JPS625443A - Diagnosis control method - Google Patents

Abstract

PURPOSE:To reduce a processing caused by the diagnosis result analysis of a service processor by permitting a service processor to diagnose a data processor in accordance with a request. CONSTITUTION:At the time of the leading of a system, the data processors 11-13 make self-diagnosis. The service processor 15 reads the self-diagnosis through a service bus 17 and stores it. The stored self-diagnosis is outputted to a main processor 14 through a system bus 16. The main processor 14 analyzes the self-diagnosis to inform an operator. When the operator requests diagnosis, the main processor 14 causes the service processor 15 to execute the diagnosis of the data processors 11-13. The results are reported to the main processor 14 through the system bus 16. Thus the processing caused by the diagnosis result analysis of the service processor can be reduced.

Description

DETAILED DESCRIPTION OF THE INVENTION (Technical Field of the Invention) The present invention relates to a diagnostic control method for a system having a main processor, a plurality of data processing devices, a service processor, a system bus, a service bus, and the like.

[Technical aspects of the invention and its problems]

In general, there is a main processor, a plurality of data processing devices, a service processor that manages and maintains the system, and a main processor, a bisbu O set month, and a data processor. 1′! It is independent of the system bus that connects the device.
In a system based on an i-bis bus, each data processing device performs a self-diagnosis to confirm the validity of the firmware-containing hardware upon startup of the system. The results of this self-diagnosis are read by the service processor from each data processing device via the service bus, and analyzed by the service processor itself. moreover,
If detailed diagnosis is required, the service processor executes diagnosis on the data processing device and analyzes the faulty part.

However, unlike 4'K, a detailed analysis of such diagnostic results requires a large amount of processing, and the burden on the service processor becomes extremely heavy. For this reason, some small and medium-sized systems do not perform diagnostic analysis, that is, failure analysis. Furthermore, when a new type of data processing device is added, it is necessary to add analysis means suitable for the added data processing device to the service processor, which also poses a problem in system expandability.

[Purpose of the invention]

The present invention has been made in consideration of the above circumstances, and aims to provide a diagnostic control method that reduces the processing involved in analyzing the diagnostic results of a service processor and also enables diagnostic analysis even in small and medium-sized systems. .

[Summary of the invention]

To achieve the above object, the present invention provides a main processor, a plurality of data processing devices, a service processor that manages and maintains these data processing devices, and a method for connecting the main processor, the data processing device, and the service processor. In a diagnostic control method for a system having a system bus and a service bus connecting the data processing device and the service processor, the data processing device is caused to perform self-diagnosis at system start-up, and the data processing device is connected to the service processor via the service bus. A service processor reads and saves the self-diagnosis results of the data processing device, 1 outputs the saved self-diagnosis results to the main processor via the system bus, and transmits the self-diagnosis results to the main processor. The main processor causes the service processor to diagnose the data processing device based on a diagnosis request from the operator, and the service processor analyzes the data processing device using a diagnostic analysis program and notifies the operator of the analysis result. The main processor is notified of the information via the system bus.

[Embodiments of the invention]

Hereinafter, the present invention will be described in detail with reference to the drawings.

FIG. 1 is a flowchart of a diagnostic control method according to an embodiment of the present invention, and FIG. 2 is a block diagram of a system to which the method of FIG. 1 is applied. As shown in FIG. 2, the system is constructed by connecting a plurality of data processing devices 11, 12, 13, a main processor 14, and a service processor 15 to a system bus 16° and a service bus 17, respectively.

Here, the service processor 15 manages and maintains the system.

When the system is powered on, each data processing device 11.
12.13 executes self-diagnosis (step 21). The results of the self-diagnosis by the data processing device 11.12.13 are read by the service processor 15 via the service bus 17 (step 22). Next, it is determined whether or not the read self-diagnosis result is an error (step 23).

If the result of this determination is an error, proceed to step 31;
It is determined whether or not this is an error in a device essential for starting up the system. If the error is in a device essential for starting up the system, the service processor 15 carries out analysis of the failure location and configuration control in step 33. On the other hand, if it is determined that the error is in a device that is not directly necessary for system startup,
In step 32, only the diagnostic results are saved.

Next, in step 24, it is determined whether or not reading of the self-diagnosis results of all data processing devices has been completed.If it has not been completed, the process returns to step 22 and the same process is executed again.

After all self-diagnosis results have been read, the service processor 15 continues to start up the system (step 25).
The self-diagnosis results stored in 5 are arranged in advance in the diagnostic analysis program in the main processor 14 (A-)
By entering the fixed area in July, the main professional 12
The diagnostic results are sent to the handle 14 (step 26).

ii' in main 10 retsuri 14? [gi analysis program] Analyzes the self-diagnosis results, displays the analysis results on the master station computer, etc., notifies system users, and also performs subsequent processing on the data processing device, e.g. Perform a detailed diagnosis or
Insert instructions such as whether to disconnect from the system (step 27).

Next, if there is a diagnosis request from each system user, the process proceeds to step 29, and if there is no diagnosis request, the process is terminated (step 28). In step 29, the main processor 14 sends input/output commands to the service processor 15 via the system bus 16 after writing the diagnostic request command channel program 43 and the diagnostic procedure into the memory of the main bus (J14). The service processor 15 that receives the input/output command retrieves the tunnel block from the memory at the address for storing the channel block sent along with the input/output command (mesotub 30).

Here, the retrieved channel []gram is a <r diagnosis request command as shown in the format 1-diagram in FIG. , executes diagnosis for the data processing device Nk indicated by the data processing device boot tunnel number, stores the results in a predetermined area of the main storage device, and ends the operation for the diagnosis request-1 mantle.

Next, 1 no-hist 1-] is set by analyzing the execution result of the diagnosis request command. The deceased RFlJ station is analyzed and the diagnosis results are notified to the main processor 14 via the system bus 16. Additionally, if further diagnosis is required as a result of the analysis, the main processor (14) similarly causes the service processor (15) to execute a diagnosis request command, and collects as much diagnostic data as necessary to analyze the failure location. Obtainable.

(Effects of the Invention) As described above, according to the diagnostic control method of the present invention, the service processor only needs to analyze the diagnostic results of the data processing device that is essential for system startup, and the diagnostic control method between the service processor and the main processor By providing a two-way interface for notification of results and diagnosis execution requests, service professionals will be able to reduce the burden of processing, and the main processor's software interface will enable flexible responses. This makes it possible to analyze diagnostic results even in small and medium-sized systems.

[Brief explanation of the drawing]

FIG. 1 is a flowchart of a diagnostic charge 12Ilty wide according to an embodiment of the present invention, FIG. 2 is a block diagram of a system to which the same diagnostic control method is applied, and FIG. 3 is a block diagram of a system to which the same diagnostic control method is applied. FIG. 11.12.13...Data processing device, 1/1...
Main processor 1, 15... Service processor, 1
6...System bus, 17...Revis bus.

Claims (1)

[Claims] A main processor, a plurality of data processing devices, a service processor that manages and maintains these data processing devices, a system bus that connects the main processor, the data processing device, and the service processor, and the data processing device and the service processor. In the diagnostic control method for a system having a service bus that connects the The saved self-diagnosis results are read and saved, and the saved self-diagnosis results are output to the main processor via the system bus, and the self-diagnosis results are analyzed by the diagnostic analysis program in the main processor, and the analysis results are presented to the operator. The main processor causes the service processor to diagnose the data processing device in response to a diagnosis request from an operator, and the service processor notifies the main processor of the diagnosis result via the system bus. A diagnostic control method characterized by: