JPH10161815A - Disk duplex management device and method therefor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To decide whether the state of a duplex disk is normal or abnormal or to grasp the occurrence of a fault of the duplex disk at an optional time. SOLUTION: The time when the normal or abnormal state of a duplex disk is confirmed is stored in a duplex disk parameter management table 2 together with the time when the detected abnormal duplex disk is restored by a normal disk and the time when the occurrence of a fault is notified if the restoration processing is not performed or abnormally finished. Then the normal or abnormal state of the duplex disk is confirmed, the abnormal duplex disk is restored and the fault of the duplex disk is notified at each time. In addition, the normal or abnormal state of the duplex disk is occasionally stored and updated. If one of both duplex disks is abnormal when a system starting request is produced by means of the abnormal disk, the other normal disk is used to start the system.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a disk duplication management apparatus and a disk system for improving the reliability and maintainability of a duplicated disk in a computer system having a duplicated disk by constantly grasping the status of a failure in the duplicated disk. It relates to a redundant management method.

[0002]

2. Description of the Related Art As a conventional dual disk management method,
Japanese Patent Laid-Open No. 81103/1993 discloses that when one disk fails, input / output is performed only on the normal disk, and the failed disk is restored from the normal disk after the failure is repaired. A method is described in which the system is restored without stopping the system by copying data.

[0003] Also, JP-A-7-36629 discloses that
A method is described in which, when it is determined that a failure is caused by a disk medium, formatting is performed in a minimum unit including a failed part of the failed disk medium, thereby restoring a duplex state.

[0004]

However, in the above-mentioned prior art, although a failed disk is recovered by a normal disk, it is determined at an arbitrary time whether or not the recovery processing has been completed normally. I couldn't do that. Therefore, for example, in a system that uses the primary disk of a dual disk to start up the system, if you try to restart the system before the failure that occurred on the primary disk is recovered, the secondary disk Even if the system is normal, the system is started by the primary disk, so that the system cannot be started.

Further, in the above-mentioned prior art, since the notification when the recovery processing of the failed disk is abnormally terminated is performed only at the time of the abnormal termination, it is noticed that the recovery processing is abnormally terminated in the unmanned system. Operation is continued without any problem, resulting in a double failure.

An object of the present invention is to provide a disk duplication management apparatus and a disk duplication management method which can grasp at any time whether the state of a duplicated disk is normal or abnormal.

Another object of the present invention is to provide a disk duplication management apparatus and a disk duplication management method which can grasp occurrence of a failure in a duplicate disk at an arbitrary time.

[0008]

According to a first aspect of the present invention, there is provided a disk duplication management apparatus for generating duplication state confirmation information indicating a timing for confirming whether a state of each disk constituting a duplicated disk is normal or abnormal. At the time indicated by the redundant disk management parameter table to be stored and the redundant status confirmation information stored in the redundant disk management parameter table, it is determined whether the status of each disk constituting the redundant disk is normal or abnormal. And a duplicated disk status checking means.

A second disk duplication management apparatus according to the present invention is a dual disk management apparatus for storing duplication state restoration information indicating a time when a disk in which an abnormality has been detected among disks constituting a duplicate disk is restored by a normal disk. A duplex disk recovery for recovering a disk in which an abnormality has been detected among the disks constituting the duplex disk with a normal disk at a time indicated by the parameter table and the duplex status recovery information stored in the duplex disk management parameter table. Means.

[0010] The third disk duplication management apparatus of the present invention is capable of detecting the occurrence of a failure when a recovery process is not performed on a disk in which an abnormality is detected among the disks constituting the duplicated disk or when the disk is abnormally terminated. A duplicated disk management parameter table for storing failure notification information indicating a notification timing; and a failure detection means for notifying the occurrence of a failure in the duplicated disk at a time indicated by the failure notification information stored in the duplicated disk management parameter table. And

[0011] A fourth disk duplication management apparatus according to the present invention comprises: a duplication status confirmation information indicating a time for confirming whether the status of each disk constituting the duplication disk is normal or abnormal; Of the disk to be recovered, which indicates when to recover a disk in which a failure has been detected by using a normal disk, and the time to notify the occurrence of a failure when the recovery processing is not performed or ends abnormally. Whether the status of each disk constituting the duplicated disk is normal at the time indicated by the duplicated disk management parameter table storing the failure notification information and the duplicated status confirmation information stored in the duplicated disk management parameter table. A duplex disk status checking means for checking whether an error has occurred, and the duplex disk management parameter; At the time indicated by the duplication state restoration information stored in the data table, duplicated disk restoration means for restoring a disk in which an abnormality has been detected among the disks constituting the duplicated disk with a normal disk, and the duplicated disk management parameter table A failure detecting means for notifying the occurrence of a failure in the duplicated disk at a time indicated by the failure notification information stored in the storage device.

A fifth disk redundancy management device according to the present invention, in the fourth disk redundancy management device, wherein the status of each of the disks constituting the duplicate disk is stored as normal or abnormal. A management table, wherein the duplicated disk status confirmation means updates the duplicated disk status management table according to the confirmed status of the duplicated disk, and the duplicated disk recovery means updates the duplicated disk according to the status of the duplicated disk after restoration processing. It is characterized in that the disk status management table is updated.

A sixth disk duplication management apparatus according to the present invention, in the fifth disk duplication management apparatus, wherein the duplicate disk comprises a system boot disk and a spare disk for the system boot disk.
When the system startup is requested, the redundant disk status management table is referred to, and when the system boot disk among the redundant disks is abnormal, the spare disk for the system boot disk is newly booted. Boot disk selecting means for selecting the disk as a disk for use.

According to a first disk redundancy management method of the present invention, redundant status confirmation information indicating a time when it is determined whether the status of each disk constituting a duplicated disk is normal or abnormal is stored in advance. At the time indicated by the duplication status confirmation information, it is checked whether the status of each disk constituting the duplication disk is normal or abnormal.

According to the second disk duplication management method of the present invention, the duplication state restoration information indicating the time when the disk in which the abnormality is detected among the disks constituting the duplicated disk is restored by the normal disk is stored in advance. At the time indicated by the duplication state restoration information, the disk in which an abnormality has been detected among the disks constituting the duplicated disk is restored with a normal disk.

The third disk duplexing management method according to the present invention is designed to prevent a failure when a recovery process is not performed on a disk in which an error is detected among the disks constituting the duplexed disk or when the disk is abnormally terminated. Failure notification information indicating the timing of notification is stored in advance, and the occurrence of a failure in the dual disk is notified to the timing indicated by the failure notification information.

According to a fourth disk duplication management method of the present invention, the duplication status confirmation information indicating a time for confirming whether the status of each of the disks constituting the duplication disk is normal or abnormal, and configuring the duplication disk. Of the disk to be recovered, which indicates when to recover a disk in which a failure has been detected by using a normal disk, and the time to notify the occurrence of a failure when the recovery processing is not performed or ends abnormally. A first step of preliminarily storing the failure notification information to be indicated and, at the time indicated by the duplex status confirmation information, confirming whether the status of each disk constituting the duplex disk is normal or abnormal, At the time indicated by the duplex status recovery information, a disk in which an abnormality is detected among the disks constituting the duplex disk is replaced with a normal disk. A second step of recovering Te, the timing at which the said fault notification information indicates, and a third step of notifying the occurrence of a failure in the redundant disk.

According to a fifth disk redundancy management method of the present invention, in the fourth disk redundancy management method, it is stored whether the status of each disk constituting the duplex disk is normal or abnormal. In the first step, the state of each of the stored disks is updated according to the confirmed state of the duplicated disk, and in the second step, the stored state of the duplicated disk is updated according to the state of the duplicated disk after recovery processing. It is characterized in that the status of each disk is updated.

A sixth disk redundancy management method according to the present invention, in the fifth disk redundancy management method, wherein the duplex disk comprises a system boot disk and a spare disk for the system boot disk.
When the system startup is requested, the state of each of the stored disks is referred to, and if the system boot disk among the duplicated disks is abnormal, the spare disk for the system boot disk is newly added. The disk is selected as a system boot disk.

[0020]

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

Referring to FIG. 1, in the embodiment of the present invention, a duplex disk management parameter setting means 1 for setting a management parameter by a system administrator using an information input / output means 6, Duplicated disk management parameter table 2 to be stored, and duplicated disk status management means 3 for managing the disk status based on the parameters stored in the duplicated disk management parameter table 2
And a redundant disk status management table 4 for storing a redundant disk status, and a boot disk selecting means 5 for selecting a boot disk with reference to the redundant disk status management table 4.

The information input / output means 6 functions as an interface with an input device, and the disk control means 7 functions as an interface with a disk device. Is omitted.

Next, the duplicated disk management parameter table 2 will be described with reference to FIG.

Referring to FIG. 2, there are three management parameters stored in the redundant disk management parameter table 2, namely, “redundant state confirmation”, “redundant state recovery”, and “failure notification”. You can set any type of start time.

Next, the dual disk status management table 4
Will be described with reference to FIG.

Referring to FIG. 3, the duplicated disk status management table 4 is provided with description columns for the partitions set for the connected internal disk and external disk, and describes the status of each disk. I can do it. The contents of the duplicated disk status management table 4 are updated by the duplicated disk status management means 3 as needed.

Next, the operation of the embodiment of the present invention will be described in detail with reference to FIG. 4, FIG. 5, and FIG.

FIG. 4 is a flowchart showing the operation of the dual disk management parameter setting means 1.

Duplicate disk management parameter setting means 1
First asks the system administrator whether to use the default values (step 40).
1). Here, if the default value is not used, the system administrator can specify the management parameter with a value suitable for the system (step 402). The value specified here is immediately reflected in the dual disk management parameter table 2.

Next, the duplicate disk management parameter setting means 1 inquires of the system administrator whether to start a management function based on the set management parameter value (step 403). The system administrator can select whether to start management for each item. When starting management, the management function is set to be valid (step 40).
4) In this case, the management function based on the value stored in the dual disk management parameter table 2 is started.
On the other hand, when the management is not performed, the management function is set invalid (step 405), and in this case, no management is performed on the item.

FIG. 5 is a flowchart showing the operation of the dual disk status management means 3.

The duplicated disk status management means 3 is operated according to the values set by the duplicated disk management parameter setting means 1.

That is, the dual disk status management means 3
Determines whether the time specified for the item “Confirm Duplex Status” of the duplicated disk management parameter table 2 has been reached or whether a failure has been detected (step 501). Alternatively, when a failure is detected, the duplexed disk status confirmation means 31 confirms the current duplexed status by the disk control means 7 (step 502). Here, when neither the designated time nor the failure is detected, nothing is performed until the next start time.

Next, the duplicated disk status management means 3 determines whether a failure has been detected as a result of the confirmation of the duplicated state by the disk control means 7 (step 503).
Here, when a failure is detected, the duplicated disk status management unit 3 immediately proceeds to the duplicated disk status management table 4.
Is updated (step 504).

The duplicated disk status management means 3 determines whether or not recovery is designated for the item of "duplicated status recovery" in the duplicated disk management parameter table 2 (step 505). Here, if the recovery is designated, the duplexed disk recovery unit 32 attempts to recover from the normal system by the disk control unit 7 (step 506), and determines whether the recovery processing has been completed normally (step 506). Step 507). Here, when the recovery is completed normally, the dual disk status management table 4 is updated again (step 508), and the next boot time is awaited. On the other hand, if there is no recovery specification for the item of “redundant state recovery” in the redundant disk management parameter table 2 or if the recovery processing ends abnormally, failure information is notified (step 509), and the next startup time is set. Meet.

FIG. 6 is a flowchart showing the operation of the boot disk selecting means 5.

When the system is started or restarted,
The boot disk selecting means 5 first checks the duplicated disk status management table 4 (step 601), and determines whether the primary disk is normal or abnormal (step 602). Here, if the primary disk (normally built-in disk) is normal, the disk is controlled so as to start with reference to the primary disk (step 603). On the other hand, if the primary disk is abnormal, the disk is controlled so as to start with reference to the secondary disk (normally an external disk) (step 60).
4). After startup, the redundant disk status management table 4
Is checked, and if an abnormality is detected, control is passed to the duplicated disk status management means 3 (step 605), and recovery is performed according to the value of the management parameter. After that, nothing is done until the next startup.

With the above, the operation of the embodiment of the present invention is completed.

According to the embodiment of the present invention, information about when to check the status of the duplexed disk can be set in the duplexed disk management parameter table, so that whether the duplexed disk is normal or abnormal can be arbitrarily determined. This has the effect of being able to ascertain at a point in time.

Also, by setting information about the timing of notifying the occurrence of a failure in the duplicated disk in the duplicated disk management parameter table,
This has the effect that the occurrence of a failure in the duplicated disk can be grasped at any time.

[0041]

Next, the configuration of an embodiment of the present invention will be described.

In this embodiment, it is assumed that one-to-one data duplication is performed between the internal disk device and the external disk device. After the data duplication is completed, the duplex disk management parameters are set as shown in FIG.

Since the setting of the management parameter has been validated, the status is checked at the set time, and if there is a fault, the duplex state is restored and a fault is notified.

Here, the partition 2 of the internal disk
Suppose that has failed.

At this time, the dual disk status management table 4 is updated as shown in FIG.

Even when a disk failure occurs, the restoration of the duplex state is performed based on the setting of the management parameter.

It is desirable that the dual disk status management unit be recorded on the disk and on the RAM.

Next, the operation of the embodiment of the present invention will be described.

When data duplication is completed, the duplicate disk management parameter setting means 1 sets management parameters via the information input / output means 6. First, the duplicated disk management parameter setting means 1 inquires of the system administrator whether to use the default value. The system administrator can specify a value when setting an arbitrary parameter by the user. The specified value is reflected in the dual disk management parameter table 2 as shown in FIG. In this case, according to FIG. 7, the dual disk status confirmation means 31 is executed every day at 3:00 am, and is also activated when the system is started and stopped. It is also activated when a failure occurs. The duplicated disk recovery means 32 is activated at the time of starting, stopping, and detecting a failure. When a failure is recognized, the failure notification using the failure detection unit 33 is performed at the time of starting, at the time of stopping, and at the time of occurrence of a failure. Whether to execute the management function can be selected by setting the validity flag to be valid or invalid. The management function is performed only when the validity is set. The processing at this time is performed according to the flow shown in FIG.

Once the management is started, the designated state of the duplicated disk 31 is activated at the designated time, and the state of the duplicated disk is confirmed. The process at this time is performed according to the flow shown in FIG. As a result of checking the duplex status, if the status is normal, nothing is performed until the next start time. In this example, the next start time is 3:00 am, when the system is stopped, when the system is restarted, or when a failure occurs. If an abnormality is detected as a result of the confirmation of the duplication status, the duplication disk status management table 4 is updated. FIG. 8 shows the contents of the duplicated disk status management table 4 in the updated state.
Here, it is assumed that a failure occurs in the partition 2. On the other hand, since the restoration is designated in the duplicated disk management parameter table 2, the duplicated disk restoring means 32 performs the restoration processing of the disk state. When the recovery processing has been completed normally, the dual disk status management table 4 is updated, and the next boot time is awaited. When the recovery processing ends abnormally, the failure information is notified via the information input / output means 6.

FIG. 6 shows the flow of processing when the system is restarted. If the system is restarted while the redundancy status management table 4 is in the state at the time of occurrence of the abnormality shown in FIG. 8, the abnormality of the primary system disk is detected. An instruction is given to the means 7. After starting by referring to the normal system of the disk, the duplicated disk status management table 4 is referenced again, and if an abnormality is detected, the duplicated disk management means 3
And performs recovery processing / failure detection processing.

As described above, when a disk failure occurs, particularly when a logical failure occurs, recovery from the failure is performed early, and even when a failure occurs, restart of the system is guaranteed.

[0053]

As described above, the present invention has an effect that whether the duplexed disk is normal or abnormal can be grasped at any time.

Further, the present invention has an effect that the occurrence of a failure in the duplicated disk can be grasped at any time.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an overall configuration of an embodiment of the present invention.

FIG. 2 is a diagram showing contents of a dual disk management parameter table 2 according to the embodiment of the present invention.

FIG. 3 is a diagram showing the contents of a dual disk status management table 4 according to the embodiment of the present invention.

FIG. 4 is a flowchart showing the operation of the dual disk management parameter setting means 1 according to the embodiment of the present invention.

FIG. 5 is a flowchart showing the operation of the dual disk status checking means 3 according to the embodiment of the present invention.

FIG. 6 is a flowchart showing the operation of the startup disk selecting means 5 according to the embodiment of the present invention.

FIG. 7 is a diagram showing the contents of a duplicated disk management parameter table 2 in one embodiment of the present invention.

FIG. 8 is a diagram showing the contents of a dual disk status management table 4 in one embodiment of the present invention.

[Explanation of symbols]

 REFERENCE SIGNS LIST 1 duplicated disk management parameter setting means 2 duplicated disk management parameter table 3 duplicated disk status management means 31 duplicated disk status confirmation means 32 duplicated disk recovery means 33 failure detection means 4 duplicated disk status management table 5 boot disk selection means 6 information input / output means 7 Disk control means

Claims (12)

[Claims] 1. A duplicated disk management parameter table for storing duplicated state confirmation information indicating a timing of confirming whether a state of each disk constituting a duplicated disk is normal or abnormal, and a duplicated disk management parameter table. A disk provided with a duplex disk status checking means for checking whether the status of each disk constituting the duplex disk is normal or abnormal at the time indicated by the stored duplex status confirmation information Redundant management device. 2. A duplicated disk management parameter table for storing duplicated state recovery information indicating a time when a disk in which an abnormality is detected among disks constituting a duplicated disk is restored by a normal disk, and the duplicated disk management parameter table At the time indicated by the duplication status restoration information stored in the duplicated disk, a duplicated disk restoring means for restoring a disk in which an abnormality has been detected among the disks constituting the duplicated disk with a normal disk. Disk duplication management device. 3. A failure notification information indicating a timing of notifying the occurrence of a failure when a recovery process is not performed on a disk in which an abnormality is detected among the disks constituting the duplexed disk or when the disk ends abnormally. A duplicated disk management parameter table, and a failure detecting means for notifying the occurrence of a failure in the duplicated disk at a time indicated by the failure notification information stored in the duplicated disk management parameter table. Redundant management device. 4. Duplication status confirmation information indicating a timing for confirming whether the status of each disk constituting the duplicated disk is normal or abnormal, and detecting an abnormality among the disks constituting the duplicated disk. A duplexed disk that stores redundant state recovery information indicating a time when a disk is recovered by a normal disk and failure notification information indicating a time when a failure occurs when the recovery processing is not performed or abnormally ends. A management parameter table, and a duplex disk status for checking whether the status of each disk constituting the duplex disk is normal or abnormal at the time indicated by the duplex status check information stored in the duplex disk management parameter table. Confirmation means; and the duplicated state stored in the duplicated disk management parameter table. At the time indicated by the recovery information, duplex disk recovery means for recovering a disk in which an abnormality has been detected among the disks constituting the duplex disk with a normal disk; and the failure notification information stored in the duplex disk management parameter table A failure detecting means for notifying the occurrence of a failure in the duplicated disk at the time indicated by (1). 5. A duplex disk status management table for storing whether the status of each disk constituting the duplex disk is normal or abnormal, wherein the duplex disk status check means checks the status of the duplicate disk. 5. The disk duplication according to claim 4, wherein the duplicated disk status management table is updated by the updating of the duplexed disk status management table according to the status of the duplicated disk after the recovery processing. Management device. 6. The dual disk includes a system boot disk and a spare disk for the system boot disk. When a system boot is requested, the dual disk refers to the dual disk status management table, and 6. The system according to claim 5, further comprising a boot disk selecting means for selecting the spare disk for the system boot disk as a new system boot disk when the system boot disk is abnormal. Disk duplication management device. 7. A duplication status check information indicating a timing for confirming whether the status of each disk constituting the duplexed disk is normal or abnormal is stored in advance, and at a timing indicated by the duplication status confirmation information, A method for managing disk duplication, comprising: checking whether the status of each disk constituting the duplicated disk is normal or abnormal. 8. Duplication state restoration information indicating a time when a disk in which an abnormality has been detected among disks constituting a duplicated disk is restored by a normal disk is stored in advance, and at a time indicated by the duplication state restoration information. And recovering a disk in which an abnormality has been detected among the disks constituting the duplicated disk by using a normal disk. 9. When failure recovery processing is not performed on a disk in which an abnormality is detected among disks constituting a duplexed disk or when the disk is abnormally terminated, failure notification information indicating a time of notifying the occurrence of a failure is stored in advance. A disk redundancy management method, wherein the method is characterized in that a failure is reported in the duplicated disk at a time indicated by the failure notification information. 10. Duplication status confirmation information indicating when to check whether the status of each disk constituting a duplicated disk is normal or abnormal, and detecting an abnormality among the disks constituting the duplicated disk. Duplicate state recovery information indicating a time when a disk is recovered by a normal disk, and failure notification information indicating a time when a failure occurs when the recovery processing is not performed or abnormally ends, is stored in advance. A first step of checking whether the status of each disk constituting the duplexed disk is normal or abnormal at the time indicated by the duplex status confirmation information; and at the time indicated by the duplex status recovery information, A second step of recovering a disk in which an abnormality has been detected among the disks constituting the duplex disk by using a normal disk; A third step of notifying the occurrence of a failure in the duplicated disk at a time indicated by the notification information. 11. storing whether the status of each disk constituting the duplexed disk is normal or abnormal, and, in the first step, storing the stored status according to the confirmed status of the duplexed disk; The disk duplication according to claim 10, wherein the status of each disk is updated, and in the second step, the stored status of each disk is updated according to the status of the duplicated disk after the recovery processing. Management method. 12. The duplicated disk includes a system boot disk and a spare disk for the system boot disk. When a system boot is requested, the stored disk status is referred to, and the duplicated disk is referred to. 12. The disk duplication management method according to claim 11, wherein when the system boot disk among the disks is abnormal, the spare disk for the system boot disk is newly selected as a system boot disk. .