts.

When a failure occurs in the master database 5, backup data is copied to the master database 5 from an external storage medium (not shown) in accordance with an instruction from the maintenance device 3 as in the above-described embodiments. Further, the data from the databases 9-1 to 9-3 corresponding to the plurality of processing devices 2-1 to 2-3 are transferred to the copy databases 6-1 to 6-3 by starting the DB transfer device 15. make a copy. Also, a single or a plurality of copy databases 6-1 to 6-3 to be compared with the backup data copied from the maintenance device 3 to the master database 5.
Is specified.

According to the designation from the maintenance device 3, the DB comparison device 14 compares the backup data copied to the master database 5 with the data of the designated copy database, and compares the difference information with the difference information corresponding to the designated copy database. Store in the storage unit. If the copy databases 6-1 to 6-3 for comparison processing are not specified from the maintenance device 3, all copy databases 6-1 to 6-3 are specified.
Data of the master database 5 are compared with each other, and the difference information 1 to 3 is obtained from the difference information storage units 7-1 to 7-3.
To be stored.

After the comparison process is completed, the difference information is transferred from the difference information storage unit to the maintenance device 3 in accordance with the request from the maintenance device 3, and the data in the master database 5 is corrected according to the judgment of the maintenance person. -Failure recovery of database 5 is performed.

FIG. 10 is an explanatory view of the tenth embodiment of the present invention, and the same reference numerals as those in the above-mentioned drawings indicate the same parts.
In this embodiment, the backup data copied to the master database 5 and the copy database 6 include the databases 9-1 to 9-n of the processing devices 2-1 to 2-n.
The case where the maintenance device 3 is notified of the difference information or the presence / absence of the difference information when comparing the data copied from the above in the DB comparison device 14 is shown.

In this case, when the maintenance device 3 gives an instruction to request the difference information of the comparison result and to start the comparison process, the input / output device 11 analyzes the instruction information and activates the DB matching device 12 to perform the DB matching. The device 12 activates the DB comparison device 14. The DB comparison device 14 performs data comparison processing between the master database 5 and the copy database 6,
The difference information is stored in the difference information storage unit 7 and transferred to the maintenance device 3 via the DB matching device 12 and the input / output device 11. The maintenance person recognizes the displayed difference information and corrects the data in the master database 5 after the comparison process is completed. When the maintenance device 3 does not request the difference information, only the presence or absence of the difference information can be notified to the maintenance device 3.

FIG. 11 is an explanatory view of the eleventh embodiment of the present invention, in which the same reference numerals as those in the above-mentioned drawings indicate the same parts,
Reference numeral 7-4 is a difference information storage unit, and 22 is a DB logic device.
The DB logic device 22 includes difference information storage units 7-1 to 7-.
The case where the difference information from 3 is subjected to logical processing such as logical product or logical sum and stored in the difference information storage unit 7-4 is shown.

That is, when a failure occurs in the master database 5, backup data is copied from an external storage medium (not shown) according to an instruction from the maintenance device 3, and the databases of the plurality of processing devices 2-1 to 2-3 are also copied. 9-1
The data 9-3 is transferred to the copy databases 6-1 to 6-3 under the control of the DB transfer device 15 and copied,
The DB comparison device 14 compares the data in the master database 5 with the data in each of the copy databases 6-1 to 6-3, and the difference information is stored in the difference information storage unit 7-.
Stored in 1 to 7-3.

Further, the maintenance device 3 instructs the DB logical device 22 to perform logical processing such as logical product or logical sum of difference information. Thereby, the difference information in the difference information storage units 7-1 to 7-3 is logically processed and stored in the difference information storage unit 7-4. Based on the information stored in the difference information storage unit 7-4, the data in the master database 5 is corrected and the failure is restored.

FIG. 12 is an explanatory diagram of the logical processing of the difference information. FIG. 12A shows the case of the logical product. For example, the comparison processing of the difference information 1 and 2 in the difference information storage units 7-1 and 7-2. If the partial data * 1 and * 2 are * 1 ≠ * 2,
The data portions of * 1 and * 2, which are not the same, are stored in the difference information storage unit 7-4.

Further, (B) shows a case of logical sum, for example,
The difference information 1 and 3 in the difference information storage units 7-1 and 7-3 are compared, and the partial data * 1 and * 3 are * 1 ≠ *.
In the case of 3, partial data * 1 is stored in the differential information storage unit 7-4 when the side of the differential information storage unit 7-1 is prioritized by the setting from the maintenance device 3, and partial data * 1, * Data excluding 3 is stored next. When the difference information storage unit 7-3 side is prioritized, a part of the data * 3 is stored in the difference information storage unit 7-3.
-4, and the data excluding the partial data * 1 and * 3 is stored next.

FIG. 13 is an explanatory view of the twelfth embodiment of the present invention, in which the same reference numerals as those in the above-mentioned drawings indicate the same parts,
Due to the failure of the master database 5, backup data is copied from an external storage medium (not shown), and the databases 9-1 to 9-n of the processing devices 2-1 to 2-n are also copied.
The difference information obtained by copying the data from any one or a plurality of databases 9-n to a copy database (not shown) and comparing the data of the master database 5 and the data of the copy database The state stored in the storage unit 7 is shown.

This difference information includes the databases 9-1 to 9-1.
The last update date information of 9-n is added, and the update date can be specified from the maintenance device 3 or the range of the update date can be specified. As a result, the DB matching device 12 reads the specified update date or the difference information within the range of the update date from the difference information storage unit 7 and transfers it to the maintenance device 3 via the input / output device 11. Therefore, the maintenance person can recognize the content including the presence or absence of the difference information of the desired date.

FIG. 14 is an explanatory view of the thirteenth embodiment of the present invention, and the same reference numerals as those in FIG. 13 indicate the same parts. In this embodiment, in the state of FIG. 13, the difference information displayed on the maintenance device 3 is recognized, and the data of the master database 5 is corrected based on the difference information. Thereby, the master database 5 in which the failure has occurred can be reliably restored.

FIG. 15 is an explanatory view of the fourteenth embodiment of the present invention, in which the same reference numerals as those in the above-mentioned drawings indicate the same parts.
In this embodiment, the processing time for reflecting the difference information to the master database 5 in the embodiment shown in FIG. 14 is registered from the maintenance device 3 in the clock management table 17. The clock management device 16 displays in the clock management table 17, for example, 17:00 on December 31, 1995,
If the DB matching device and the reflection instruction information of the difference information are registered, the DB matching device 12 is activated at that date and time to start the reflection process.

The activated DB matching device 12 reads the difference information from the difference information storage unit 7 and corrects the backup data copied to the master database 5. The difference information storage unit 7 in this case is shown in FIG.
The difference information storage unit 7-4 shown in FIG. 6 can correct the data in the master database 5 based on the difference information that has been subjected to the logical processing.

FIG. 16 is an explanatory view of the fifteenth embodiment of the present invention, in which the same reference numerals as those in the above-mentioned drawings indicate the same parts.
This embodiment shows a case where the difference information is reflected in the database of the processing device. In the state where the difference information is stored in the difference information storage unit 7, the difference information storage unit 7 is instructed according to the instruction from the maintenance device 3. From the difference information to the maintenance device 3, the maintenance person recognizes and selects the displayed difference information,
If the difference information with respect to the backup data is an error on the database side or an update omission, the processing device or database is specified. For example, when the maintenance person recognizes that the data in the database 9-2 needs to be corrected, the processor 2-2 or its database 9-2 is designated and the correction processing is performed based on this difference information.

FIG. 17 is an explanatory diagram of the sixteenth embodiment of the present invention, in which the same reference numerals as those in the above-mentioned drawings indicate the same parts.
In this embodiment, when a failure occurs in the master database 5 of the database management device 1, a plurality of processing devices 2-
One of the databases 1-2-1 to 9-n is selected, and the data is transferred to the master database 5 via the DB transfer device 15 and copied.

In this case, a database for transferring data from the maintenance device 3 to the master database 5 is
Specify in advance or each time, or in FIGS. 5, 6 and 7.
As shown in, the database can be selected by recognizing the maintenance person or automatically recognizing the latest update date based on the last update date information.

A printer or the like can be connected to the database management device 1 to print out the trouble record and the recovery record of the master database 5. It is also possible to input a command from the maintenance device 3 for each recovery procedure of the master database 5. In addition, the present invention can be applied to the intelligent network system shown as a conventional example, but the present invention can be applied to other distributed databases that hold data of the same content in a distributed manner when a failure occurs in the master database. The recovery method of can be applied.

[0078]

As described above, according to the present invention, when the failure of the master database 5 in the distributed database occurs, the latest data in the distributed database can be used for restoration. Therefore, there is an advantage that the reliability of the master database 5 can be maintained. Further, since the actual processing is executed by accessing the databases 9-1 to 9-n of the processing devices 2-1 to 2-n, the failure recovery of the master database 5 is performed by the processing devices 2-1 to 2-1. It can be controlled to execute at a date and time corresponding to the 2-n state, and the influence on the entire system can be reduced, and the failure recovery of the master database 5 can be performed.

Further, the database management device 1 is a master
The database 5, the copy database 6, the difference information storage unit 7, the comparison processing unit 8 and the like are provided, and the database 9-1 to 9-n is normally used after the master database 5 is changed according to the data change request. The update control is performed from the master database 5 side according to the free access state of the master database 5. However, when the master database 5 fails, the backup data and the database 9
-1 to 9-n is used to restore the latest data, and there is an advantage that the failure recovery can be performed by a relatively simple process and the reliability of the master database can be improved. is there.

[Brief description of drawings]

FIG. 1 is an explanatory diagram of a first embodiment of the present invention.

FIG. 2 is an explanatory diagram of a second embodiment of the present invention.

FIG. 3 is an explanatory diagram of a third embodiment of the present invention.

FIG. 4 is an explanatory diagram of a fourth embodiment of the present invention.

FIG. 5 is an explanatory diagram of a fifth embodiment of the present invention.

FIG. 6 is an explanatory diagram of a sixth embodiment of the present invention.

FIG. 7 is an explanatory diagram of a seventh embodiment of the present invention.

FIG. 8 is an explanatory diagram of an eighth embodiment of the present invention.

FIG. 9 is an explanatory diagram of a ninth embodiment of the present invention.

FIG. 10 is an explanatory diagram of a tenth embodiment of the present invention.

FIG. 11 is an explanatory diagram of an eleventh embodiment of the present invention.

FIG. 12 is an explanatory diagram of logical processing of difference information.

FIG. 13 is an explanatory diagram of the twelfth embodiment of the present invention.

FIG. 14 is an explanatory diagram of a thirteenth embodiment of the present invention.

FIG. 15 is an explanatory diagram of the fourteenth embodiment of the present invention.

FIG. 16 is an explanatory diagram of the fifteenth embodiment of the present invention.

FIG. 17 is an explanatory diagram of the sixteenth embodiment of the present invention.

FIG. 18: Intelligent network of conventional example
It is an explanatory view of a system.

FIG. 19 is an explanatory diagram of a system including a distributed database.

[Explanation of symbols]

 1 Database Management Device 2-1 to 2-n Processing Device 3 Maintenance Device 4 External Storage Medium 5 Master Database 6 Copy Database 7 Difference Information Storage Unit 8 Comparison Processing Unit 9-1 to 9-n Database

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Yayoi Ito 3-9-18 Shin-Yokohama, Kohoku-ku, Yokohama-shi, Kanagawa Fujitsu Communication Systems Limited (72) Inventor Hiroshi Morita 3 Shin-Yokohama, Kohoku-ku, Yokohama-shi, Kanagawa 9-9-18 Fujitsu Communication Systems Limited (72) Inventor Masaki Watanabe 3-9-18 Shin-Yokohama, Kohoku Ward, Yokohama City, Kanagawa Prefecture Fujitsu Communication Systems Limited

Claims (14)

[Claims] 1. A failure recovery method for a distributed database having a plurality of databases that are update-controlled so as to have the same contents as the master database, wherein when a failure occurs in the master database, an instruction from a maintenance device is followed. A failure recovery method for a distributed database, comprising: transferring data of one of the selected databases to the master database to recover the failure of the master database. 2. A database management apparatus having a master database, a copy database, and a difference information storage unit, an external storage medium connected to the database management apparatus for backing up the master database, and the database management apparatus. A plurality of processing units each having a database connected to the master database and controlled to be updated so as to have the same content as the master database;
A maintenance device connected to the database management device, wherein when a failure occurs in the master database, backup data is copied from the external storage medium to the master database according to an instruction from the maintenance device; Data of a single or a plurality of databases selected from among the above databases is copied to the copy database, the data of the copy database is compared with the backup data of the master database, and the difference information is converted into the difference. A failure recovery method for a distributed database, comprising the step of modifying backup data stored in an information storage unit and copied to the master database based on the difference information. 3. The distribution according to claim 2, further comprising the step of copying the backup data from the external storage medium to the faulty master database based on the time information registered in the clock management table. Database failure recovery method. 4. The method according to claim 4, further comprising the step of transferring data of a database selected from the plurality of databases to the copy database based on time information registered in the clock management table and copying the data. Item 2
Disaster recovery method for the described distributed database. 5. A process for comparing data copied from the database to the copy database with backup data copied to the master database based on time information registered in a clock management table is included. The method of recovering a failure of a distributed database according to claim 2. 6. The last update date information according to the contents of the master database is added to the database of the processing device, and the last update date is requested by the maintenance device when a failure occurs in the master database. Transferring and displaying a processing unit name or a database name including information, designating a single or a plurality of databases based on the displayed last update date, and transferring the data to the copy database. Claim 2
Disaster recovery method for the described distributed database. 7. The distributed database failure recovery method according to claim 2, further comprising a step of collecting and comparing the last update date information and automatically selecting a database having the latest update date. 8. A database for transferring data to the copy database is designated in advance from the maintenance device, and when a failure occurs in the master database, the copy is performed from the predesignated database according to a transfer instruction from the maintenance device. The method according to claim 2, further comprising the step of transferring data to the database. 9. When a failure occurs in the master database, a plurality of databases are selected, data corresponding to the plurality of databases is copied to the copy database, and data corresponding to each database and the master database are copied. The backup data is compared with backup data, the difference information is stored in the difference information storage unit in correspondence with a plurality of databases, and the backup is copied to the master database based on information obtained by logical processing of the difference information corresponding to the plurality of databases. 9. The distributed database failure recovery method according to claim 2, further comprising a step of correcting data. 10. The last update date information of the database is added to the difference information stored in the difference information storage unit,
10. The method according to claim 2, further comprising the step of designating a date or a date range and modifying the backup data copied to the master database based on the difference information within the date or the date range. A failure recovery method for a distributed database according to any one of items. 11. A database management device comprising a master database for controlling update of a plurality of databases so as to have the same contents, wherein input / output between the master database and an external storage medium for backup. A device, a transfer device for transferring data from the specified database when a failure occurs in the master database, a copy database for copying the data transferred via the transfer device, and a copy for the copy database A comparison processing unit that compares the created data with the backup data copied to the master database; a difference information storage unit that stores difference information as a result of comparison by the comparison processing unit; and instruction information from the maintenance device. And a control unit for controlling each unit based on Management device. 12. A clock management table for registering a device to be activated and time, and a clock management device for controlling activation of the registration device according to registration and registration time in the clock management table. Item 11. The database management device according to item 11. 13. Maintaining designated information such as a database name designated in advance by a maintenance device from a plurality of the databases, controlling the transfer device according to the designated information, and copying the data from the designated database 13. The database management device according to claim 11, further comprising a transfer information unit for transferring to the database. 14. With respect to the last update date information added to a plurality of said databases, designated information which designates the last updated date or updated date range from a maintenance device is held in advance, and said transfer device is controlled according to said designated information. 13. The database management apparatus according to claim 11, further comprising: a transfer information unit for selectively connecting the databases and transferring the data of the selectively connected databases to the copy database.