JP2000324105A - Collection system or fault information - Google Patents

Abstract

PROBLEM TO BE SOLVED: To collect the fault information centering around the data of higher priorities and also to collect the representative data of lower priorities hen plural faults simultaneously occur by locking the fault information which cannot be updated if the fault occurs while another fault is processed. SOLUTION: An input/output processor 5 notifies a central processor 3 of a fault occurring at a channel part 6a and also locks a fault information saving area 2 to prepare to transfer the fault information data. Then a fault information control part recognizes that the transfer of the fault information data is prepared. Meanwhile, if the fault information control part recognizes a fault of a channel part 6b, the fault of the part 6b is informed to the processor 5. When the processor 5 notifies the processor 3 of the fault of the channel 6b, the fault information control part acquires a lock flag and inhibits writing to a fault information data part while the lock flag is acquired.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

TECHNICAL FIELD The present invention relates to a fault information collecting method,
In particular, the present invention relates to a failure information information processing apparatus.

[0002]

2. Description of the Related Art Conventionally, in this type of fault information collecting method, when a plurality of fault information has occurred, effective information is stored in a fault information save area on a main storage unit in the information processing apparatus. Exclusive control of the failure information save area was performed by providing a lock byte indicating that the information was not collected. For this reason, only the failure information for which the lock byte of the failure information save area has been acquired is subject to failure information collection regardless of the priority of the failure information, and failure information for which the lock byte cannot be acquired has been discarded. The data of the fault information collected was the same regardless of the cause of the fault.

On the other hand, for example, Japanese Patent Application Laid-Open No. 7-23041
As disclosed in the “line fault notification method” of Japanese Patent Application Laid-Open Publication No. H10-207, a technique of collecting high-priority fault information is used by using a technique of collecting only high-priority fault information in an information processing apparatus. .

FIG. 5 is a block diagram showing an example of a conventional fault information collecting method. This failure information collecting method includes a main storage device 1 including a failure information save area 2, a central processing unit 3, a diagnostic processor 4, an input / output processing device 5, a channel unit (a to
d) A fault detector 9 including 6, peripheral devices (a to d) 8, and a general-purpose processing device interface bus 7.

An operating system (OS) (not shown) sends an I / O signal to the central processing unit 3 connected to the main storage device 1 via the general-purpose processing device interface bus 7.
Execute the instruction. The central processing unit 3 prepares a channel program for executing I / O (input / output) instructions on the main storage device 1, and sends the channel program to the availability processing device 5 via the general-purpose processing device interface bus 7. Notify that has been prepared.

After several conversations between the central processing unit 3 and the input / output processing unit 5 via the general-purpose processing unit interface bus 7, the channel units 6a to 6
The channel program is executed by performing data transfer between the peripheral devices 8a to 8d under d and the main storage device 1. Further, a diagnosis processor 4 for diagnosing a failure of the information processing device, collecting data of the failure information, removing a lock byte in the failure information save area 2 after the data collection, and initializing the information processing device is provided. It is connected via a general-purpose processor interface bus 7.

In the configuration processing apparatus as described above, for example, when a failure occurs in the channel units 6a and 6b, the failure detection unit 9 detects the failure. And the input / output processing device 5
To each of them. The input / output processing device 5 recognizing this notifies the central processing device 3 that there has been a failure from the channel.

As a result, the OS operates in the channel sections 6a and 6a.
It recognizes that a failure has occurred in b. The input / output device 5
The lock byte in the failure information save area 2 is acquired, and the failure information is stored in the failure information save area 2 in the main storage device 1.
To transfer failure information to the server. At the same time, by notifying the diagnostic processor 4 that a failure has occurred,
The diagnostic processor 4 collects the fault information from the fault information save area 2 and releases the lock byte on the fault information save area 2. At this time, the fault information save area 2
Since there is only one per device, the first channel 6a
When the failure is written, the lock byte on the failure information save area 2 is acquired and the failure information save area 2 is locked.

In the subsequent failure of the channel section 6b, the failure information save area 2 cannot be locked, so that the failure information of the channel section 6b is discarded. The reason why there is only one fault information save area 2 per device is that when there are multiple faults at the same time, it is impossible to predict how much resources should be held.

Next, FIG. 6 is a detailed configuration diagram of a portion including the fault detecting section 9 in the fault information processing device shown in FIG. The failure detection unit 9 is connected to the channel units (a to d) 6 and includes a failure factor analysis unit 34 and a control unit 32. The fault information from each of the channel units (a to d) 6 is input by being grouped into three types of faults 33a to 33c of A to C. The priority is 3 among the A to C failure types 33a to 33c.
A priority determining means (not shown) in which 3a>33b> 33c is set is provided. When failure information is generated from a plurality of groups among the three groups, one failure information given priority by the above-described priority order determination means is notified from the failure factor analysis unit 34 to the control unit 32. in this case,
Since the processing is performed for the failure notification of one of the groups that has priority, a more effective failure information analysis can be performed.

[0011]

The above-mentioned problem of the prior art is that, for a plurality of faults, only one fault information which has occurred earlier can be collected, and there is a possibility that important fault information cannot be collected. .

In the techniques of FIGS. 5 and 6, priorities are assigned to a plurality of faults, and when fault information for a fault with a higher priority is collected, fault information for a lower priority is not collected at all. Can not. Furthermore, since the failure information to be collected is fixed data even if the failure factors are different, there is a problem that unnecessary data or detailed data is insufficient depending on the factors.

[0013] An object of the present invention is to collect fault information mainly on data of fault information with high priority and collect representative data on fault information with low priority when a plurality of faults occur simultaneously. The purpose is to effectively use a limited fault information save area by flexibly changing data to be collected depending on the cause of a fault.

[0014]

SUMMARY OF THE INVENTION In order to solve the above-mentioned problems, a failure information collecting method according to the present invention employs the following characteristic configuration.

(1) In a fault information collecting method having a fault detecting unit for detecting a fault and a fault processing unit for the fault which has occurred, when another fault occurs while the fault processing unit is executing fault processing. A failure information collecting method for determining a priority order and setting a lock state in which the failure information cannot be updated.

(2) In a fault information collecting method for collecting fault information including a plurality of channel units each connected to a peripheral device and a fault detecting unit for detecting the occurrence of a fault in the channel unit, the system is connected to the fault detecting unit. A fault information collecting method including a fault information collecting unit including a fault information buffer for storing the fault information.

(3) The fault information collecting method according to the above (2), wherein the fault information buffer includes a fault information data section which is in a lock state and a lock release state for protecting stored data.

(4) The failure information data section includes a common section for writing common data regardless of the cause of the failure,
The fault information collecting method according to the above (3), further comprising a flexible section for writing optimum data according to the factor.

(5) The fault information collecting unit has a fault information control unit that compares the priorities of fault information that occur sequentially and notifies the fault information buffer of the fault information. Failure information collection method.

[0020]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A preferred embodiment of a failure information collecting system according to the present invention will be described below in detail with reference to FIGS. Note that the same reference numerals are used for the same elements as those of the conventional failure information collecting method shown in FIGS. 5 and 6 for convenience.

FIG. 1 is a block diagram showing a preferred embodiment of a failure information collecting system according to the present invention. This failure information collecting method includes a main storage device 1 including a failure information save area 2, a central processing unit 3, a diagnostic processor 4, an input / output processing device 5, a channel unit (a to d) 6, a general-purpose processing device interface bus 7, Peripheral devices (a to d) 8, failure detection unit 9
And a failure information collecting unit 10.

An OS (not shown) executes an I / O instruction for the central processing unit 3 connected to the main storage device 1 via the general-purpose processing device interface bus 7. A channel program for executing an I / O instruction is prepared on the central processing unit 3, and the I / O processing unit 5 is notified via the general-purpose processing unit interface bus 7 that the channel program has been prepared.

After some communication between the central processing unit 3 and the input / output processing unit 5 via the general-purpose processing unit interface bus 7, the channel units 6a to 6
The channel program is executed by performing data transfer between the peripheral devices 8a to 8d under d and the main storage device 1. Further, a diagnostic processor 4 for diagnosing a failure of the information processing apparatus, collecting data of the failure information, removing a lock byte in the failure information save area 2 after the data collection, and initializing the information processing apparatus is generally used. It is connected via a processing device interface bus 7.

A failure detector 9 for detecting that a failure has occurred in any of the channel units 6a to 6d is connected to the corresponding channel unit 6. The failure information collecting unit 10 for notifying that a failure has occurred in the failure detection unit 9 is connected between the input / output processing device 5 and the failure detection unit 9.

Next, FIG. 2 shows the fault information collecting unit 10 shown in FIG.
It is a detailed block diagram of. This failure information collecting unit 10
It comprises a control unit 12 connected to the failure detection unit 9 and a failure information buffer 11 including a right holder register 13, a lock flag 14, and a failure information data unit 15.

Next, the operation of the fault information collecting method according to the present invention will be described with reference to FIGS. In the failure information collection method of FIG. 1, it is assumed that a failure of a high priority factor B occurs, for example, from the channel unit 6b immediately after a failure of the low priority factor A occurs, for example, from the channel unit 6a. The failure detection unit 9 recognizes the failure of the channel unit 6a where the failure has occurred first, and notifies the failure information collection unit 10 of the failure.

The control circuit 12 in the fault information collecting unit 10
Upon detecting this, the cause A from the channel section 6a is written into the right holder register 13 in the failure information buffer 11. At this time, the failure information control unit 12 sets the lock flag 1
4 and the right holder register 13 are checked. Then, it is confirmed that the failure information control unit 12 does not come out of the locked state and that there is no valid failure information in the failure information data unit 15.

FIG. 3 is a detailed configuration diagram of the fault information data section 15 in FIG. As shown in FIG. 3, the failure information data section 15 includes a common section 16 and a flexible section 17. Common data is written to the common unit 16 irrespective of the factor, but data to be written to the flexible unit 17 changes depending on the factor. That is, at this point, the factor A
The most suitable data is collected and written to the flexible unit 17.

The fault information collecting unit 10 is provided with the input / output processing unit 5
Is notified that a failure has occurred in the channel unit 6a. The input / output processing device 5 recognizing this notifies the central processing device 3 that a failure has occurred from the channel unit 6a. Accordingly, the OS recognizes that a failure has occurred in the channel unit 6a via the central processing unit 3, and performs a failure occurrence process for the channel unit 6a. Generally, the channel section 6
Recovery processing is performed after the disconnection processing of a or the reset of the channel unit 6a.

On the other hand, the input / output processing device 5 prepares for transferring the failure information data to the failure information save area 2 on the main storage device 1. However, if the failure detection unit 9 recognizes the failure from the next failed channel unit 6b before the data transfer and notifies the failure information collection unit 10 of the failure, the failure information Sampling unit 1
The control unit 12 in the error information buffer 11 detects the cause of the cause B from the channel unit 6b.
Attempts to write to the right holder register 13 in the.

At this time, the lock flag 14 and the right holder register 13 are checked as in the case described above. In this case, although the lock flag 14 has not been acquired, the data of the factor A from the channel unit 6a has already been
The fault information control unit 12 writes the
And the priority of the factor B. As a result, since the factor B has a higher priority than the factor A, the control unit 12 extracts only the representative data from the factor A of the fault information data unit 15 and sends the representative data of the factor A from the channel unit 6a to the common unit 16. Write down the failure and representative data. Thereafter, the failure information of the factor A from the channel unit 6a in the flexible unit 17 is discarded, and the optimal data for the factor B is written to the flexible unit 17. Further, data indicating that the factor B from the channel unit 6b has occurred is written into the right holder register 13.

On the other hand, the fault information collecting unit 10 notifies the input / output processing unit 5 that a fault has occurred in the channel unit 6b. The input / output processing device 5 that recognizes this notifies the central processing device 3 that there has been a failure from the channel unit 6b. Thus, the OS recognizes that a failure has occurred in the channel unit 6b via the central processing unit 3, and the OS performs a failure occurrence process for the channel unit 6b.

Further, after acquiring the lock byte in the failure information save area 2, the input / output processor 5 finishes preparing the transfer of the failure information data. Inform. Upon receiving this notification, the failure information collecting unit 10 acquires the lock of the lock flag 14 in the failure information buffer 11 to protect the data of the failure information data unit 15. When this is completed, the data in the fault information data section 15 is actually
The data is transferred to the failure information save area 2 in the main storage device 1. When the data transfer is completed, the failure information control unit 12
Clears the right holder register 13 and the lock flag 14 in the failure information buffer 11.

On the other hand, the input / output processing unit 5 notifies the diagnostic processor 4 that valid fault information has been transferred to the fault information save area 2. As a result, the diagnostic processor 4 collects the fault information from the fault information save area 2, and releases the lock byte on the fault information save area 2 when the collection is completed.

Next, an example of the above-described series of operations will be described with reference to FIG.
This will be described in detail with reference to FIG. When recognizing a fault in the channel unit 6a, the fault information control unit 12 checks whether the lock flag 14 is locked (step 1). Here, if the lock is set, the collection of the failure information is abandoned, and the processing is skipped to the failure notification to the input / output processing device 5.

If it is not locked in step 1,
The right holder register 13 is checked, and if it is not free, the priorities are compared (step 2). Right holder register 13
Is free or has a higher priority than the data written in the right holder register 13, if the factor A from the channel unit 6a
Is written in the rights holder register 13 and
The data is written in the failure information data section 15 in the format of the cause A, and the representative data of the data written in the right holder register 13 is left in the failure information data section 15 as a history (step 3).

The fault information control unit 12 is provided with the input / output processing unit 5
Is notified that a failure has occurred from the channel unit 6a.
The input / output processing device 5 includes a channel unit 6a in the central processing unit 3.
In addition to notifying that a failure has occurred, the lock of the failure information save area 2 is taken, and preparations are made for data transfer of failure information. When you are ready to transfer data,
The failure information control unit 12 is notified that the preparation is completed.
During this time, if the failure information control unit 12 recognizes a failure in the channel unit 6b, it executes steps 1, 2, and 3 as described above. In step 3, if the priority is lower than the data written in the right holder register 13, only the representative data is written in the failure information data section 15 as a history.

The fault information control unit 12 is provided with the input / output processing unit 5
Is notified that there is a failure from the channel unit 6b.
The input / output processing device 5 includes a channel unit 6b in the central processing unit 3.
Notify that there was an obstacle from.

When the failure information control unit 12 is notified that the data transfer is ready, the failure information control unit 12 acquires the lock flag 14. While the lock flag 14 is being acquired, writing to the failure information data section 15 is prohibited. The failure information control unit 12 activates data transfer from the failure information data unit 15 to the failure information save area 2 in the main storage device 1. When the data transfer is completed, the lock flag 14 is released and the right holder register 13 is released.
Also clear.

The input / output processing device 5 notifies the diagnostic processor 4 that the failure data has been transferred, so that the diagnostic processor 4 collects the failure information from the failure information save area 2 and, when the collection is completed, Then, the lock byte on the failure information save area 2 is released.

With these operations, when a plurality of faults occur simultaneously, more data of fault information having a higher priority is collected, and fault information having a lower priority is also collected using only representative data. The purpose is to use limited data effectively by flexibly changing the data collected depending on the cause of the failure.

The configuration and operation of the preferred embodiment of the failure information collecting system according to the present invention have been described above in detail. However, the present invention should not be limited to only such specific examples, and those skilled in the art can easily understand that various modifications can be made without departing from the gist of the present invention.

[0043]

As can be understood from the above description, according to the fault information collecting method of the present invention, even when a plurality of faults occur simultaneously, more fault information of higher priority is collected and fault information of lower priority is obtained. The information has a practically remarkable effect that the representative data can be collected and the collected data can be flexibly changed depending on the cause of the failure.

[Brief description of the drawings]

FIG. 1 is a block diagram of a preferred embodiment of a failure information collecting method according to the present invention.

FIG. 2 is a detailed block diagram of a failure information collecting unit in FIG.

FIG. 3 is a configuration diagram of a failure information data section in FIG. 2;

FIG. 4 is an operation time chart of the failure information collecting method shown in FIGS. 1 to 3;

FIG. 5 is a block diagram illustrating an example of a conventional failure information collection method.

FIG. 6 is a block diagram illustrating a configuration of a fault information detection unit in FIG. 5;

[Explanation of symbols]

 6a to 6d Channel unit 8a to 8d Peripheral device 9 Failure detection unit 10 Failure information collection unit 11 Failure information buffer 12 Failure information control unit 15 Failure information data unit 16 Common unit 17 Flexible unit

Claims (5)

[Claims] In a fault information collecting method having a fault detecting unit for detecting a fault and a fault processing unit for the fault that has occurred, when another fault occurs during execution of fault processing in the fault processing unit, A failure information collecting method for determining a priority order and setting a lock state in which the failure information cannot be updated. 2. A fault information collecting method for collecting fault information including a plurality of channel units connected to peripheral devices and a fault detecting unit for detecting occurrence of a fault in the channel unit, wherein the fault information collecting system is connected to the fault detecting unit; A failure information collection method, comprising: a failure information collection unit that includes a failure information buffer that stores the failure information. 3. The fault information collecting method according to claim 2, wherein said fault information buffer includes a fault information data section which is in a lock state and a lock release state for protecting stored data. 4. The fault information data section has a common section for writing common data regardless of the cause of the fault, and a flexible section for writing optimum data according to the cause. The failure information collection method according to claim 3. 5. The fault information collection unit according to claim 2, further comprising a fault information control unit that compares priorities of the fault information sequentially generated and notifies the fault information buffer.
The failure information collection method described in 3 or 4.