JP2002215399A - Computer system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a computer system having a plurality of start devices switched, when a power supply is turned on. SOLUTION: This computer system comprises a plurality of storage devices (starting devices) 5 storing operating systems (OS#1 to OS#5) and a start processing part (BIOS) 11 for determining the priority of a plurality of storage devices 5, investigating the plurality of storage devices 5 in the order of priority, and starting an OS stored in the storage device 5 with the highest priority of the normal storage devices 5.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a computer system, and more particularly, to a computer system having a plurality of start-up devices and switching between them for continuous operation.

[0002]

2. Description of the Related Art Generally, a computer system such as a personal computer is put into an operating state and activated by the activation of an operating system (OS) as a control program. Specifically, when the power of the personal computer is turned on, the BIOS is loaded into the main memory and started. The started BIOS loads the OS stored in the hard disk device into the main memory and starts it, for example, and brings the computer system into an operating state. In this case, a device (boot device) for storing an OS to be loaded when the computer system starts up
Is specified in advance as a hard disk device.

[0003]

When the computer system is started, the hard disk device which is the starting device may fail. In this case, the computer system cannot operate. In order to operate the computer system, the user manually turns off the power once turned on, switches the boot device to another device (for example, a floppy disk device, etc.), turns on the power again, and turns on the new boot device. It is necessary to load and start the OS from. Therefore, the burden on the user is inevitable and cumbersome.

Further, in a computer system such as various servers on the Internet that operates continuously for 24 hours, a failure may occur in the OS that is normally started and is operating for some reason. In this case, the system hangs up as it is. That is, continuous operation becomes impossible. In order to prevent this, it is necessary to prepare a plurality of computer systems so that if one computer system fails, the other computer system can be replaced immediately. Therefore, the cost for improving the reliability of the computer system increases.

[0005] The present invention comprises a plurality of activation devices,
It is an object of the present invention to provide a computer system that switches between them when power is turned on.

It is another object of the present invention to provide a computer system which includes a plurality of boot devices and switches between them when an operating system failure occurs.

[0007]

A computer system according to the present invention examines a plurality of storage devices each storing an operating system (OS) and a plurality of storage devices in accordance with the priorities. A boot processing unit for booting the OS stored in the highest priority one of the normal storage devices.

According to the computer system of the present invention,
Even if a certain storage device (boot device) storing the OS fails at the time of the boot, the boot device can be booted by selecting other boot devices according to a predetermined order. Therefore, even when the boot device fails, the computer system can be operated without the burden of the user such as resetting the boot device.

Further, the computer system of the present invention comprises:
A plurality of storage devices each storing an OS, a boot processing unit that starts the OS stored in any of the plurality of storage devices, and an OS that is monitored by the boot processing unit and monitors the OS.
And a monitoring device that outputs a monitoring signal when a failure occurs in the storage device. The activation processing unit activates an OS stored in any of the plurality of storage devices according to the output of the monitoring signal.

According to the computer system of the present invention,
In a computer system that operates continuously for 24 hours, even if a failure occurs in an operating system that is normally started and is operating, another operating system without failure can be started. Therefore, even if a failure occurs in the operating OS, it is possible to prevent the system from hanging up and to enable continuous operation without preparing a plurality of expensive computer systems and switching between them.

[0011]

FIG. 1 is a computer system configuration diagram showing the configuration of a computer system according to the present invention.

In FIG. 1, the computer system includes:
It includes a CPU (Central Processing Unit) 1, a main memory 2, a monitoring device 3, a BIOS storage device 4, and a plurality of storage devices (hereinafter simply referred to as boot devices) 5 as boot devices. CPU
1 is the BI loaded and stored on the main memory 2
An OS, an operating system (OS), and an application program are executed. The OS controls the computer system by being executed on the CPU 1. The BIOS is stored in the BIOS storage device 4, and the OS
Is stored in each of the plurality of activation devices 5 (51 to 55), and an application program (not shown) is also stored in a storage device similar to the OS.

Each of the boot devices 5 has the same OS (OS #) as a control program of the computer system.
1 to OS # 5). The boot device 5 is composed of a nonvolatile storage device. Specifically, the activation device 5
Are a floppy disk device (floppy disk drive, FDD) 51, a disk-on-chip device (semiconductor disk device or silicon disk device) 52 composed of flash memory, a CD-ROM device (CD-ROM drive) 53, a hard disk device (hard disk drive) , HDD) 54 and flash memory (CompactFlash (registered trademark)) 55. In addition, CD-R
Instead of the OM device 53, a CDR device, a CDRW device, D
A VD device may be used. Further, some or all of the storage devices 51 to 55 may be replaced with another nonvolatile storage device.

In this example, the priorities of these boot devices 5 are determined by a boot processing unit 11 described later, for example, in a floppy disk device 51, a hard disk device 54,
CD-ROM device 53, disk-on-chip device 52,
The order is the flash memory 55. Note that a plurality of boot devices 5 may have the same priority.
This priority order can be changed for the floppy disk device 51, the hard disk device 54, and the CD-ROM device 53, and can be set, for example, by inputting a user's instruction from a well-known BIOS menu screen. The BIOS menu screen is displayed when the computer system starts up.
Displayed in response to a predetermined input from the keyboard by the user. The priority order of the disk-on-chip device 52 and the flash memory 55 may be changed by changing the setting of the BIOS menu screen.

In this example, the OSs stored in the respective boot devices 5 are the same, but are distinguished from one another by adding symbols # 1 to # 5. In this example, the symbols # 1 to # 5 are assigned in descending order of priority, but it is not always necessary to follow this order. Thus, the boot processing unit 11 and the like can know from which boot device 5 the OS is loaded. Therefore, instead of this, information indicating the boot device 5 storing the OS or information indicating the machine number of the boot device 5 may be used.

The activation device 5 is connected to different buses in principle. That is, in this example, the floppy disk device 51, the hard disk device 54, the CD-R
The OM device 53 and the disk-on-chip device 52
They are connected to different buses. On the other hand, the hard disk device 54 and the flash memory 55 are connected to the same bus. Thus, even if any of the buses of the computer has a failure, the OS of the boot device 5 connected to a bus other than the failed bus can be started. Note that the hard disk device 54 and the flash memory 55 may be connected to different buses of the same type (for example, an IDE bus).

The floppy disk device 51 is connected to an I / O (input / output) control circuit 6 via a dedicated connection line 96. The I / O control circuit 6 is connected to the ISA bus 91, and via this is connected to the bus conversion circuit 7. I /
The O control circuit 6 connects various input / output devices (for example, a keyboard and a mouse) to the ISA bus 91 and controls input and output of various signals to and from these devices. The ISA bus 91 has
The monitoring device 3 and the BIOS storage device 4 are also connected. The disk-on-chip device 52 is connected to the ISA bus 91 via an X bus (a bus similar to the ISA bus 91) 95.

The bus conversion circuit 7 has two IDE (PCI-
IDE) buses 93 and 94. The hard disk device 54 is connected to the bus conversion circuit 7 via the first IDE bus 93. As described above, the flash memory 55 is also connected to the first IDE bus 93. CD-R
The OM device 53 is connected to the bus conversion circuit 7 via the second IDE bus 94. Further, the bus conversion circuit 7
A connection is made between the CI bus 92 and the ISA bus 91. That is, the signal on the PCI bus 92 and the signal on the ISA bus 91 are mutually converted.

The bus control circuit 8 connects between the CPU 1 and the main memory 2 and the PCI bus 92. Bus control circuit 8
Performs transmission and reception of signals between the CPU 1 and the main memory 2;
It outputs signals from the PU 1 and the main memory 2 to the PCI bus 92, transmits signals from the PCI bus 92 to the CPU 1 or the main memory 2, and controls the main memory 2 (performs memory mapping).

The start-up processing unit 11 is provided with a BIOS (Basic Inpu
t Output System) program to BIOS storage device 4
And is loaded onto the main memory 2 and executed by the CPU 1. The BIOS storage device 4 is composed of, for example, a flash memory. The BIOS storage device 4
In addition to the area for storing the BIOS, a log information writing area (hereinafter, referred to as a CPU) for writing the log information by the activation processing unit 11 (that is, the BIOS) or the recording processing unit (or the interrupt processing unit) 13 (that is, the CPU 1) described later , Log area) 41 (for example, see FIG. 5A). Therefore, the log information is written to the storage device 4 in which the program (BIOS) constituting the boot processing unit 11 is stored. Since the log information is information on a fault that has occurred, it is possible to know which boot device 5 or its OS has a fault by referring to the log information. Instead of the log information, any information may be used as long as the information indicates which boot device 5 or its OS has failed.

The activation processing unit 11 activates the OS stored in any one of the activation devices 5 when the power is turned on and when an OS failure occurs. The activation processing unit 11 performs activation processing in response to turning on the power of the computer system. That is,
The OS stored in any one of the boot devices 5 without failure is started. At the time of this start processing, the start processing unit 11
Writes the log information in the log area 41 when the failed boot device 5 exists. Also, the activation processing unit 1
1 performs a startup process each time a monitoring signal to be described later is output. That is, the OS stored in any one of the boot devices 5 other than the boot device 5 in which the faulty OS that caused the output of the monitoring signal is stored is started.

The activation processing unit 11 determines the priority of the activation device 5 as described above. Also, the activation processing unit 1
1 checks the boot device 5 in accordance with the priority, and stores the boot device 5 in the normal boot device (that is, the boot device other than the boot device 5 in which the failed OS is stored) 5 having the highest priority. The loaded OS is loaded into the main memory 2 and activated. Specifically, the activation processing unit 1
1 examines the boot device 5 having the highest priority, as described later with reference to FIG. 2, and activates the OS stored therein if the boot device 5 is normal. Check the high storage device and repeat these processes. First, it is checked whether all of the plurality of boot devices 5 are normal or not, and based on the result, the O stored in the normal boot device 5 with the highest priority is stored.
S may be loaded.

The monitoring device 3 monitors the OS activated by the activation processing unit 11, and outputs a monitoring signal when a failure occurs in the OS. The monitoring device 3 of this example is composed of a well-known watchdog timer circuit, and outputs a hardware interrupt signal to the CPU 1 as a monitoring signal via the interrupt signal line 97 (and the bus conversion circuit 7). That is, CPU1
The OS (control processing unit) executed above updates (resets) the watchdog timer 3 at a predetermined cycle so that the timer does not time out. Therefore, the monitoring signal indicates that the OS has not updated the watchdog timer 3 in a predetermined cycle, that is, that the OS is not performing processing correctly (has a fault). By using the monitoring device 3 as independent hardware, even if there is a failure in the software OS, the failure can be monitored.

CPU 1 receiving hardware interrupt signal
Performs a predetermined interrupt process. That is, the recording processing unit 13 (accordingly, also referred to as an interrupt processing unit), which is an interrupt processing unit, records log information on the OS in which a failure has occurred in the log area 41. Therefore, the log information is written by the CPU 1 receiving the hardware interrupt signal as the monitoring signal. In response to the output of the monitoring signal, the activation processing unit 11 activates the OS stored in any one of the activation devices 5 based on the log information as described above. That is, the OS related to the failure
An OS other than the (OS in which the fault has occurred or the OS of the faulty boot device 5) is started.

FIG. 2 to FIG. 4 are flow charts of the OS startup process, showing the OS startup process and the monitoring process in the computer system of the present invention.

In FIG. 2, when the power supply of the computer system is turned on (step S11), the BIOS transmits the BIOS
The main memory 2 is loaded from the OS storage device 4 and activated (step S12). The activated BIOS, that is, the activation processing unit 11 activates the watchdog timer 3 (step S13), and refers to the log information written in the BIOS storage device 4 to refer to the log information written in the BIOS storage device 4.
Is detected (step S14). After that, the activation processing unit 11 updates the watchdog timer 3 at a predetermined cycle so that the watchdog timer 3 does not time out until the OS is activated.

The activation processing unit 11 checks each activation device 5 (storage device) in accordance with the priority order based on the log information referred to. It is checked whether or not the floppy disk device 51, which is the boot device 5 having the highest priority, has been booted in response to power-on (step S15).
The OS is loaded and started (step S16),
The computer system enters a normal operation state (step S17).

If the disk drive has not been started up normally, the priority is determined in the same manner as in the above-mentioned floppy disk drive 51 in accordance with the priority order.
Check each activation device 5. That is, it is checked whether or not the hard disk device 54, which is the boot device 5 having the next highest priority, has been started (step S18). If this is not normal, the CD-ROM device 53 having the next highest priority is started. Is checked (step S1).
9) If this is not normal, it is checked whether or not the disk-on-chip device 52 with the next highest priority is activated (step S110). If this is not normal, the next highest priority is given (the highest priority). It is checked whether or not the flash memory 55 (low) is activated (step S11).
1). If the flash memory 55 is not normal, step S19 is repeated. That is, the computer system enters a deadlock state. However, this probability is extremely low,
In practice, it hardly occurs. If any of the examined boot devices 5 has been normally booted, the boot device 5
OS is loaded and started (step S1).
6) The computer system enters a normal operation state (step S17).

In this process, the startup processing unit 11 skips (omits) the process of checking each startup device 5 (storage device) based on the log information referred to. For example, when the log information indicates that the floppy disk device 51 or the OS stored in the floppy disk device has a failure, the startup processing unit 11 skips step S15 after step S14 and executes step S18. I do. The same applies to the other boot device 5 and its OS.

FIG. 3A shows the operation of the activation device 5. In FIG. 3A, the power is supplied to the boot device 5 of the computer system when the power of the computer system is turned on (step S21). If the boot device 5 is normal according to the power supply, ,
A response signal is sent to CPU 1 (step S22). Thereafter, although not shown, the boot device 5 is in a waiting state, and when there is a loading request of the OS from the boot processing unit 11, the boot device 5 sends out the OS in response thereto.

FIG. 3B shows the processing executed by the activation processing unit 11. That is, steps S15, S18, S1 in FIG.
9, S110 and S111 are shown. FIG.
In (B), the activation processing unit 11 checks whether there is a response signal from the activation device 5 to be processed (step S3).
1). When receiving the response signal, the activation processing unit 11 stores the OS stored from the activation device 5 in the main memory 2.
And start it (step S3).
2). If no response signal is received, the activation processing unit 11
The log information indicating that the boot device 5 is not normal is written in the BIOS storage device 4 (step S33). At this time, an alarm may be output together with the writing of the log information.

FIG. 3C shows a process executed by the OS.
In FIG. 3C, the activated OS updates the watchdog timer 3 after a predetermined time has elapsed (step S41) so that the watchdog timer 3 does not time out (step S42), and repeats this during operation. When there is a failure in the OS, or when a failure occurs during loading to the main memory 2 (for example, when loading fails), the OS cannot update the watchdog timer 3 and a timeout occurs. . After this, the OS
Execute the application software (step S4
3) In the meantime, step S41 and subsequent steps are repeated at a predetermined cycle.

FIG. 4A shows the operation of the watchdog timer 3. In FIG. 4A, when the operation is started by the activation processing unit 11, the watchdog timer 3 starts counting time (step S51), and determines whether or not the count value is equal to or more than a predetermined value (whether or not time is over). Is checked (step S52), and if it is smaller than the predetermined value, the steps from step S51 are repeated. If the value is equal to or greater than the predetermined value, the watchdog timer 3 outputs an interrupt signal as a monitoring signal to the CPU 1 (step S53). The count value of the watchdog timer 3 is reset by the (normally operating) OS.

FIG. 4B shows a recording processing unit (interruption processing unit) 1.
3 shows the processing executed. In FIG. 4B, the CPU
1 checks whether there is an interrupt signal from the watchdog timer 3 (step S61), and if there is no interrupt signal, repeats step S51. If there is an interrupt signal, the CPU 1 writes log information indicating that the OS loaded from the boot device 5 is not normal to the BIOS storage device 4 (step S62), and starts the BIOS, that is, the boot processing unit 11 (step S63). . Thereafter, the activation processing unit 11 performs the same processing as in FIG. In addition,
An alarm may be output together with the writing of the log information.

FIGS. 5 to 7 are explanatory diagrams of the OS boot process, and show the OS boot process of the present invention.

In FIG. 5A, for example, when the power supply of the computer system shown in FIG. 1 is turned on (step S1).
1), the BIOS is started (step S12). At this time, the BIOS holds information on the priority order of the plurality of boot devices 5. The priority information includes, for example, OS # 1, OS # 2, OS # 3, OS # 4, O
This indicates that the order is S # 5.

In FIG. 5B, the boot processing unit 11 based on the BIOS attempts to load and start the OS # 1 from the floppy disk device 51 with reference to the log information # 1 at that time (step S1). S15). Note that, since the log information # 1 is at the time of startup, the log information about the startup device 5 is not normally included. Assuming that there is no response signal from the floppy disk device 51 (step S31), the start-up processing section 11 returns log information (#) indicating that the floppy disk device 51 has a failure.
2) is written (step S33). Log information # 2 is
Regardless of whether the OS # 1 of the floppy disk device 51 actually has a failure, the information is information indicating that the OS # 1 has a failure.

If there is no failure in the floppy disk device 51, the OS # 1 of the floppy disk device 51 is loaded and started.

In FIG. 6A, the activation processing unit 11
If the hard disk device 54 having the next highest priority is normal (step S18), the OS # 2 is loaded from the hard disk device 54 and activated (step S1).
6) The computer system enters a normal operation state (step S17).

In FIG. 6B, the control processing unit 12 by the OS # 2 in which the OS is loaded normally and no trouble occurs.
Updates the watchdog timer 3 at a predetermined cycle (step S42). Therefore, watchdog timer 3
Is not detected (step S5).
2) It is found that there is no obstacle in OS # 2.

In FIG. 7A, if a failure occurs in the OS # 2 thereafter, the watchdog timer 3 is not updated, so a timeout occurs (step S52), and an interrupt signal is output to the CPU 1 (step S53). ). CP
The interrupt processing unit 13 of U1 detects this (step S6).
1), O loaded from the hard disk drive 54
The log information (# 3) indicating that there is a failure in S # 2 is written (step S62), and the BIOS is started again (step S63).

In FIG. 7B, the activation processing unit 11
By referring to the log information (# 1 to # 3) up to that point (step S14), it is known that there is a failure in the floppy disk device 51 and the hard disk device 54 (OS # 1 and OS # 2). Step S, which is a startup process
Steps S19 and S18 are omitted, and Step S19 is executed. That is, the CD-ROM device 53 having the next highest priority
Is normal (step S19), the OS # 3
Is loaded and started (step S16), and the computer system enters a normal operation state (step S1).
7).

At this time, if the CD-ROM device 53 is not normal (step S31), the log information is written (step S33), and if the disk-on-chip device 52 having the next highest priority is normal. (Step S1
8) Then, the OS # 4 is loaded and activated (step S16), and the computer system enters a normal operation state (step S17).

Thereafter, when the user restores the floppy disk device 51 and its OS # 1 (only) to a normal one, the following processing is performed.

That is, the power of the computer system is once turned off, the system is restarted to display the BIOS menu screen, and from this screen, the log information # 1 indicating that the floppy disk device 51 has a failure is deleted (or Add a flag to invalidate). Thereafter, when the processing in FIG. 2 is performed, step S15 is executed without being omitted.
That is, the start-up processing unit 11 sends the floppy disk device 51
Is checked (step S15), and if it has been normally started, O is sent from the boot device 5 to the main memory 2.
Loading and starting S # 1 (step S16),
The computer system enters a normal operation state (step S17). The same applies to the restoration of the other activation devices 5.

[0046]

As described above, according to the present invention,
In the computer system, even if the boot device storing the OS at the time of booting is faulty, the boot device is selected by selecting boot devices having no other faults according to the priority and booting the OS. The computer system can be operated without the burden of the user such as fixing.

Further, according to the present invention, in a computer system which operates continuously for 24 hours, even if a failure occurs in a normally started and operating OS, another OS without failure is started by starting up another OS without failure. The system can be prevented from hanging up and continuous operation can be performed without employing a configuration in which a plurality of high-cost computer systems are prepared and switched between them.

[Brief description of the drawings]

FIG. 1 is a configuration diagram of a computer system.

FIG. 2 is an OS startup processing flow.

FIG. 3 is an OS startup processing flow.

FIG. 4 is an OS startup processing flow.

FIG. 5 is an explanatory diagram of an OS activation process.

FIG. 6 is an explanatory diagram of an OS startup process.

FIG. 7 is an explanatory diagram of an OS startup process.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 CPU 2 Main memory 3 Monitoring device 4 BIOS storage device 5 Plural storage devices (boot devices) 51 Floppy disk device (FDD) 52 Disk-on-chip device 53 CD-ROM device 54 Hard disk device (HDD) 55 Flash memory

Claims (15)

[Claims] 1. A plurality of storage devices each storing an operating system, a priority order is determined for the plurality of storage devices, the plurality of storage devices are checked according to the priority order, and the highest priority among normal storage devices is determined. A computer system comprising: an activation processing unit that activates an operating system stored in a device having a higher rank. 2. The boot processing unit checks a storage device with the highest priority, starts the operating system stored in the storage device when the storage device is normal, and stores a storage device with the next highest priority when the storage device is not normal. The computer system according to claim 1, wherein the step of repeatedly checking is performed. 3. The computer system further comprises a monitoring device that monitors an operating system started by the boot processing unit and outputs a monitoring signal when a failure occurs in the operating system. The computer system according to claim 1, wherein the activation processing unit activates an operating system stored in any of the plurality of storage devices in response to the output. 4. The computer system according to claim 1, wherein said plurality of storage devices comprise a nonvolatile storage device. 5. The plurality of storage devices include the floppy disk device, a hard disk device, a CD-ROM device, and a disk-on-a-chip device.
The computer system according to claim 1, wherein the computer system is connected to different buses. 6. The plurality of storage devices further include a flash memory, and the hard disk device and the flash memory are connected to the same bus or different buses of the same type. Claim 5
A computer system according to claim 1. 7. A plurality of storage devices each storing an operating system, a start processing unit for starting an operating system stored in any of the plurality of storage devices, and an operating system started by the start processing unit And a monitoring device that outputs a monitoring signal when a failure occurs in the operating system. The boot processing unit stores the monitoring signal in one of the plurality of storage devices according to the output of the monitoring signal. A computer system for activating a selected operating system. 8. The system according to claim 7, wherein the activation processing unit activates an operating system stored in one of storage devices other than the storage device storing the operating system in which the failure has occurred. A computer system according to claim 1. 9. The boot processing unit determines a priority order for the plurality of storage devices, checks the plurality of storage devices according to the priority order, and selects a storage device other than the storage device in which the failed operating system is stored. 9. The computer system according to claim 8, wherein an operating system stored in a storage device having the highest priority among normal storage devices is started. 10. The boot processing unit checks a storage device with the highest priority, starts the operating system stored in the storage device if the storage device is normal, and stores a storage device with the next highest priority if the storage device is not normal. 10. The computer system according to claim 9, wherein examining is repeated. 11. The computer system further includes a recording processing unit that records log information, and the activation processing unit activates an operating system other than the operating system related to the failure based on the log information. The computer system according to claim 8, wherein: 12. The computer system according to claim 11, wherein the log information is written in a storage device in which a program constituting the boot processing unit is stored. 13. The log information is written by a CPU that has received a hardware interrupt signal as the monitoring signal into a storage device in which a program constituting the boot processing unit is stored. 12. The computer system according to claim 11. 14. The computer system according to claim 7, wherein said monitoring device comprises a watchdog timer circuit, and outputs a hardware interrupt signal as said monitoring signal. 15. The computer according to claim 7, wherein the activation processing unit performs activation processing in response to turning on of the power of the computer system, and performs activation processing each time the monitoring signal is output. Computer system.