JP2001236602A - Testing method of disk operation - Google Patents

Abstract

PROBLEM TO BE SOLVED: To much more surely detect the failure possible to generate at the access time of a disk. SOLUTION: The disk access system, the length of test data and the data pattern are set as parameters at the time of performing the operation test. At this time, the plural methods are allowed to set in the plural by the disk access system. The contents of the test are set to the unit of the disk objective for the test, and the setting of access timing with respect to the plural disks is also allowed. Then, the data are written by issuing the demand of disk access, or the like, at the same time in the plural with respect to the plural disks in accordance with the parameters manually or automatically set inside, and after that, the failed place is detected by comparing them with the data read from the disk.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a disk operation test method, and more particularly to an improvement in detection accuracy of a defective portion in an operation test.

[0002]

2. Description of the Related Art When a system using a computer involves access to a fixed disk device, a disk operation test including a fixed disk device and a fixed disk control device at the time of product development, product manufacture, or system delivery. I do. This operation test is performed using an operation test program created by determining the test content and assuming various cases so that the undetected defective portion does not occur as much as possible. The content of the test is set, for example, so that a desired address can be accessed with a desired data length. The disk access method is performed by sequential access or random access. Further, when a plurality of fixed disk devices are connected to the fixed disk device, the fixed disk device may be overloaded while taking into account the timing of accessing the plurality of fixed disk devices. Perform a test including.

[0003]

However, in the prior art, when performing continuous disk access, the disk access method is usually fixed, and even if a test is performed using a plurality of disk access methods, for example, the sequential access is performed. A random access was performed after the access test. That is, when the operation is actually started, a plurality of programs having different disk access methods may simultaneously issue an access request. However, when this access request is regarded as a series of disk accesses, the fixed disk controller Means that an access request is issued to the fixed disk device by mixing different disk access methods. However, in the past,
Since such a test environment that matches the operation at the time of actual operation is not set, a defective part may be detected after the operation is started. That is, in the conventional operation test method, there is a case where a defective portion cannot be detected even if an operation test is performed.

In addition, when a test program is used, the contents to be tested (test execution items) are particularly likely to be fixed, and the operation test to the fixed disk device according to the system is not always executed. For this reason, it is easy to overlook even when a defective part can be easily detected by setting differently.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-described problems, and an object of the present invention is to provide a disk operation test method capable of more reliably detecting a defect that may occur during disk access. Is to do.

[0006]

In order to achieve the above object, a disk operation test method according to the present invention comprises:
In a disk operation test method for performing an operation test including a fixed disk device alone or a fixed disk control device,
The operation test is performed by accessing the fixed disk device to be tested by mixing the disk access methods.

In a disk operation test method for performing an operation test including a fixed disk unit or a fixed disk controller, at least a disk access method, a test data pattern and a data It is possible to set a test content for changing any of the lengths irregularly.

Further, operation tests on a plurality of fixed disk devices are performed simultaneously in parallel.

[0009]

Preferred embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a schematic configuration diagram of a computer system in which an embodiment of a disk operation test method according to the present invention is performed. There is no characteristic in the system configuration itself, and a general CPU 1, memory 2, and two disk controllers 3 are connected by a bus 4. A plurality of fixed disk devices (hereinafter, simply referred to as “disks”) 5 are connected to each disk controller 3 which is a fixed disk control device. The disk operation test method according to the present embodiment is implemented by loading an operation test program into the memory 2 and executing the program by the CPU 1.

A feature of the present embodiment is that various parameters can be set as test patterns for testing the disk controller 3 and the disk 5. This allows the disk controller 3 to perform an operation close to the actual use environment of the system, and allows a test to be performed in an overloaded state, so that a defective portion can be detected more reliably and in a short period of time. Can be.

Next, a disk operation test method according to the present embodiment will be described with reference to the flowchart shown in FIG.

In this embodiment, the disk controller 3 and the disk 5 to be tested are tested by executing an operation test program. This embodiment is characterized in that various test contents can be set, so that the test contents,
That is, it is necessary to set the disk access method and test data specifications in advance. Therefore, first, the pattern of the disk access method is determined (step 10).
0). This is determined for each disk 5 to be tested.
In the present embodiment, four types of disk access patterns are assumed as patterns of the disk access method: sequential, random, random for a fixed number of times, and random for an indefinite number of times. A plurality is selected and set. Of course, we are not trying to limit to this pattern.

By the way, in the present embodiment, the test contents for performing the operation test can be set manually as parameters as well as automatically. Automatic is equivalent to the process of setting a disk access pattern using pseudo-random numbers described later, and manual is
This corresponds to a process of setting a disk access pattern using a fixed value or a fixed pattern. The setting by pseudo-random numbers means that sequential / random / random at a fixed number of times / random at an indefinite number of times are determined based on the result of the pseudo-random numbers. Also, in the case of random for every fixed number of times / random for every indefinite number of times, the number of times is also determined by a pseudo random number. Setting with a fixed value means that given parameters are sequential / random / random /
Random for each indefinite number of times is determined. Also, in the case of random for every fixed number of times / random for every indefinite number of times, the number of times is also determined by a given parameter. The setting based on the fixed pattern determines sequential / random / random for a fixed number of times / random for an indefinite number of times with given parameters. Also, in the case of random for every fixed number of times / random for every indefinite number of times, the number of times is also determined by a given parameter. The difference from the setting using the fixed value is that the access pattern is changed depending on the number of times of activation.

The above-mentioned "start" is caused by performing a disk access by activating a test program. However, a plurality of disk accesses are generated by the test program, and a plurality of test programs are executed. In some cases. Also, the disk controller 3
In some cases, disk access is merged under the control of. Therefore, it is appropriate to interpret the term “activation” in this embodiment as “issuing an access request”. Hereinafter, the details of the process of determining the disk access pattern will be described with reference to the flowchart shown in FIG.

First, if parameters set in advance from outside are stored internally, they are read (steps 101 and 102). If the parameter is a fixed value, a disk access pattern is set based on the fixed value (steps 103 and 104). If the parameter set in advance is a fixed pattern, a disk access pattern is set according to the fixed pattern (steps 105 and 106). On the other hand, when the parameter is not set in advance, or when the set parameter is neither a fixed value nor a fixed pattern, a pseudo random number is generated internally, and one of the disk access patterns is selected based on the random value. Set (Step 1
01, 107).

Next, the number of executed disk accesses constituting the disk access pattern is determined (step 1).
10). For example, in the case of the sequential access, since the address is sequentially accessed from the address a to the address (a + m) by one activation, the number of executions in this case is one. The number of executions in the case of accessing two continuous areas from address a to address (a + m) and further from address b to address (b + n) by one activation is two. The number of executions can be set either automatically or manually as in the above disk access pattern. The term “automatic” corresponds to a process of setting the number of executions using a pseudo-random number described later, and the term “manual” corresponds to a process of setting the number of executions using a fixed value or a fixed pattern. Setting by pseudo random numbers means that the number of executions for each disk access is determined based on the result of the pseudo random numbers. Setting with a fixed value determines the number of executions for each disk access with given parameters.
Setting by a fixed pattern determines the number of executions for each disk access by a given parameter. The difference from the setting by the fixed value is that the number of executions is changed depending on the number of times of activation. Hereinafter, the details of the process of determining the number of disk accesses to be executed will be described with reference to the flowchart shown in FIG.

First, if parameters set in advance from outside are stored internally, they are read (steps 111 and 112). If the parameter is a fixed value, the number of executions is set based on the fixed value (steps 113 and 114). If the parameter set in advance is a fixed pattern, the number of executions is set according to the fixed pattern (steps 115 and 116). On the other hand, when the parameter is not set in advance, or when the set parameter is neither a fixed value nor a fixed pattern, a pseudo random number is generated internally, and one of the number of executions is selected and set according to the random number value (Step 11
1,117).

When the disk access method and the number of executions are determined as described above, the address of the disk 5 to be accessed is determined according to the determined disk access method (steps 120 to 140). Then, disk 5
Then, the data length and data pattern of the test data to be written into are determined (step 150). The data length and data pattern can also be set automatically or manually similarly to the disk access pattern. The term “automatic” corresponds to a process of setting a data length or a data pattern by using a pseudo-random number described later, and the term “manual” corresponds to a process of setting a data length or a data pattern by a fixed value or a fixed pattern. Setting the data length by a pseudo random number determines the data length based on the result of the pseudo random number. Setting the data length by a fixed value determines the data length by a given parameter. To set the data length by a fixed pattern, the data length is determined by a given parameter. The difference from the setting by the fixed value is that the data length is changed by the number of times of activation. Setting a data pattern using a pseudo-random number means determining a fixed value (identical value) / identical pattern (constant increment value, worst pattern, etc.) / Data pattern for random number generation based on the result of the pseudo-random number. Setting a data pattern by a fixed value determines a fixed value / identical pattern / random number generation data pattern with given parameters. Setting a data pattern using a fixed pattern involves determining a fixed value / identical pattern / random number generation data pattern using given parameters. The difference from the setting by the fixed value is that the data pattern is changed by the number of times of activation. FIGS. 5 and 6 show flowcharts showing the details of the processing for determining the data length and the data pattern of the test data, respectively, and FIGS. 3 and 5 and FIG. 6 is the same as the process itself except for the content to be set, so the detailed description is omitted.

As described above, when the parameters necessary for the operation test have been set for all the disks to be tested (step 160), the disk 5 is actually set.
An attempt is made to detect a defective portion by reading and writing the test data to the memory (step 170). FIG. 7 shows a flowchart of a process of writing to the disk 5 for a process for accessing the disk and detecting a defective portion.
A description will be given with reference to FIG. 8 showing a flowchart of the reading from the disk 5 and the data comparison processing, and FIG.

In FIG. 7, all the start information to be sent to the disk controller 3 is generated based on the disk access pattern, the number of executions, the data length of the test data and the parameters of the data pattern set in the above processing. They are set in the area in the order of execution (steps 171 and 172). Then, by sending startup information to the disk controller 3 to which the disk 5 to be tested is connected, the disk controller 3 is instructed to execute data writing in accordance with the set parameters (step 173). As a result, an interrupt occurs, and the disk controller 3 writes test data to the specified disk 5 (step 174).
If an error has occurred during the data writing, the fact and information necessary for error analysis such as error status information are recorded (steps 175 and 176).

In FIG. 8, all the start information to be sent to the disk controller 3 is generated based on parameters such as the disk access pattern set in the above processing,
The execution order is set in a predetermined area of the memory 2 (steps 181 and 182). Then, by sending startup information to the disk controller 3 to which the disk 5 to be tested is connected, the disk controller 3 is instructed to execute data reading in accordance with the set parameters (step 183). As a result, an interrupt occurs, and the disk controller 3 reads test data from the specified disk 5 (step 18).
4). If an error occurs during the reading of the data, the fact and information necessary for error analysis such as error status information are recorded (steps 185 and 186).

When the test data is read from the designated address, the read data value is compared with the expected value (step 187). As the expected value, test data held in the memory 2 at the time of data writing may be used.
As a result of the comparison, if the values are not the same, it is determined that an error has occurred, and error information indicating that fact is recorded (step 1).
89). The write and read access timings of the data may be such that the write and the read are alternately performed, or the test data may be continuously written and then the data may be continuously read. This follows the parameter settings.

The above process is repeated for the number of activations set as a parameter (step 180). When the test is completed, a test result is output to report whether an error has occurred during disk access, that is, whether a failure has been detected in the operation test (step 190).

According to the present embodiment, items to be tested in the operation test, that is, how much data is to be written to which address of which disk 5, and how the test data pattern to be written is In addition, it is possible to set the timing and order of disk access. That is, parameters of test contents can be set so that operation tests can be performed in various patterns. Thus, the tester can perform the test in the same pattern as the disk access in the actual system. Also,
It is easy to perform disk access irregularly or in an overloaded state. Since various test environments can be set as described above, a defective portion can be detected more reliably and in a short period of time, so that the efficiency and accuracy of the operation test can be improved.

In the above description, a case has been described as an example where all the changeable items such as the disk access method, the test data pattern and the data length have been changed. Operation tests can be performed by selecting only ones and combining them.

[0027]

According to the present invention, the test contents can be set with parameters so that the operation test can be performed in various patterns. Therefore, the test can be performed in the same pattern as the disk access that can occur in an actual system. it can. Further, it is possible to easily perform an operation test while changing the disk access method and mixing the disk access methods. In addition, an operation test can be easily performed in an overloaded state. As described above, by setting various test environments, a defective portion can be detected more reliably and in a short period of time.

[Brief description of the drawings]

FIG. 1 is a schematic configuration diagram of a computer system in which an embodiment of a disk operation test method according to the present invention is implemented.

FIG. 2 is a flowchart showing a disk operation test method according to the embodiment.

FIG. 3 is a flowchart showing details of a process for determining a disk access pattern in the present embodiment.

FIG. 4 is a flowchart showing details of processing for determining the number of executions for each disk access in the present embodiment.

FIG. 5 is a flowchart showing details of processing for determining a test data length in the present embodiment.

FIG. 6 is a flowchart showing details of processing for determining a test data pattern in the present embodiment.

FIG. 7 is a flowchart showing details of a writing process to a disc in the present embodiment.

FIG. 8 shows reading from a disk according to the embodiment;
9 is a flowchart illustrating details of a comparison process.

FIG. 9 is a conceptual diagram showing the contents of a disk access in the present embodiment.

[Explanation of symbols]

1 CPU, 2 memories, 3 disk controllers (fixed disk controller), 4 buses, 5 disks (fixed disk device).

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat ゛ (Reference) G11B 20/18 501 G11B 20/18 501F 572 572F

Claims (3)

[Claims] In a disk operation test method for performing an operation test including a fixed disk device alone or a fixed disk controller, an operation test is performed by accessing a fixed disk device to be tested by mixing disk access methods. A disk operation test method. 2. A disk operation test method for performing an operation test including a single fixed disk device or a fixed disk controller, wherein at least a disk access method, a test data pattern and a data are included in an operation test including a series of disk accesses. A disk operation test method, wherein test contents for changing any of the lengths irregularly can be set. 3. The disk operation test method according to claim 2, wherein operation tests on a plurality of fixed disk devices are performed simultaneously in parallel.