JPH0244578A - Trouble processing system - Google Patents

Abstract

PURPOSE:To improve the reliability of information, to shorten a re-test processing time, to reduce the load of a CPU and to reduce the burden of a maintainer to defect alternate processing, etc., by directly executing the re-test processing and defect alternate processing to a magnetic disk with a data transfer control device. CONSTITUTION:The information are written through a data transfer control device 2 to a duplexed disk 3 by writing indication from a CPU 1 and the information are read from one side of the disk 3 in correspondence to reading indication. In this device 2, a writing re-testing means 101, a reading re-testing means 102 and a defect alternate means 200, etc., are provided and the writing re-testing and reading re-testing processing to the disk 3 is directly and easily executed without being passed through the CPU 1. Further, the alternation of the disk 3, which fails re-testing, to an alternate area is directly and easily executed by the device 2 through the means 200 automatically. As a result, the reliability of the information is improved and the re-test processing time is shortened. Then, the load of the CPU is reduced and further, the burden of the maintainer is reduced to the defect alternate processing.

Description

[Detailed Description of the Invention] [Summary] Regarding a failure handling method in a data transfer control device that stores and reads information from a duplicated magnetic disk, the present invention provides a method for shortening the time required for retry processing for magnetic disks, and also improving retry processing. With the aim of reducing the load on the central processing unit required for the In a data transfer control device that writes the same information to a disk and reads information from one of the duplicated magnetic disks based on a read instruction transmitted from the central processing unit, writing of information to the magnetic disk is unsuccessful. a write retry means for retrying writing to the same magnetic disk when the information is completed, and retrying reading from the other magnetic disk when reading information from one magnetic disk is unsuccessful. A storage area of the magnetic disk where the retry by the write retry means was unsuccessful, and a storage area of the magnetic disk where the reading was unsuccessful are provided in the magnetic disk. It is configured to include a defect replacement means for replacing the spare storage area.

[Industrial application field]

The present invention stores information on duplicated magnetic disks,
The present invention relates to a failure handling method in a data transfer control device for reading.

In order to improve the reliability of information stored on magnetic disks, the same information is written on duplicated magnetic disks, and even if one magnetic disk becomes damaged, normal information can be read from the other magnetic disk. Methods that enable this are widely adopted.

In addition, a spare storage area is prepared in advance on the magnetic disk,
A so-called failure replacement process is also widely adopted in which when a part of the normally used storage area becomes damaged, a spare storage area is used to perform writing and reading from the affected storage area.

[Conventional technology]

FIG. 8 is a diagram showing an example of a conventional information processing system, FIG. 9 is a diagram showing an example of the write execution process in FIG. 8, and FIG. 1O is an example of the read execution process in FIG. FIG.

In FIGS. 8 and 9, in order to write the same information to the duplicated magnetic disks 3-0 and 3-1, the central processing unit 1 first instructs the O-system write start to the O-magneto-optical disk 3-0. a8° is transmitted to the data transfer control device 2.

In the data transfer control device 2, the disk control unit 22
O system write start instruction a8 transmitted from the central processing unit 1
When it receives ゜, it sends an O-system write start signal to the 0-system magnetic disk 3-0. . Communicate.

0 series write start signal. . 0 series magnetic disk 3 that received
If a failure occurs in -0 and writing is not completed normally, the 0-system magnetic disk 3-0
An O-system write failure response signal C1 indicating that the write was unsuccessful is returned to the write failure response signal C1.

Upon receiving the 0-system write failure response signal C-0, the disk control unit 22 sends an O-system write failure report d1 to the central processing unit 1.
Return ゜.

The central processing unit 1 that received the O-system write failure report d8゜,
A 0-system write restart instruction a that starts the retry processing unit 11 and causes it to retry writing to the 0-system magnetic disk 3-0. .

“ is transmitted to the data transfer control device 2.

0 system write restart instruction a. . ° Received data transfer system?
The disk control unit 22 of the II unit W2 transmits a 0-system write restart signal S,4' to the 0-system magnetic disk 3-0.

The failure that occurred in the 0-system magnetic disk 3-0 is a temporary failure, and this time, when the 0-system magnetic disk 3-0 completes writing normally, the 0-system magnetic disk 3-0 will be transferred to the disk controller 22. An O-system write normal response signal e indicating that writing has been successfully completed. . to be sent back.

Upon receiving the 0-system write normal response signal e4°, the disk control unit 22 sends a 0-system write normal report f8 to the central processing unit 1.
Return ゜.

The central processing unit l that received the 0-system write normal report r8゜,
In order to write the same information to the 1-system magnetic disk 3-1, an I-system write activation instruction aWI is transmitted to the disk control unit 22.

The disk control unit 22 of the data transfer control device 2, which has received the l-system write activation instruction aWI, transmits an l-system write activation signal □ to the l-system magnetic disk 3-1.

1-system magnetic disk 3- that received the 1-system write activation signal b□
1, when writing is completed normally, the disk control unit 22
A 1-system write normal response signal e1 is returned to the 1-system write normal response signal e1.

The disk control unit 22, which has received the 1-system write normal response signal eWI, sends an 1-system write normal report f to the central processing unit l.

Return 1.

■System write normal report f. 1, the central processing unit 1 receives
O magneto-optical disk 3-0 and 1-series magnetic disk 3-1
It is determined that writing of the same information has been completed.

Note that the data storage memory 21 provided in the data transfer control device 2 temporarily stores write information transmitted from the central processing unit 1 or read information transmitted from the magnetic disk 3.

Next, in FIGS. 8 and 10, the central processing unit 1
However, in order to read the same information stored on the dual magnetic disks 3-0 and 31, first the 0-series magnetic disk 3-0 is read out.
0 system read start instruction a for 0. is transmitted to the data transfer control device 2.

In the data transfer control device 2, the disk control unit 22
is the 0-system read start instruction a7 transmitted from the central processing unit 1.
Upon receiving ゜, it transmits an O-system read activation signal bjlo to the 0-system magnetic disk 3-0.

0-series magnetic disk 3 that received the 0-series read activation signal bR6
If a failure occurs in the 0-system magnetic disk 3-0 and reading is not completed normally, the 0-system magnetic disk 3-0
A 0-system read failure response signal C1° indicating that the read has ended unsuccessfully is returned to the 0-system read failure response signal C1°.

Upon receiving the 0-system read failure response signal Cll0, the disk control unit 23 sends a 0-system read failure report d to the central processing unit 1.
Return the RO.

The central processing unit 1 that received the 0-system read failure report dRO,
The retry processing unit 11 is started and the 1st system read start instructions a and 11 for the 1st system magnetic disk 3 - 1 are transmitted to the data transfer control device 2 .

The disk control unit 22 of the data transfer control device 2, which has received the 1-system read activation instruction aRI, transmits the 1-system read activation signal bRI to the 1-system magnetic disk 3-1.

1-system magnetic disk 3 that received the 1-system read activation signal bR1
-1, when the reading is completed normally, the disk controller 2
2, a 1-system read normal response signal e7 is returned.

Upon receiving the 1-system read normal response signal e□, the disk control unit 22 sends a 1-system read normal report f+ to the central processing unit 1.
Return t+.

The central processing unit 1 that received the system 1 read normal report fR1,
It is determined that reading of the required information from the 1-system magnetic disk 3-1 has been completed.

In addition, in FIGS. 8 and 9, the disk controller 2
2 transmits the 0 system write restart signal bW (1°) to the O magneto-optical disk 3-0, the 0 system magnetic disk 3-0
The failure that occurred at 0 is a fixed failure, and this time also
If the magneto-optical disk 3-0 does not complete writing normally, the 0-system magnetic disk 3-0 returns the O-system write failure response signal Cwo and sends the 0-system write failure response signal C1゜. The disk control unit 22 that received the data also sends the O-system write failure report dw to the central processing unit 1 again. to be sent back.

0 system write failure report d again. . Central processing unit 1 that received
outputs to the console 4 off-site information indicating that the θ-system write failure report d,. has been received again in response to the 0-system write restart instruction a9゜.

The maintenance person analyzes the fault information output to the console 4 and determines that part of the storage area of the magnetic disk 3 has a fixed fault.
If it is determined that the O magneto-optical disk 3-0 is unusable, the defective replacement processing unit 12 is activated by inputting the required command from the console 4, and the write and write operations are performed on the storage area of the O magneto-optical disk 3-0 that is unusable. A defect replacement process is executed in which reading is performed on behalf of a replacement area selected from a spare storage area provided in advance in the O magneto-optical disk 3-0.

Similar defect replacement processing is shown in FIGS. 8 and 10 when the 0-system read activation signal b9° is received from the disk control unit 22, the read cannot be completed normally, and the 0-system read failure response signal CMII is sent. Returned 0 series magnetic disk 3-0
It is also executed for

[Problem to be solved by the invention]

As is clear from the above description, in a conventional information processing system, retry processing when writing to or reading from the magnetic disk 3 is unsuccessful is carried out by the retry processing unit 11 provided in the central processing unit 1. Therefore, the retry process takes time and the load on the central processing unit 1 increases, which may put pressure on the processing capacity of the central processing unit 1.Furthermore, a fixed failure may occur in the magnetic disk 3, causing the storage There is a problem in that determining whether a part of an area is unusable and processing defective replacement for an unusable area requires the intervention of a maintenance person.

The present invention aims to shorten the time required for retry processing for magnetic disks, reduce the load on a central processing unit required for retry processing, and further reduce as much as possible the burden on maintenance personnel associated with defective replacement processing for magnetic disks. With the goal.

[Means to solve the problem]

FIG. 1 is a diagram showing the principle of the present invention.

In FIG. 1, 1 is a central processing unit, 2 is a data transfer control unit, and 3 is a duplicated magnetic disk.

101 is a write retry means provided in the data transfer control device 2 according to the present invention.

102 is a read retry means provided in the data transfer control device 2 according to the present invention.

Reference numeral 200 denotes a defect replacement means provided in the data transfer control device 2 according to the present invention.

[Effect]

The data transfer control device 2 writes the same information to the duplicated magnetic disk 3 based on the write instruction transmitted from the central processing unit 1, and writes the same information to the duplicated magnetic disk 3 based on the read instruction transmitted from the central processing unit 1. Information is read from either one of the magnetic disks 3.

The write retry unit 101 retries writing to the same magnetic disk 3 when writing information to the magnetic disk 3 is unsuccessful.

The read retry means 102 retries reading from the other magnetic disk 3 when reading information from one magnetic disk 3 is unsuccessful.

The defect replacement means 200 provides in the magnetic disk 3 a storage area of the magnetic disk 3 whose retry by the write retry means 100 was unsuccessful, and a storage area of the magnetic disk 3 whose readout was unsuccessful. the reserved storage area.

Therefore, the data transfer control device autonomously executes retry processing and defective replacement processing for magnetic disks directly on the magnetic disks, improving the reliability of information and reducing the time required for retry processing. In addition, the load on the central processing unit due to retry processing is reduced, and the burden on maintenance staff regarding defective replacement processing is also reduced.

〔Example〕

An embodiment of the present invention will be described below with reference to the drawings.

FIG. 2 is a diagram showing an information processing system according to an embodiment of the present invention, FIG. 3 is a diagram illustrating the write process in FIG. 2, and FIG. 4 is a diagram showing the write execution in FIG. FIG. 5 is a diagram illustrating a write execution process accompanied by the defect replacement process in FIG. 2; FIG. 6 is a diagram illustrating a read process in FIG. 2; 7th
The figure is a diagram illustrating the read execution process in FIG. 2.

Note that the same reference numerals indicate the same objects throughout the figures.

In FIG. 2, the write retry means 10 in FIG.
1, the write retry processing section 231 is used as the read retry processing section 231 as the read retry means 102 in FIG.
are provided in the disk control unit 23 in the data transfer control device 2, respectively, and the defect replacement means 20 in FIG.
0, the defective replacement processing unit 233 transfers data to the data transfer control device 2.
The central processing unit 1 is not provided with a retry processing section II and a defective replacement processing section 12 as shown in FIG.

In FIGS. 2, 3, and 4, the central processing unit 1
transmits a write start instruction a8 to the data transfer control device 2 in order to write the same information on the duplicated magnetic disks 3-0 and 3-1.

In the data transfer control device 2, the disk control unit 23
is the write activation instruction a transmitted from the central processing unit 1. When it receives (step 511 in FIG. 3), it first sends a 0-system write activation signal to the 0-system magnetic disk 3-0. . is transmitted (step 512).

0 series write start signal. . 0 series magnetic disk 3 that received
If a failure occurs in the O magneto-optical disk 3-0 and writing is not completed normally, the O magneto-optical disk 3-0
A 0-system write failure response signal CWO is returned to the 0-system write failure response signal CWO.

Upon receiving the 0-system write failure response signal CWO (step 313), the disk control unit 23 activates the write retry processing unit 231.

The activated write retry processing unit 231 sends a 0 system write restart signal in order to retry writing to the 0 system magnetic disk 3-0. ” (step 515).

The failure that occurred in the 0-system magnetic disk 3-0 is a temporary failure, and this time, when the O-magneto-optical disk 30 completes writing normally, a 0-system write normal response signal is sent to the disk control unit 23. e. . to be sent back.

Having received the θ system write normal response signal e8° (step 516), the disk control unit 23 writes 1 to the 1 system magnetic disk 3-1 in order to write the same information to the 1 system magnetic disk 3-1. Transmit system write activation signal bWl (step 51
4).

■When the 1st system magnetic disk 3-1 receives the 1st system write activation signal S11, the disk control unit 2
3, the 1-system write normal response signal el, 11 is returned.

1 system write normal response signal e, 1. The disk control unit 23 that has received this (step 5112) sends the central processing unit 1 to the 0-system magnetic disk 3-0 and the 1-system magnetic disk 3-0.
Write normal report f indicating that writing of the same information to No. 1 has been completed. is sent back (step 5114).

Next, in FIGS. 2, 3, and 5, the disk control unit 23 of the data transfer control device 2 executes steps Sll to S15 in FIG.
As a result of transmitting the 0-system write restart signal b1゜' to the 0-system magnetic disk 3-0 (step 5I5), it was determined that the failure that had occurred in the 0-system magnetic disk 3-0 was a fixed failure, and that writing was successful this time as well. If it cannot be executed, 0 series magnetic disk 3-0
is a 0-system write failure response signal C to the disk control unit 23.
80 will be sent back again.

The disk control unit 23, which has received the 0-system write failure response signal C80 again (step 316), writes the O-magneto-optical disk 3-
It is determined that a fixed failure has occurred in 0, and the defective replacement processing unit 233 is activated.

The activated defective replacement processing unit 233 performs a known defective replacement process on the 0-series magnetic disk 3-0 (step 317), and then performs a predetermined process to perform a write/read test (step 518). The O system write/read test signal g0 is transmitted to the 0 system magnetic disk 3-0, and the O system write/read test signal g0 is transmitted to the O system magnetic disk 3-0.
0 to O system test normal completion signal. is returned (step 519), it is determined that the defective replacement process has been successfully completed, and the write retry processing unit 231 is activated.

The activated write retry processing unit 231 transmits a 0 system write retry activation signal S,1゜'' in order to retry writing to the 0 system magnetic disk 3-0 (step 5il).
o).

The 0-system magnetic disk 3-0, which has received the 0-system write re-activation signal b8゜'', returns a 0-system write normal response signal ew to the disk control unit 23 when the writing is completed normally.

Upon receiving the 0-system write normal response signal e8°, the disk control unit 23 executes step 514 and subsequent steps in order to write the same information to the 1-system magnetic disk 31 as described above.

Next, in FIG. 2, FIG. 6, and FIG. A read start instruction aR is transmitted to the host.

In the data transfer control device 2, the disk control unit 23
When it receives the read start instruction aR transmitted from the central processing unit 1 (step S21 in FIG. 6), it first sends an O system read start signal to the 0 system magnetic disk 3-0. (step 522).

O magneto-optical disk 3 that received the 0-system read activation signal bRo
If a failure occurs in -0 and reading is not completed normally, the 0-system magnetic disk 3-0
An O-system read failure response signal CRO is returned to the O-system read failure response signal CRO.

Upon receiving the 0-system read failure response signal C3° (step 523), the disk control unit 23 activates the read retry processing unit 232.

The activated read retry processing unit 232 transmits the 1-system read activation signal bR1 to the 1-system magnetic disk 3-1 in order to read the same information from the 1-system magnetic disk 3-1 (step 525). .

1-system magnetic disk 3 that received the I-system read activation signal bR1
-1, when the reading is completed normally, the disk controller 2
3, the system 1 read normal response signal eR1 is returned.

The disk control unit 23, which has received the 1-system read normal response signal "R1" (step 526), returns a read normal report f7 indicating that the read based on the read start instruction aR has been completed to the central processing unit 1 (step 526). 527).

Next, in order to restore the affected O-magneto-optical disk 3-0 to a normal state, the disk control unit 23 uses the defective replacement processing unit 23.
Start 3.

The activated defect replacement processing unit 233 performs a known defect replacement process on the 0-system magnetic disk 3-0 (step 528), and then performs a predetermined O-system write/read test in order to perform a write/read test. In order to transmit the test signal g0 to the O magneto-optical disk 3-0 (step 529), the test signal g0 is transferred from the O-series magnetic disk 3-0 to the O
When the system test normal end signal h0 is returned (step 52
10) It is determined that the defect replacement process has been completed normally, and the write retry processing unit 231 is activated.

The activated write retry processing unit 231 writes the 1-system magnetic disk 3-0 to the O magneto-optical disk 3-0 that has undergone the defect replacement process.
In order to store the same information as 1, a 0-system write activation signal b8° is transmitted (step 5211).

When the O magneto-optical disk 3-0 that has received the 0-system write start signal S, 1° completes writing normally (step 521
2) Send a 0-system write normal response signal to the disk control unit 23. to be sent back.

As described above, the disk control unit 23 controls the O magneto-optical disk 3.
The storage of the same information on the -0 and 1-system magnetic disks 3-1 is completed.

As is clear from the above description, according to the present embodiment, the write start instruction a8 or the read start instruction a is sent from the central processing unit 1.
The disk control unit 23 in the data transfer control device 2 that has received the request attempts writing or reading, and if it ends in failure, the write retry processing unit 231 or the read retry processing unit 2
32 autonomously executes the retry process, and if it is determined that a fixed failure has occurred, the defective replacement processing unit 233 autonomously executes the defective replacement process, so the central processing unit 1 has no retry processing. There is no need to execute processing or defective replacement processing, and there is no need to place a burden on maintenance personnel to determine whether or not defective replacement processing is necessary and to execute it.

Note that FIGS. 2 to 7 are only one embodiment of the present invention, and for example, FIGS. 4 and 5 show the process in which the l-system magnetic disk 3-1 successfully completes writing. However, if writing to the 1-system magnetic disk 3-1 has not been completed normally, the disk control unit 23 activates the write retry processing unit 231 and, if necessary, the defective replacement processing unit 232, and The processes corresponding to steps 315 to 5111 in FIG. 3 that were executed for the magnetic disk 3-0 are also executed for the 1-system magnetic disk 3-1 (step 31 in FIG. 3).
13). In addition, in FIG. 7, 1-system magnetic disk 3-1
A failure has also occurred in the 1st system read normal response signal “Il”.
If l is not returned, the disk control unit 23 returns an abnormal termination report to the central processing unit 1 indicating that the read based on the read start instruction aB ended unsuccessfully (step 5213). Although many other modifications may be considered in addition to those described above, the effects of the present invention remain the same in any case.

〔Effect of the invention〕

As described above, according to the present invention, in the data transfer control device, the retry process and the defective replacement process for the magnetic disk are autonomously executed directly on the magnetic disk, and the information is reliable. In addition to improving performance, the time required for retry processing is also reduced, the load on the central processing unit associated with retry processing is also reduced, and the burden on maintenance personnel regarding defective replacement processing is also reduced.

[Brief explanation of the drawing]

FIG. 1 is a diagram showing the principle of the present invention, FIG. 2 is a diagram showing an information processing system according to an embodiment of the present invention, FIG. 3 is a diagram illustrating the write processing process in FIG. 2, and FIG. is a diagram illustrating the write execution process in FIG. 2, FIG. 5 is a diagram illustrating the write execution process accompanied by defect replacement processing in FIG. 2, and FIG. 6 is a diagram illustrating the read process in FIG. 2. Figure, 7th
8 is a diagram illustrating an example of the read execution process in FIG. 2, FIG. 8 is a diagram illustrating an example of a conventional information processing system, FIG. 9 is a diagram illustrating an example of the write execution process in FIG. FIG. 10 is a diagram showing an example of the read execution process in FIG. 8. In the figure, 1 is a central processing unit, 2 is a data transfer control unit, 3 is a magnetic disk, 4 is a console, 11 is a retry processing unit, 12 and 233 are defective replacement processing units, 21 is a data buffer memory, 22 and 23 101 is a write retry means, 102 is a read retry means,
200 is a defective replacement means, 231 is a write retry processing unit, 2
32 is a reading of the principles of the opening month. i:, to +731 ru and stop Oshimine 4 Figure 2 Figure (-); + υ Ziyu 3 to generation 5 Go 7. Ri and Hakunaten 1
Shigome Tora f-E Jijima De Toyomine 5 2 Hands 2 Diagrams (Niba (Japanese vomit beast and 7 shots 1000)

Claims (1)

[Claims] Based on a write instruction transmitted from the central processing unit (1),
The same information is written on the duplicated magnetic disks (3), and the information is read from either one of the duplicated magnetic disks (3) based on a read instruction transmitted from the central processing unit (1). In the data transfer control device (2), if writing of the information to the magnetic disk (3) is unsuccessful, a write retry means (101) retries writing to the same magnetic disk (3). ), and if the reading of the information from one of the magnetic disks (3) is unsuccessful, the other magnetic disk (3)
3) read retry means (102) for retrying read from
), the storage area of the magnetic disk (3) where the retry by the write retry means (101) ended unsuccessfully, and the storage area of the magnetic disk (3) where the readout ended unsuccessfully.
) with a spare storage area provided in the magnetic disk (3);
A failure handling method characterized by providing.