JPH0340121A - Disk duplication method - Google Patents

Abstract

PURPOSE:To surely assure the file updating contents by storing an original file and its back-up file into a duplexed disk and using a spare disk when a working disk has a fault. CONSTITUTION:When an original file 12 of a working disk 10 is updated, a working write flag 20 is set and then reset when the file 12 is updated. Simultaneously, the copy 16 of a new file 14 is produced on a main storage 18. When the back-up file 24 of a spare disk 26 is updated, a spare write flag 28 is set. Then the file 24 is updated with the copy 16 of the storage 18 when the disk 10 has a fault and the flag 28 is set. Furthermore the file 12 is updated with the file 24 when the flag 20 is not set in a restarting state caused by a system down. Then the file 24 is updated with the file 12 when the flag 28 is set. Consequently, the file contents are surely assured.

Description

[Detailed description of the invention] [Table of contents] Overview Industrial field of application Conventional technology Problems to be solved by the invention Kth means for solving the problems Effects of the invention [Summary] Updating files on the active disk Regarding the disk duplication method that is reflected on the spare disk, we aim to provide a method that can reliably guarantee the contents of file updates even if a failure occurs on the active disk or the system goes down. Sets the active write flag when the original file is updated, resets this flag when the update is completed, creates a copy of the new file in main memory, and updates the backup file on the spare disk. Set the spare side write flag,
When a failure occurs on the active disk, the backup file is copied to the main memory and updated when the backup write flag is set, and when the system is restarted due to system failure, the active write flag is set. When the original file was saved as a backup file,
It is characterized in that when the backup side write flag is set, each backup file is updated with the original file.

[Industrial Field of Application] The present invention relates to a disk duplication system in which file updates on a working disk are reflected on a backup disk.

The SVP (service processor) is used for operation, maintenance, control, etc. of the main system, and is composed of a small computer system so that appropriate measures can be taken in case of an abnormality in the main system.

Therefore, the occurrence of a failure that damages the contents of files used for SVP processing has a serious impact on the main system, and it is therefore necessary to guarantee the contents of these files.

Therefore, the SVP disks are duplicated, and the original file and its backup file are each stored on these disks, and if a failure occurs in the active disk, the spare disk is used in its place.

[Prior art] Fig. 11 shows the configuration of SVP, and its bus 3
0 has MPU 32. Display adapter 34 (
DPA) and file adapter 3B (FA) are connected.

A main storage device 38 is connected to the MPU 32, and a console 40 (C
9L) and line printer 4.2 (LP) are connected.

In addition, the file adapter 36 is equipped with an m-disk on the active side.
44 (MDO), spare disk unit 4B (MDI)
are connected together with a floppy disk device 48 (FD), and these magnetic disk devices 44, 46j store original files and their backup files, respectively.

In FIG. 12, the conventional method is explained using a flowchart. When the MPU 32 executes a file write command to the active magnetic disk device 44 (step 200), the specified file is written to the active magnetic disk device 44 as an original file. is written (step 202).

Then, a file write command to the backup magnetic disk device 4 is executed (step 204), and the designated file is written to the backup magnetic disk device 46 as a backup file of the original file (step 206).

[Problems to be Solved by the Invention] However, in the past, the original file and its backup file were written to the active magnetic disk device 44 and the backup magnetic disk device 46, so If a failure of the disk device 44 occurs during file backup, a contradiction will occur between the processing contents at that time and the contents of the backup file used for SVP processing instead of the original file.

If the SVP system goes down at that time, a contradiction will occur between the original file stored in the active magnetic disk device 44 and the backup file in the backup magnetic disk device 46.

The present invention has been made in view of the above-mentioned conventional circumstances, and
The purpose is to provide a method that can reliably guarantee file contents even if a failure occurs in the active magnetic disk or the system goes down.

[Means for Solving the Problems] In order to achieve the above object, the following features are taken in the present invention.

In FIG. 1, when the original file 12 on the active disk 10 is rewritten with the new file 14, the active write flag 20 is set in the storage area 22 (step 100).

When writing of the new file 14 to the active disk 10 is completed, the active write flag 20 is reset and a copy 16 of the new file 14 is created on the main memory 18 (step 102).

Further, when the new file 14 is written to the spare disk 26 as a backup file 24 of the original file 12, the spare write flag 28 is set to the storage area 22.
(step 104).

When writing of the new file 14 to the spare disk 26 is completed, the spare write flag 28 is reset (step 106).

Here, when a failure occurs on the active disk 10,
As explained in FIG. 2, when the backup write flag 28 is set, the copy 16 on the main memory 19-8 is written to the backup disk 26 as a backup file 24 (step 108).

When the system is restarted due to a system failure, as shown in FIG. 12 is rewritten (step 110), and the backup write flag 28 is set, the active disk 10
The backup file 24 on the spare disk 26 is rewritten with the original file 12 (step 11).
2).

[Function] In the present invention, the active side disk 10 and the spare side disk 26
The original file 12 and the backup file 24 are sequentially written to the original file 12 and the backup file 24, and while these files are being written, the active side write flag 20 and the backup side write flag 28 are set to -10- respectively. It is set.

When writing is performed on the active disk 10, a copy 16 of the new file 14 is created on the main memory 18.

Here, if the backup write flag 28 was set when a failure occurred on the active disk 1o, it is confirmed from the flag set that the file was being backed up. A copy 16 with the same content as the new file 14 to be written to is written from the main memory 18 to the spare disk 26 as a backup file 24.

On the other hand, if the current writing plaque 20 is set when the system is restarted due to system failure, the setting of the flag 20 indicates that the original file 12 is being updated with the new file 14. The original file 12 on the active disk 1o is updated with the backup file 24 on the backup disk 26.

Furthermore, if the backup side write flag 28 is set to 1- at that time, the setting of the flag 28 indicates that the backup file 24 is being updated by the copy 16, so that the backup file 24 is the original. File 12 is updated.

Note that if the new file 14 (original file 12, backup file 24) is part of the specified file, this can be used by writing information indicating the location and range of the file writing to the storage area 22. It is also possible to limit the update of the disk file to the necessary parts (see examples).

[Example] Hereinafter, a preferred example of the method according to the present invention will be described based on the drawings.

In FIG. 4, the configuration of the embodiment is divided into ζ blocks according to functions.

In addition, Figure 5 explains the structure of the file control table created in the system control table area that the system refers to when starting up, and this file control table contains the file name, file write start address, file The write size, active side write flag, and backup side write flag are written.

Furthermore, the operation of the embodiment is explained in flowcharts in FIGS. 6, 7, and 8, and FIG.

In FIG. 10, the overall operation of the embodiment is explained using a timing chart.

When an update mode file access command (a write command to the active magnetic disk device 44) is executed in a process for operating, maintaining, controlling, etc. the main system in the disk access control unit 5o of FIG.
In step 600 of FIG. 6), the disk write control processing unit 52 is activated (step 602 of FIG. 6 and of FIG. 9).

Next, when the file control table processing section 54 is activated by the disk write control processing section 52 (Fig. 9), an entry is created in the file control table in the system control table area, and the file name, write start address, etc. The write size is set, and when writing is performed from the main storage device 38 to the active side magnetic disk device 44, the active side write flag is set (step 604 in FIG. 6, (2) in FIG. 9).

Thereafter, when the current disk write processing section 58 finishes writing the file to the current magnetic disk device 44, the file control table processing section 54 resets the current write flag, and a copy of the new file for updating the original file is stored in the main memory. It is created on the main memory by the device write processing unit 56 (step 606 in FIG. 6 and (2) in FIG. 9).

When a copy of the new file is created on the main memory in this way, the backup disk write processing unit 60 starts writing the file to the backup magnetic disk device 46.
At this time, the spare side write flag of the file control table is set (608 in FIG. 6, @ in FIG. 9).

When the file writing to the spare magnetic disk 46 is completed, the spare write flag is reset 14 and the corresponding file name is deleted from the file control table (step 60 in FIG. 6).
8. Figure 9 [Co., Ltd.]).

Here, between the time when file writing to the active side magnetic disk device 44 is started and the writing is completed, and the file writing is performed to the backup side magnetic disk device 46 (Fig. 9), the current side magnetic disk When a failure occurs in the device 44, the file recovery processing section 64 is activated by the system recovery processing section 62 (step 700 in FIG. 7, FIG. 10 (2)).

When the file recovery processing unit 64 is activated, the file control table is referenced (FIG. 10), and the back file in the spare magnetic disk device 46 is updated based on the contents of the file control table (FIG. 7, step 7). 702
．． FIG. 10 (1) to (1)) At that time, write flags on the active side are set for all entries registered in the file control table.

The presence or absence of the backup side write flag is searched, and only for entries for which the active 15- side write flag or the backup side write flag is set, the backup side magnetic disk device 46
Writing is performed for the write size from the write start address in the back file t in the file, and the entry to which this write has been performed is deleted from the file control table.

Thereafter, SVP processing is continued using the spare magnetic disk device 46 as the active disk device (step 704 in FIG. 7).

If a system down occurs during the above period (Fig. 9), the file recovery processing part 64 is activated by the system startup processing part 66 (step 800 in Fig. 8. ]) File recovery processing unit 6
4 is activated, the file table is referred to, and all entries registered in the file table are searched for whether or not the active side write flag and backup side write flag are set.

When an entry in which the current write flag is set is confirmed, the write size is read from the write start address in the corresponding file of the backup magnetic disk device 16-46, and the write size is read out from the corresponding file of the backup magnetic disk device 16-46. The entry for that file is deleted from the file control table (Step 802 in Figure 8 and ■ in Figure 10). Also, when the entry for which the write flag on the backup side was set is confirmed, the entry for the file on the active side is The write size is read from the write start address in the corresponding file of the disk device 44 and written to the spare magnetic disk 46, and the entry of this file is deleted (step 802 in FIG. 8).

Thereafter, the process of using the magnetic disk device 44 as the active side continues (step 804 in FIG. 8).
Figure 10 @).

As explained above, according to this embodiment, even if a failure occurs in the active magnetic disk device 44 during file backup, the new file to be backed up is transferred from the main memory to the backup magnetic disk device 46. Since it will be written,
SVP processing can be continued without any inconvenience by using a backup file whose contents do not conflict with the processing at the time of failure.

Furthermore, according to this embodiment, if a system failure occurs during file backup, files with correct contents between the active side and the backup side are written to the other side, so it is possible to reliably guarantee that the files on the active side and the backup side match. .

Furthermore, according to this embodiment, only the necessary portion of the file can be recovered using the write start address and write size set in the file control table, making it possible to facilitate and speed up file recovery. .

[Effects of the Invention] As explained above, according to the present invention, even if a failure occurs on the active disk during file backup or the system goes down, the correct file can be written to the active disk and spare disk. Therefore, even if a failure occurs on the active disk or the system goes down, it is possible to reliably guarantee the file contents.

[Brief explanation of drawings]

1, 2, and 3 are diagrams explaining the principle of the invention, Figure 4 is a functional block diagram explaining the configuration of the embodiment, Figure 15 is a diagram explaining the configuration of the file control table, Figure 6, Figures 7 and 8 are flowcharts explaining the operation of the embodiment, Figures 9 and 10 are timing charts explaining the operation of the embodiment, Figure 11 is an explanatory diagram of the SVP configuration, and Figure 12 is the conventional This is an explanatory diagram of the technology. 32 ・ 38 ・ 44 ・ 46 ・ 50 ◆ MPU. Main storage device, active magnetic disk device, spare magnetic disk device, disk access control section, 19-52 ・ 54 ・ 56 ・ 58 ・ 60 ・ 62 ・ 64 ・ 66 ・ Disk write control processing section, file control table processing section , main storage device write processing section, active disk write processing section, spare disk write processing section, system recovery processing section, file recovery processing section, system startup processing section.

Claims (2)

[Claims] (1) Original file (12) on the active disk (10)
is rewritten with a new file (14), the active write flag (20) is set in the storage area (22) (step 100), and the new file (14) is written to the active disk (10). When it is finished, the current write flag (20) is reset, a copy (16) of the new file (14) is created on the main memory (18) (step 102), and the new file (14) is used as a backup. Backup file (24) of original file (12) to side disk (26)
When written as , the spare side write flag (28)
is set in the storage area (22) (step 104
), when writing of the new file (14) to the spare disk (26) is completed, the spare side write flag (28) is set.
(step 106), and the active disk (1
When the backup side write flag (28) was set when a failure occurred in the copy (1) on the main memory (18),
6) is written to the spare disk (26) as a backup file 24 (step 108). (2) Original file (12) on the active disk (10)
is rewritten with a new file (14), the current write flag (20) is set in the storage area (22) (step 100), and writing of the new file (14) to the current disk (10) is completed. When the current write flag (20) is reset, a copy (16) of the new file (14) is created on the main memory (18) (step 102), and the new file (14) is transferred to the backup side. Backup file (24) of original file (12) to disk (26)
When written as , the spare side write flag (28)
is set in the storage area (22) (step 104
), when writing of the new file (14) to the spare disk (26) is completed, the spare side write flag (28) is set.
(step 106), and when the active side write flag (20) is set when the system is restarted due to a system failure, the backup file (24) of the spare disk (26) is used to write the active side disk (10). Rewrite the original file (12) of (step 110)
, the original file (12) on the active disk (10) when the backup side write flag (28) was set.
Backup file (2) on the spare disk (26)
4) (step 112).