JP2011053780A - Restoration system, restoration method and backup control system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a system which can be surely restored while reducing the stop time of the system as much as possible even if failure recovery by its own server is difficult. <P>SOLUTION: The restoration system includes a local storage means which stores files to be used by the server while classifying them to each type; an external storage means which stores the files to be stored in the local storage means by types; an information management means which manages at least first information determining types of backup files to be acquired according to the modes of system failures of the server; and an external backup control means which receives, upon occurrence of failure which can be hardly restored by the own server occurs in the server, a restore request according to the first information, acquires a file of the type designated in the restore request from the external storage means, and transfers it to the server. The server includes a restoring means which performs restoration by use of the file acquired from the backup control means. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a recovery system, a recovery method, and a backup control system, and can be applied to, for example, a data or program system for restoration when a failure occurs in a hard disk of a server or the server itself.

  In a server that is always operating, it is common to make backups of servers and regularly back up important data and programs.

  Furthermore, in systems where high reliability is required, in order to enable restoration even if normal backup data is overwritten, multiple generations of backup data are retained, and appropriate generations of backup data are stored when a failure occurs. Select and restore.

  For example, in a server that requires extremely high reliability, such as a call server that provides an IP telephone service, a mechanism for autonomously restarting a process when a failure occurs is provided in order to minimize system downtime. Have. Further, when it is determined that the active file cannot be restarted, the backup data is used for recovery. At this time, restoration is required to restore the file as new as possible and to minimize the system stop time.

  Considering reliable and quick recovery when a server failure occurs, it is common to store backup data on the local server or a redundant standby server.

  Japanese Patent Application Laid-Open No. 2004-151561 describes a technique for backing up OS information of an OS master volume of an active server to an OS backup volume of a standby server in a redundant configuration including an active server and a standby server.

Japanese Patent Publication No. 2008-198152

  As described above, conventionally, backup data is generally held in its own server, a redundant standby server, or the like.

  However, when the hard disk itself of the own server that holds the backup data fails or when the data or program is defective, the backup data held in the standby server may also be defective.

  For this reason, there is a need for a system that can be reliably recovered by shortening the system downtime as much as possible even if the hard disk of the server itself fails or if the backup data held by the standby server fails. Yes.

  The present invention includes a backup server that saves backup data and stores the backup data in the backup server. The backup server classifies and stores the backup data according to the type of backup. A recovery system, a recovery method, and a backup control system for acquiring backup data quickly and surely based on an acquisition destination and performing a recovery process are provided.

  In order to solve such a problem, the recovery system of the first aspect of the present invention includes (1) a local storage unit that classifies and stores files used by the server, and (2) a file that is stored in the local storage unit. (3) Information management means for managing first information in which at least the type of backup file to be acquired is determined according to the system failure mode of the server, and (4) the server When a failure that makes it difficult to restore by the local server occurs, a restore request according to the first information is received, the type of file specified in the restore request is acquired from the external storage means, and transferred to the server External backup control means for performing restore using the file acquired from the backup control means by the server. Characterized in that it.

  The recovery method of the second aspect of the present invention includes (1) a local storage step in which the local storage means classifies and stores the files used by the server for each type, and (2) the external storage means stores in the local storage means. An external storage step for storing files to be classified for each type, and (3) information management for managing first information in which the information management means negotiates at least the type of backup file to be acquired according to the system failure mode of the server The process and (4) the external backup control means receives a restore request according to the first information when a failure that makes it difficult for the server to restore by the local server occurs, and the type of the type specified in the restore request An external backup control step of acquiring a file from the external storage means and transferring it to the server, and the server from the backup control means Using the resulting file and having a restoring step of restoring.

  The backup control system of the third aspect of the present invention includes (1) a local storage means for classifying and storing files used by the server for each type, and (2) an external for storing the files stored in the local storage means for each type. Storage means, (3) at least information management means for managing the first information in which the type of backup file to be acquired is determined according to the mode of the system failure of the server, and (4) it is difficult for the server to restore by the own server An external backup control unit that receives a restore request according to the first information, acquires a file of the type specified in the restore request from the external storage unit, and transfers the file to the server when a failure occurs. It is characterized by providing.

  According to the present invention, even when the hard disk of the server itself fails or when a failure occurs in the backup data held by the standby server, the system can be reliably restored by shortening the system down time as much as possible.

It is a block diagram which shows the structure of the backup control system of 1st Embodiment. It is explanatory drawing explaining the structure of backup data acquisition information. It is a flowchart which shows a backup process. It is explanatory drawing explaining a backup process. It is a flowchart which shows the process by the backup control of an application server. It is explanatory drawing explaining the backup control process of an application server. It is a flowchart which shows the process by the backup control of a backup server. It is explanatory drawing explaining the process of backup control of a backup server. It is a block diagram showing a case where a backup server backs up multiple generations of previous day files and startup guarantee files. It is a block diagram which shows the case where a backup server connects with an external disk or a dedicated server.

(A) Main Embodiment Hereinafter, a first embodiment of a recovery system, a recovery method, and a backup control system of the present invention will be described with reference to the drawings. This embodiment will be described by exemplifying a case where the present invention is applied to a server that requires extremely high reliability, such as a call server that provides an IP telephone service.

(A-1) Configuration of Embodiment FIG. 1 is a configuration diagram illustrating a configuration of a backup control system according to the embodiment. In FIG. 1, the backup control system 1 according to the first embodiment includes an application server 10 and a backup server 20 provided outside the application server 10.

  Note that the application server 10 and the backup server 20 are connected via a connection line, and each includes a CPU, a ROM, a RAM, and the like, and the processing of the OS and backup control unit described later is executed by the CPU in terms of hardware. Is.

  The application server 10 corresponds to a server that requires high reliability, such as a call server. As shown in FIG. 1, the application server 10 includes at least an application 11, a backup control unit 12, an operating system (OS) 13, a hard disk 14, and the like.

  The application server 10 includes a general OS 13 such as Linux (registered trademark), for example, and the processes of the application 11 and the backup control unit 12 operate on the OS 13.

  The application 11 executes a process related to a service process (for example, a call processing service) provided by the application server 10, and is executed by referring to an execution file on the hard disk 14 during operation, Or update it.

  The hard disk 14 is a storage area for storing the OS 13 and application 11 programs, application execution files, backup data acquisition information 17 and the like. The hard disk 14 stores backup data classified by type.

  The types of backup data include, for example, the previous day file 15, the startup guarantee file 16, and the full backup file, and these files are classified by type.

  In the previous day file 15, for example, a file used for operation is stored every day at a predetermined time. As a result, the file one generation before the current operation can be stored. The startup guarantee file 16 stores a file that guarantees the start-up time of the new file after the application 11 is upgraded. The backup data includes an execution file, configuration file, various data, and the like of the application 11.

  The backup control unit 12 is activated by a command from the OS 12 or the application 11. In addition, when the backup server 12 backs up each backup data after the application server 10 is started, the backup controller 12 classifies the backup data for each type of backup data. Further, the backup control unit 12 sets the backup data storage destination to the hard disk 14 of the server 10 and the hard disk 23 of the backup server 20.

  Further, the backup control unit 12 includes a regular processing thread and a backup control thread.

  The periodic processing thread backs up the executable file in operation on the day 15 to the previous day file 15 on the hard disk 14 at a predetermined time every day after the application server 10 is started.

  As will be described later, the backup control thread performs a process for backing up each backup file to the hard disk 14 of the server 10 and a backup process for transferring each backup file to the external backup server 20 for backup.

  Further, when a failure occurs, the backup control thread receives a restore command, acquires a file of a type specified by this command from a specified acquisition destination, and performs a restore process.

  The backup control thread backs up the startup guarantee file of the application to be upgraded to the startup guarantee file 16 of the hard disk 14 at the time of application upgrade.

  Further, when a command for requesting transfer of the previous day file 15 or the startup guarantee file 16 is given, the backup control thread transfers the previous day file 15 or the startup guarantee file 16 stored in the hard disk 14 to the backup server 20. Backup to the hard disk 23 of the backup server 20.

  Further, when a command for requesting transfer of the full backup file is given, the backup control thread acquires the full backup file of its own server from the hard disk 14, transfers the full backup file to the backup server 20 and backs up to the hard disk 23. Is.

  When a backup control thread performs a restore, when a failure that triggers backup file acquisition occurs, it receives a command specifying the backup data type and acquisition destination according to the failure mode, and acquires the backup data according to the contents of this command . As a destination for acquiring the backup data, the hard disk 14 of the local server 10 or the hard disk 23 of the backup server 20 is used as described later. The types of backup data to be acquired include the previous day file, startup guarantee file, full backup file, and the like.

  A method of referring to the backup data acquisition information 17 and specifying the backup data type and acquisition destination in the command can be applied.

  FIG. 2 is a configuration diagram showing the configuration of the backup data acquisition information 17. The backup data acquisition information 17 is set by associating a backup data type, a backup acquisition destination, a failure type, and a command request destination.

  For example, when the hard disk 14 of the application server 10 fails, as shown in FIG. 2, the type of backup data to be acquired is “full backup file”, and the acquisition destination of this backup data is “hard disk 23 of the backup server 20”. This indicates that the request command to be transmitted is “backup control unit 22 of backup server 20” as a request destination.

  The backup server 20 is a server that backs up an executable file executed by the application 11 of the application server 10. The backup server 20 includes at least a backup control unit 21, an OS 22, and a hard disk 23.

  The backup server 201 stores the backup data for each backup target server by classifying the backup data for each type. Further, the backup server 20 may store backup data of a plurality of backup target servers as well as a single backup target server.

  In the backup server 20, the OS 22 is mounted with a general OS 22 such as Linux (registered trademark), for example, and the backup control unit 21 operates on the OS 22, and the data of the hard disk 23 is operated during operation. Browse or update.

  Note that the backup to the hard disk 23 of the backup server 20 is realized by the processing of the backup control unit 12 of the application server 10 as described above.

  The backup control unit 21 receives a command after the failure recovery of the application server 10 or the hard disk 14 in the case of the failure of the hard disk 14 of the application server 10 or the failure of the application server 10 itself, and the backup data stored in the hard disk 23 Is given to the application server 10. As a result, the OS and application programs can be reinstalled in the application server 10 and restored using necessary file data.

  Here, the request command describes the type of backup data to be acquired, and according to this, the backup control unit 21 selects one of the previous day file 24, the startup guarantee file 25, and the full backup file 26 to Transfer to server 10.

(A-2) Operation of Embodiment Next, the processing operation of the backup control system according to this embodiment will be described with reference to the drawings.

(A-2-1) Backup Process First, the operation of the backup process according to the embodiment will be described. FIG. 3 is a flowchart illustrating the operation of the backup processing according to the embodiment. FIG. 4 is an explanatory diagram illustrating the operation of the backup processing according to the embodiment.

  When the application server 10 is activated, the backup control unit 12 activates a periodic processing thread at a predetermined time. Further, when receiving a command from the OS 13 or the application 11, the backup control unit 12 activates a backup control thread that performs processing according to the command request (step S101).

  The periodic processing thread starts at a predetermined time every day, acquires an execution file necessary for the application 11 to execute as backup data, and stores it in the previous day file 15 of the hard disk 14 (step S102).

  Further, the backup control thread is activated upon receiving a command request from the OS 13 or the application 11 (step S103).

  At this time, when the program of the application 11 is upgraded, the application 11 gives a command for requesting the backup acquisition of the startup guarantee file to the backup control unit 12. Upon receiving this, the backup control thread obtains a startup guarantee file for ensuring startup of the upgrade application as a backup file, and stores it in the startup guarantee file 16 of the hard disk 14 (step S104).

  When the previous day file backup server transfer request is received by inputting the command, the backup control thread transfers the previous day file 15 stored in the hard disk 14 to the backup server 20 (step S105). At this time, the backup server 20 stores the previous day file received from the application server 10 in the previous day file 24 of the hard disk 23.

  Here, the previous day file backup server transfer request command is input, for example, immediately after the previous day file 15 is stored in the hard disk 14, after a predetermined time has elapsed after the acquisition of the previous day command 15, or at a predetermined time every day. Can do.

  Further, when a startup guarantee file backup server transfer request is received by inputting a command, the backup control thread transfers the startup guarantee file 16 held on the hard disk 14 to the backup server 20 (step S106). At this time, the backup server 20 stores the startup guarantee file received from the application server 10 in the startup guarantee file 25 of the hard disk 23.

  Here, the startup security file backup server transfer request command may be input immediately after the startup security file 16 is stored in the hard disk 14 or after a predetermined time has elapsed after the acquisition of the startup security file 16 is completed. it can.

  When a full backup file acquisition / transfer request is received by inputting a command, the backup control thread acquires a full backup of the own server held in the hard disk 14 and transfers the full backup file to the backup server 20 ( Step S107). Then, the backup server 20 stores this full backup file in the full backup file 26 of the hard disk 23.

  The command input of the full backup file acquisition / transfer request can be performed immediately after storing either the previous day file 15 or the startup guarantee file 16 or after a predetermined time has elapsed after the storage.

  The full backup file may be combined with the previous full backup file as an incremental file when the file 15 on the previous day is backed up. The incremental file of the second day is combined with the first full backup file (first day). Further, a full-back file is created by synthesizing the third-day incremental file with the synthesized full-back file.

  As described above, the server 10 and the backup server 20 can hold the previous day file, the startup guarantee file, and the full backup file as backup data.

(A-2-2) Restore Process by Backup Control Unit 12 Next, the operation of the restore process by the backup control unit 12 of the application server 10 will be described with reference to the drawings.

  FIG. 5 is a flowchart showing the restore processing according to the embodiment. FIG. 6 is an explanatory diagram for explaining the restore processing according to the embodiment.

  The restore process is executed by the backup control thread of the backup control unit 12 by receiving a restore command from the OS 12 or the application 11.

  When a failure occurs in the application server 10, the OS 13 starts a shutdown process. In this shutdown process, the OS 13 gives the backup control unit 12 a restore command for designating the backup data acquisition destination and acquisition type, and the backup control thread is activated. The backup control thread may be activated when the application 11 inputs a restore command.

  When the restore command is input, the backup data type and acquisition destination corresponding to the failure that occurred are specified.

  Here, the backup data type and acquisition destination designation method is, for example, backup data acquisition information 17 in which the type of backup data to be acquired and the acquisition destination are set according to the failure mode that has occurred as shown in FIG. The method of specifying with reference to can be applied.

  In the backup data acquisition information 17 of FIG. 2, for convenience of explanation, the backup data acquisition destination is described as “application server”, “backup server”, but the acquisition destination server address information and link destination Etc. may be described.

  The backup data type may also be identification information for identifying the type, storage data (link destination) information of the backup data in the hard disks 14 and 23, and the like. In this way, by including the acquisition destination and type of backup data to be acquired in the command, a quick restore process can be realized.

  For example, as shown in FIG. 2, if a failure due to a bug or the like occurs in the application 11, a command is sent to the backup control unit 12 of the application server 10.

  When the backup control thread waiting for command reception receives the restore command (step S201), the backup data acquisition destination and type specified by the command are confirmed (step S202).

  When the backup data acquisition destination is the local server (step S203), the backup control thread loads the backup data (the previous day file 15, the startup guarantee file 16) specified by the command from the hard disk 14 of the local server 10. Then, after the OS 13 is rebooted, the application 11 starts autonomously and performs restoration by reading the designated backup data (step S204).

  When the backup data acquisition destination is the backup server 20 (step S203), the backup control thread reads the backup data (the previous day file 24, the startup guarantee file 25, the full file) specified by the command from the hard disk 23 of the backup server 20. Obtain and load the backup file 26). Then, the started application 11 performs restoration by reading the backup data (step S205).

(A-2-3) Restore Process by Backup Control Unit 21 Next, the operation of the restore process by the backup control unit 21 of the backup server 20 will be described with reference to the drawings.

  FIG. 7 is a flowchart showing a restore process by the backup control unit 21 of the backup server 20. FIG. 8 is an explanatory diagram for explaining a restore process by the backup control unit 21 of the backup server 20.

  As shown in FIG. 2, when the hard disk 14 of the application server 10 fails or when the main body of the application server 10 fails, a command is sent to the backup control unit 21 of the backup server 20.

  For example, a failure occurs in the application server 10, and 0S13 starts the shutdown process. Thereafter, the failure of the hard disk 14 of the application server 10 or the failure of the application server 10 main body is recovered.

  Then, after recovery from the failure, a command is input to the backup control unit 21 of the backup server 20. This command input may be automatically input as part of the restoration process after the failure recovery of the application server 10, or may be input in response to an instruction from an administrator terminal (not shown). May be.

  In the backup control unit 21, a backup control thread is activated by inputting a command (step S301). The backup control thread confirms the acquisition destination and type of the backup data specified by the command (step S302).

  Here, the full backup file 26 is designated as the backup data to be acquired. The backup control thread reads the specified full backup file 26 from the hard disk 23 based on the link destination specified by the command, and transfers it to the hard disk 14 of the application server 10 (step S303).

  Since the full backup file 26 also includes an OS and application programs, the application server 10 performs a restart process using the full backup file 26 received from the backup server 10 and restores the server environment. As a result, the work of installing the OS and application programs can be reduced, so that a quick response can be taken.

  Next, in the application server 10, after restoring from the full backup file of the hard disk 14 of the own server 10, processing when the application of the application server is an old file will be described with reference to FIG. 7B and FIG. To do.

  In this case, it is assumed that the time when the application server 10 backs up the full backup file is different from the time when the backup server 20 backs up the previous day file 24 and the startup guarantee file 25 (when the backup acquisition timing is different). .

  In the application server 10, after restoring from the full backup file of the hard disk 14, if the application of the application server 10 is an old file, a command is input to the backup control unit 12 of the application server 10 (step S 401).

  This command specifies that the backup data acquisition destination is the bird disk 23 of the backup server 10 and that the backup data type is the previous day file 24 or the startup guarantee file 25 (see FIG. 2).

  In the application server 10, the backup control thread confirms the backup data acquisition destination and type specified by the command (step S402), and transfers the backup data acquisition destination and type specified by the command to the backup server 20. Request.

  In the backup server 10, the backup control unit 21 activates the backup control thread, and transfers the previous day file 24 or the startup guarantee file 25, which is the backup data type specified by the input command, to the application server 10. The application server 10 uses the previous day file 24 or the startup guarantee file 25 received from the backup server 10 to perform the restart process using the latest file, and restores the server environment (step S403).

(A-3) Effect of Embodiment As described above, according to the embodiment, the restoration process by the backup server is possible even when the hard disk of the server or the server body fails. In addition, even when a plurality of servers exist due to redundancy or the like, restoration processing by the backup server can be performed.

  In the embodiment, backup data is saved in an external backup server, the backup data to be backed up is classified and backed up by type, and the backup data acquisition trigger and the backup data acquisition destination are clarified. This makes it possible to clarify which type of backup data is read from which acquisition source when backup data is acquired, thereby realizing quick and early restoration.

  Furthermore, in the embodiment, when the local server or the redundant standby server fails, the backup data saved in the backup server can be dealt with surely and quickly at the time of restoration.

(B) Other Embodiments In the above-described embodiment, for the sake of convenience of explanation, the case where the previous day file and the startup guarantee file are held in only one generation in the hard disk of the backup server has been described. As described above, the backup server 20 may manage a plurality of generations of previous day files and startup guarantee files.

  In the above-described embodiment, the case where the backup server itself does not include an application is exemplified. However, even when the application is mounted and the data program requires backup acquisition, the backup server is similar to the backup control unit of the application server. Can be backed up and restored by the backup controller.

  Furthermore, for example, as shown in FIG. 10, by providing a backup medium (for example, a hard disk, an optical disk, a backup tape, etc.) or a server (backup server for the backup server) of the backup server itself, backup / restore of the backup server itself can be performed. Is possible.

  When a redundant configuration is adopted, a backup server may be provided for one active server, and another backup server may be provided for one standby server. Further, the active server and the standby server may share the backup server.

1 ... Backup control system,
DESCRIPTION OF SYMBOLS 10 ... Application server, 20 ... Backup server 11 ... Application, 12 ... Backup control part, 13 ... OS, 14 ... Hard disk, 15 ... Previous day file, 16 ... Startup security file, 17 ... Backup data acquisition information,
21 ... Backup control unit, 22 ... OS, 23 ... Hard disk, 24 ... Previous day file, 25 ... Startup guarantee file, 26 ... Full backup file.

Claims (4)

Local storage means for storing files used by the server classified by type, and
External storage means for storing the file to be stored in the local storage means for each type;
At least information management means for managing the first information in which the type of backup file to be acquired is determined according to the system failure mode of the server;
When a failure occurs that makes it difficult for the server to restore the server, it receives a restore request according to the first information, and obtains the type of file specified in the restore request from the external storage means. And an external backup control means for transferring to the server,
A recovery system, wherein the server has a restore means for restoring using a file acquired from the backup control means.   If the application is a past file as a result of restoring the full backup file in the server, the external backup control means receives a restore request according to the first information, and receives the type specified in the restore request. The recovery system according to claim 1, wherein the file is acquired from the external storage unit and transferred to the server. A local storage step in which the local storage means classifies and stores the files used by the server for each type;
An external storage means for storing a file to be stored in the local storage means for each type;
The information management means manages at least the first information in which the type of the backup file to be acquired according to the system failure mode of the server is determined, and the external backup control means performs restoration by the own server in the server. An external device that receives a restore request according to the first information, acquires a file of the type specified in the restore request from the external storage means, and transfers it to the server Backup control process,
A restoration method, wherein the server includes a restoration step of restoring using a file acquired from the backup control means. Local storage means for storing files used by the server classified by type, and
External storage means for storing the file to be stored in the local storage means for each type;
At least information management means for managing the first information in which the type of backup file to be acquired is determined according to the system failure mode of the server;
When a failure occurs that makes it difficult for the server to restore the server, it receives a restore request according to the first information, and obtains the type of file specified in the restore request from the external storage means. An external backup control means for transferring to the server.

Images