JP2570104B2 - Failure information collection method in information processing equipment - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for collecting fault information of an information processing apparatus. The present invention relates to a failure information collecting method in an information processing apparatus that collects memory information via a system bus and collects hardware register information of a failed unit via a diagnostic bus.

[0002]

2. Description of the Related Art In an information processing apparatus, when a fault occurs, fault information is collected to analyze the cause of the fault.

FIG. 4 is a block diagram for explaining a failure information collecting method in a conventional information processing apparatus.

[0004] In FIG. 4, a main storage device 1, a diagnostic processor 2 ′, and a plurality of units 3 to 5 are connected to a system bus 6.

The main storage device 1 has a save area in which local memory information sent from a unit in which a fault has occurred is temporarily saved.

Each of the units 3 to 5 has dump firmware for sending local memory information to the main storage device 1 therein. When the dump firmware is started under the control of the diagnostic processor 2 ′, The local memory information of the own unit is transmitted to the main storage device 1.

The diagnostic processor 2 'is connected to the units 3 to 5 by a diagnostic bus 7, and has a maintenance / diagnosis function such as collecting fault data from a faulty unit.

[0008] The system bus 6 is a data bus capable of simultaneously transferring a plurality of bits of data, and enables high-speed data transfer between the units 3 to 5 and the main storage device 1.

The diagnostic bus 7 is a bus dedicated to maintenance / diagnosis for transferring data serially.

The floppy disk drive 8 is connected to the diagnostic processor 2 'and outputs fault information collected by the diagnostic processor 2'.

A conventional fault information collecting method in the information processing apparatus having such a configuration is as follows.

Now, for example, if a fault has occurred in the unit 3, the diagnostic processor 2 '
The information of the hardware register (hardware register information) is collected via the diagnostic bus 7.

Next, the diagnostic processor 2 ′ activates the dump firmware of the failed unit 3, and saves information (local memory information) of the local memory in the failed unit 3 to a save area on the main storage device 1. Let it.

Thereafter, the diagnostic processor 2 ′ outputs the hardware register information collected via the diagnostic path 7 to the floppy disk device 8 and the main storage device 1.
The local memory information saved in the upper save area is captured and output to the floppy disk device 8.

[0015]

In the above-mentioned conventional system, there is a problem that when the fault of the failed unit is related to the system bus, the local memory information of the failed unit cannot be collected.

An object of the present invention is to provide a fault information collecting method in an information processing apparatus capable of collecting local memory information of a faulty unit even when a fault related to a system bus occurs. It is in.

[0017]

In order to achieve the above object, the present invention provides: (A) a diagnostic processor comprising:
Register information is collected via a diagnostic bus that serially transfers data, and the dump firmware in the failed unit is collected.
Activating the air causes the localization of the faulty unit
In a fault information collecting method in an information processing apparatus for collecting memory information via a system bus for parallel data transfer , the diagnostic processor sends hardware register information of the fault unit to the diagnostic bus. Collecting hardware register information through the system bus based on the hardware register information collected by the hardware register information collecting means. Means for judging whether or not the faulty unit is possible, and when it is judged that the faulty unit can be collected, the dump firmware in the faulty unit is activated to execute the faulty unit via the system bus. Local
Evacuation collection means for collecting memory information; and local memory information collection means for collecting local memory information of the faulty unit via the diagnostic bus when the collection is determined to be impossible by the collection possibility determination means. Things.

In order to achieve the above object, the present invention provides: (B) a diagnostic processor comprising:
Register information is collected via a diagnostic bus that serially transfers data, and the dump firmware in the failed unit is collected.
Activating the air causes the localization of the faulty unit
In a fault information collecting method in an information processing apparatus for collecting memory information via a system bus for parallel data transfer , the diagnostic processor sends hardware register information of the fault unit to the diagnostic bus. Hardware register information collecting means for collecting the local memory information of the failed unit via the system bus by activating the dump firmware in the failed unit; and Evacuation success determining means for determining whether or not the local memory information has been successfully collected via the bus; Local memory information collecting means for collecting local memory information.

[0019]

In the configuration (A), the hardware register information collecting means collects the hardware register information of the failed unit via the diagnostic bus.

If the collection possibility determining means determines that the local memory information of the faulty unit can be collected via the system bus based on the hardware register information, the evacuation collecting means determines whether the dump firmware in the faulty unit is available. When the firmware is started, local memory information is collected via the system bus, and if it is determined that collection is not possible, the local memory
Collect memory information.

In the configuration (B), the hardware
Register information collecting means collects hardware register information of the failed unit via the diagnostic bus.

The collection of the local memory information of the failed unit is performed via the system bus when the save collection unit activates the dump firmware of the failed unit.

The evacuation success determination means determines whether or not the collection of local memory information via the system bus has succeeded. If the determination has failed, the evacuation success determination means uses the local memory information collection means via the diagnostic bus. Collect local memory information.

[0024]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, embodiments of the present invention will be described in detail with reference to the drawings.

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

In FIG. 1, a diagnostic processor 2 is connected to a main storage device 1 and a plurality of units 3 via a system bus 6.
5 and a plurality of units 3 via a diagnostic bus 7.
To 5 are connected. Further, a floppy disk device 8 is connected to the diagnostic processor 2.

As described above, the diagnostic processor 2 connected to each device includes the hardware / register information collecting means 2
1, evacuation determination means 22, evacuation means 23, evacuation success determination means 24, local memory information collection means 2,
5 and storage means 26.

Hardware register information collecting means 21
Has a function of collecting the hardware register information of the failed unit via the diagnostic bus 7.

The collection possibility judging means 22 includes a hardware
On the basis of the hardware register information collected by the register information collecting means 21, the local
It has a function of determining whether or not collection of memory information is possible.

The saving means 23 has a function of saving the local memory information of the failed unit to the save area on the main storage device 1 by activating the dump firmware in the failed unit via the diagnostic bus 7. Have.

The save success judgment means 24 has a function of judging whether or not the save of the local memory information by the dump firmware activated by the save means 23 is successful.

The local memory information collecting means 25 has a function of collecting local memory information of a failed unit via the diagnostic bus 7.

The storage means 26 has a function of storing the failure information in a floppy disk mounted on the floppy disk device 8.

FIGS. 2 and 3 are flow charts showing different processing examples of the diagnostic processor 2 at the time of occurrence of a fault. The processing can be instructed by a user's instruction or the like. First, a case where an instruction to perform the processing of FIG. 2 is given will be described.

Assume that a failure has occurred in the unit 3 among the units 3 to 5.

When the diagnostic processor 2 detects a failure in the unit 3, it starts the processing shown in the flowchart of FIG.
The hardware register information of the faulty unit 3 is obtained by the hardware bus
The sample is collected via (step S1).

Next, the diagnostic processor 2 determines whether or not the local memory information of the faulty unit 3 can be collected via the system bus 6 by using the collection possibility determining means 22 (step S2). This determination is made based on a plurality of hardware registers collected by the hardware register information collection unit 21.
This is performed based on the hardware register information indicating whether there is an abnormality in the access to the main storage device 1 in the register information. That is, if it is indicated that there is an abnormality, it is determined that collection is not possible, and otherwise, it is determined that collection is possible.

If it is determined in step S2 that collection is possible, the diagnostic processor 2 activates the dump firmware in the faulty unit 3 using the saving means 23 (step S3). As a result, the dump firmware transfers the local memory information of the failed unit 3 to the system bus 6.
Through the evacuation area on the main storage device 1 via the e-mail.

After that, the diagnostic processor 2
6, the evacuation unit 23 fetches the local memory information of the failed unit 3 evacuated to the evacuation area on the main storage device 1 and the hardware register information collected by the hardware register information collection unit 21. Is output to the floppy disk device 8 (step S4). As a result, the floppy disk device 8 writes the hardware register information and the local memory information to the mounted floppy disk.

If it is determined in step S2 that collection is impossible, the diagnostic processor 2 uses the local memory information collecting means 25 to collect local memory information of the failed unit 3 via the diagnostic bus 7 (step S5). ).

Next, the diagnostic processor 2
6, the hardware register information of the failed unit 3 collected by the hardware register information collecting means 21 and the local memory information of the failed unit 3 collected by the local memory information collecting means 25 are stored in the floppy disk drive 8. (Step S6). This allows
The floppy disk device 8 writes the hardware register information and the local memory information to the mounted floppy disk.

Next, a case where an instruction to perform the processing of FIG. 3 when a failure occurs is given will be described.

Now, for example, assume that a failure has occurred in the unit 3.

When the diagnostic processor 2 detects a failure of the unit 3, it starts the processing shown in the flowchart of FIG.
The hardware register information of the faulty unit 3 is obtained by the hardware bus
The sample is collected via the terminal (step S11).

Next, the diagnosis processor 2 includes
Then, the dump firmware in the failed unit 3 is started by using the firmware 3 (step S12). This allows you to
The firmware saves the local memory information of the failed unit 3 to the save area on the main storage device 1 via the system bus 6.

Thereafter, the diagnostic processor 2 determines whether or not the local memory information has been successfully saved to the main storage device 1 by using the save success determination means 24 (step S1).
3).

This determination is made, for example, when a predetermined time (a time sufficient for completing the save processing by the dump firmware) has elapsed since the start of the dump firmware, the execution of the failed unit 3 is executed. The determination is made based on whether the address is held at a predetermined address. That is, when the local memory information is saved to the main storage device 1 normally, the execution address of the failed unit 3 is held at the end address of the dump firmware. If it is held, it is determined that the evacuation has succeeded, and if not, it is determined that the evacuation has failed.

If it is determined in step S13 that the evacuation to the main storage device 1 has succeeded, the diagnostic processor 2 uses the storage means 26 to cause the evacuation means 23 to evacuate to the evacuation area on the main storage device 1. Fetches the local memory information of the failed unit 3 and outputs it and the hardware register information collected by the hardware register information collecting means 21 to the floppy disk device 8 (step S).
14). Thus, the floppy disk device 8 stores the hardware register information and the local memory information on the mounted floppy disk.

If it is determined in step S13 that the evacuation to the main storage device 1 has failed, the diagnostic processor 2
Collects the local memory information of the failed unit 3 via the diagnostic bus 7 using the local memory information collecting means 25 (step S15).

Next, the diagnostic processor 2
6, the hardware register information of the failed unit 3 collected by the hardware register information collecting means 21 and the local memory information collected by the local memory information collecting means 25 are output to the floppy disk device 8 ( Step S16). Thus, the floppy disk device 8 stores the local memory information and the hardware register information on the mounted floppy disk.

[0051]

As described above, according to the present invention, when the local memory information can be collected via the system bus, the evacuation collection means including the evacuation means 23 and the storage means 26 is used. It collects local memory information via the system bus quickly and collects local memory information via the diagnostic bus even if local memory information cannot be collected via the system bus. Thus, there is an effect that the local memory information can be collected as quickly and reliably as possible according to the state of the failure.

[Brief description of the drawings]

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

FIG. 2 is a flowchart illustrating an example of a process of a diagnostic processor 2;

FIG. 3 is a flowchart illustrating another processing example of the diagnostic processor 2;

FIG. 4 is a block diagram for explaining a conventional method.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Main storage device 1, 2, 2 '... Diagnostic processor 3-5 ... Unit 6 ... System bus 7 ... Diagnostic bus 8 ... Floppy disk device 21 ... Hardware / register information collecting means 22 ... Collecting possible judgment means 23 ... Evacuation means 24 ... evacuation success determination means 25 ... local memory information collection means 26 ... storage means

Claims (2)

(57) [Claims] A diagnostic processor collects hardware register information of a failed unit via a diagnostic bus for serially transferring data, and a dump file in the failed unit.
Activating the firmware causes the faulty unit
In a failure information collection method in an information processing apparatus that collects local memory information through a system bus that transfers data in parallel , the diagnostic processor includes hardware register information of the failure unit. Hardware register information collecting means for collecting the data via the diagnostic bus, and the local memory of the faulty unit via the system bus based on the hardware register information collected by the hardware register information collecting means. A collection possibility judging means for judging whether or not information can be collected; and when the collection possibility judgment means judges that the information can be collected, the dump firmware in the faulty unit is activated to start the system bus. Evacuation collection means for collecting the local memory information of the failed unit via Means for collecting local memory information of the faulty unit via the diagnostic bus when it is determined that the fault information cannot be collected by the means. method. 2. A diagnostic processor collects hardware register information of a failed unit via a diagnostic bus for serially transferring data, and a dump file in the failed unit.
Activating the firmware causes the faulty unit
In a failure information collection method in an information processing apparatus that collects local memory information through a system bus that transfers data in parallel , the diagnostic processor includes hardware register information of the failure unit. Means for collecting hardware register information via the diagnostic bus, and evacuation collection for activating the dump firmware in the failed unit and collecting local memory information of the failed unit via the system bus Means, evacuation success judging means for judging whether or not collection of local memory information via the system bus has succeeded, and when the evacuation success judgment means determines that collection of local memory information has failed, Local memory information collecting means for collecting local memory information via the diagnostic bus. In the failure information collection system in an information processing apparatus that.