JP5288331B2 - I/O command failure recovery circuit, I/O command failure recovery method, and I/O command failure recovery program - Google Patents

Description

The present invention relates to timeout detection when a minor error such as a CRC (Cyclic Redundancy Check) error occurs in data communication at a host-disk device interface.

There is a technology that can detect minor errors, such as CRC errors, that occur in data communication at the host-disk device interface.

For example, in the technology described in Patent Document 1, when a host computer (CPU) incorporates a peripheral device, it turns on an event notification flag indicating that for commands whose execution time is longer than the I/O monitoring time given by the CPU, the host computer notifies the CPU of the respective extension time and the extension of the monitoring time. Then, when the CPU issues an I/O, if the command notifies an extension of the monitoring time, the peripheral device notifies the CPU of the extension time, and upon receiving the notification of the monitoring time extension, the CPU adds a value equivalent to the notified extension time to the counter value of the monitoring time counter. Also, if the peripheral device finishes executing a command within the extension time at the time it finishes, it sends a notification event of the shortened monitoring time, and the CPU subtracts a value equivalent to the notified shortened time from the time monitoring count value.

JP 2001-147866 A

However, with the technology described in Patent Document 1 above, if a minor error such as a CRC error occurs in data communication at the host-disk device interface, the received information is undefined and it is not possible to determine what kind of instruction or response it is. In such cases, the port that detects the minor error typically simply counts the occurrence of the error and does not actively handle the error.

The abnormality that a fault has been detected is detected by timeout monitoring implemented by the upper layer, and fault processing is then initiated. As a result, an I/O command that would normally be completed in less than 1 second can take several to tens of seconds to complete. As a result, the entire system is delayed. In this case, considering that an internal fault in the disk device may require a response time that is two to three orders of magnitude longer than normal, it is not possible to simply shorten the timeout value of the upper layer.

The present invention aims to provide an I/O command fault recovery circuit, an I/O command fault recovery method, and an I/O command fault recovery program that can reduce processing delays caused by timeout detection when a minor fault such as a CRC error occurs in the host-disk device interface.

According to a first aspect of the present invention, there is provided an I/O command failure recovery circuit incorporated in a host connected to a disk device via an interface, comprising: a counter for counting the number of failures when a failure is detected in the host; a first time monitoring means for time-monitoring a response when the host issues an I/O command to the disk device, and notifying an input/output processing means of the fact that there is no response within a first certain time period; a second time monitoring means for time-monitoring a response when the host issues an I/O command to the disk device, and notifying the input/output processing means of the fact that there is no response within a second certain time period that is shorter than the first certain time period; and input/output processing means for starting fault processing when a notification is received from the second time monitoring means and when a notification is received from the second time monitoring means and the counter has counted up, wherein when a notification is received from the second time monitoring means and the counter has not counted up, the input/output processing means inquires of the disk device as to whether or not the issued I/O processing is being performed internally, and when a reply is received indicating that the I/O processing is not being performed internally of the disk device, it determines that a fault has occurred in the interface and performs the fault processing for the fault in the interface .
According to a second aspect of the present invention, there is provided an I/O command failure recovery circuit incorporated in a host connected to a disk device via an interface, comprising: a first counter for counting the number of failures when a failure is detected in the host; first time monitoring means for time-monitoring a response when the host issues an I/O command to the disk device, and notifying an input/output processing means of the fact that there is no response within a first certain time period; second time monitoring means for time-monitoring a response when the host issues an I/O command to the disk device, and notifying the input/output processing means of the fact that there is no response within a second certain time period which is shorter than the first certain time period; and a second time monitoring means for monitoring a response when the host issues an I/O command to the disk device, and notifying the input/output processing means of the fact that there is no response within a second certain time period which is shorter than the first certain time period. an input/output processing means for initiating fault processing when a notification is received from the monitoring means and the first counter has counted up, wherein when a notification is received from the second time monitoring means and the first counter provided in the I/O instruction fault recovery circuit has not counted up, the input/output processing means inquires whether a second counter provided in the disk device, which counts the number of times a fault has occurred when a fault is detected in the disk device, has counted up, and when a reply is received indicating that a count up has occurred, the input/output processing means determines that this is due to the detection of the fault that counted up, and performs the fault processing for the fault at the count up.

Furthermore, according to a third aspect of the present invention, there is provided an I/O command failure recovery method performed by an I/O command failure recovery circuit incorporated in a host connected to a disk device via an interface, comprising: a step of preparing a counter for counting the number of failures when a failure is detected in the host; a first time monitoring step of time-monitoring a response when the host issues an I/O command to the disk device, and notifying an input/output processing step of the fact that there is no response within a first certain time; a second time monitoring step of time-monitoring a response when the host issues an I/O command to the disk device, and notifying the input/output processing step of the fact that there is no response within a second certain time which is shorter than the first certain time; an input/output processing step for initiating fault processing when a notification is received in the first time monitoring step and when a notification is received in the second time monitoring step and the counter has counted up, wherein the input/output processing step, when a notification is received in the second time monitoring step and the counter has not counted up, queries the disk device as to whether or not the issued I/O processing is being performed internally, and when a reply is received indicating that the I/O processing is not being performed internally in the disk device, determines that a fault has occurred in the interface, and performs the fault processing for the fault in the interface .
Furthermore, according to a fourth aspect of the present invention, there is provided an I/O command failure recovery method performed by an I/O command failure recovery circuit incorporated in a host connected to a disk device via an interface, comprising the steps of: preparing a first counter for counting the number of failures when a failure is detected in the host; a first time monitoring step for monitoring a response to an I/O command issued by the host to the disk device over time, and notifying an input/output processing step of the fact that there is no response within a first certain time; a second time monitoring step for monitoring a response to an I/O command issued by the host to the disk device over time, and notifying the input/output processing step of the fact that there is no response within a second certain time which is shorter than the first certain time; and and an input/output processing step for starting fault processing when a notification is received in the second time monitoring step and the first counter has counted up, wherein the input/output processing step, when a notification is received in the second time monitoring step and the first counter provided in the I/O instruction fault recovery circuit has not counted up, queries whether a second counter provided in the disk device, which counts the number of times a fault is detected in the disk device, has counted up, and when a reply is received indicating that a count up has occurred, determines that the count up is due to the fault detection, and performs the fault processing for the fault at the count up.

Furthermore, according to a fifth aspect of the present invention, there is provided an I/O command failure recovery program installed in an I/O command failure recovery circuit incorporated in a host connected to a disk device via an interface, comprising: a counter for counting the number of failures when a failure is detected in the host; a first time monitoring means for time-monitoring a response when the host issues an I/O command to the disk device, and notifying an input/output processing means of the fact that there is no response within a first certain time period; a second time monitoring means for time-monitoring a response when the host issues an I/O command to the disk device, and notifying the input/output processing means of the fact that there is no response within a second certain time period that is shorter than the first certain time period; and and input/output processing means for starting fault processing when a notification is received from the second time monitoring means and when a notification is received from the second time monitoring means and the counter has counted up, wherein when a notification is received from the second time monitoring means and the counter has not counted up, the input/output processing means inquires of the disk device as to whether or not the issued I/O processing is being performed internally, and when a reply is received indicating that the I/O processing is not being performed internally of the disk device, it determines that a fault has occurred in the interface, and causes the computer to function as an I/O instruction fault recovery circuit that performs the fault processing for the fault in the interface .
Furthermore, according to a sixth aspect of the present invention, there is provided an I/O command failure recovery program to be mounted on an I/O command failure recovery circuit incorporated in a host connected to a disk device via an interface, the program comprising: a first counter for counting the number of failures when a failure is detected in the host; first time monitoring means for time-monitoring a response when the host issues an I/O command to the disk device, and notifying an input/output processing means of the fact that there is no response within a first certain time period; second time monitoring means for time-monitoring a response when the host issues an I/O command to the disk device, and notifying the input/output processing means of the fact that there is no response within a second certain time period which is shorter than the first certain time period; and and input/output processing means for starting fault processing when a notification is received from the second time monitoring means and the first counter provided in the I/O instruction fault recovery circuit has counted up, wherein when the input/output processing means receives a notification from the second time monitoring means and the first counter provided in the I/O instruction fault recovery circuit has not counted up, the input/output processing means inquires whether a second counter provided in the disk device, which counts the number of times a fault has occurred when a fault is detected in the disk device, has counted up, and when a reply is notified that a count up has occurred, determines that the count up is due to the fault detection, and causes the computer to function as an I/O instruction fault recovery circuit that performs the fault processing for the fault at the count up.

According to the present invention, when a minor fault such as a CRC error occurs in the host-disk device interface, it is possible to detect the abnormality after a certain period of time has elapsed and start fault processing, so that fault processing for the I/O command that has occurred can be started earlier, and delays in the entire system can be prevented.

FIG. 1 is a diagram illustrating a basic configuration of an embodiment of the present invention. FIG. 1 is a diagram (1/2) showing a basic operation of an embodiment of the present invention. FIG. 2 is a diagram (2/2) showing the basic operation of an embodiment of the present invention. 1 is a flowchart (1/2) for explaining the effects of the embodiment of the present invention. 1 is a flowchart (2/2) for explaining the effects of the embodiment of the present invention.

Next, an embodiment of the present invention will be described in detail with reference to the drawings. Referring to FIG. 1, this embodiment has a host 100 and a disk device 200. The host 100 and the disk device 200 are connected via a serial interface 300.

The serial interface 300 is a serial interface such as a Fibre Channel interface.

The host 100 also has a central processing unit 110, a main memory unit 120, an input/output processing unit 130, a first time monitoring mechanism 141, a second time monitoring mechanism 142, a CRC error counter 150, and a transmission/reception circuit 160.

The central processing unit 110 is, for example, a CPU, and performs arithmetic processing in the host 100. The main memory device 120 is a memory device that can be directly accessed by the central processing unit 110.

The input/output processing unit 130 is a device that reads and writes data from and to the disk device 200 in response to I/O commands from the central processing unit 110.

The I/O processing device 130 also has a first time monitoring mechanism 141 and a second time monitoring mechanism 142 that monitor the time of responses from the disk device to detect I/O command abnormalities. In addition, the I/O processing device 130 has a CRC error counter 150 that stores the number of serial interface failures, such as CRC errors, detected when the transmission and reception circuits of the host 100 and the disk device 200 transmit information.

The transmission/reception circuit 160 is a transmission/reception circuit that transmits information to the disk device 200 via the serial interface 300 in response to instructions from the input/output processing device 130.

On the other hand, the disk device 200 has a transmission/reception circuit 210, a CRC error counter 220, a disk 230, and a disk control device 240.

The transmission/reception circuit 210 is a circuit that transmits information to the host 100 via the serial interface 300. The CRC error counter 220, like the CRC error counter 150, is a counter that stores the number of serial interface failures, such as CRC errors, detected when the transmission/reception circuits of the host 100 and the disk device 200 transmit information.

Disk 230 is a disk that stores data.

The disk control device 240 is a device that analyzes I/O commands from the host using information from the transmission/reception circuit 210 and reads and writes data to the disk 230.

Next, the operation of this embodiment will be explained with reference to the flowchart in Figure 2.

First, we will explain the operation when the central processing unit 110 in the host 100 issues an I/O command to the input/output processing unit 110 to execute data transfer with the disk device 200.

When the central processing unit 110 instructs the input/output processing unit 110 to execute an I/O command, the input/output processing unit 110 accepts the instruction (step S401).

The I/O processing device 110 then reads the CRC error counter 150 before starting the I/O command and stores it internally (step S402).

Then, the transmission/reception circuit 160 is instructed to send an I/O command start to the disk device 200. Upon receiving this instruction, the transmission/reception circuit 160 transmits the I/O command start to the disk device 200 via the serial interface 300 (step S403). If the transmission/reception circuit 160 detects a serial interface failure such as a CRC error when transmitting information via the serial interface 300, it counts up the CRC error counter 150.

The transmission/reception circuit 210 inside the disk device 200 receives the start of an I/O command from the host 100 via the serial interface 300. The transmission/reception circuit 210 then notifies the disk control device 240 of this (step S404).

The disk control device 240 that has received the notification analyzes the I/O command received from the host 100 and executes the reading and writing of data from and to the disk 230. When the execution of the I/O command is completed, the disk control device 240 instructs the transmission/reception circuit 210 to send an I/O command response to the host 100 (step S405). At this time, if the transmission/reception circuit 210 detects a serial interface failure such as a CRC error while transmitting information via the serial interface 300, it counts up the CRC error counter 220.

In order to monitor the time until an I/O command response is returned in order to detect an abnormality in the disk device 200, the I/O processing device 110 instructs the first time monitoring mechanism 141 to monitor for a time sufficient for the disk device 200 to execute an I/O command (hereinafter, this time will be referred to as "time (A)") (step S406). Furthermore, the I/O processing device 110 instructs the second time monitoring mechanism 142 to monitor for a time shorter than the aforementioned time (A) and sufficient for a normal I/O command response to be returned (hereinafter, this time will be referred to as "time (B)") (step S407).

Then, the input/output processing device 110 continues to check whether an "I/O command response from the disk device 200" has occurred from the transmission/reception circuit 160, whether a "first time monitoring over" has occurred from the first time monitoring mechanism 141, or whether a "second time monitoring over" has occurred from the second time monitoring mechanism 142.

When an I/O command with the disk device 200 is completed normally (Yes in step S408), the I/O processing device 110 recognizes an "I/O command response from the disk device 200" via the transmission/reception circuit 160, and instructs the first time monitoring mechanism 141 and the second time monitoring mechanism 142 to stop time monitoring. Then, in response to this instruction, the first time monitoring mechanism 141 and the second time monitoring mechanism 142 stop time monitoring (steps S409 and S410).

Finally, the I/O processor 110 reports to the central processor 110 that the I/O command has been completed successfully.

On the other hand (No in step S408), if the first time monitoring mechanism 141 detects that the "first time monitoring has expired," i.e., if there is no I/O command response from the disk device 200 even after a certain period of time (A) has elapsed (Yes in step S412), the I/O processing device 110 determines that the I/O command is abnormal and initiates fault processing (step S415).

If the answer is No in step S412 and the second time monitoring mechanism 142 detects that the "second time monitoring has expired", i.e., the time (B) has elapsed but there has been no I/O command response from the disk device 200 (Yes in step S413), the I/O processing device 110 reads the CRC error counter and compares it with the value read before the I/O command was started (step S413).

Then, based on the result of the comparison, it is confirmed whether a CRC error or the like occurred during the execution of the I/O command. If a CRC error or the like occurred (Yes in step S414), it is determined that there has been no response to the I/O command even after a certain period of time (B) has elapsed due to this fault, and fault processing is initiated (step S415).

In addition, if the second time monitoring has not been exceeded (No in step S412) and if a CRC error or the like has not occurred (No in step S404), it is determined that the response is delayed for some other reason, and it continues to check again whether an "I/O command response from the disk device 200", a "first time monitoring exceeded" from the first time monitoring mechanism 141, or a "second time monitoring exceeded" from the second time monitoring mechanism 142 has occurred.

[Other embodiments]
The above-described embodiment is the simplest way to realize the present invention. However, the above-described embodiment has a drawback in that it is not effective when the transmission/reception circuit 210 on the disk device 200 side detects a fault on the serial interface 300. Therefore, the following two embodiments will be described as modified examples.

In the first variation, when the second time monitoring mechanism 142 detects that there has been a "second time monitoring over," i.e., when there has been no I/O command response from the disk device 200 after a certain period of time (B), the I/O processing device 110 reads the CRC error counter 150 and compares it with the value read before the I/O command was started to check whether a CRC error or the like has occurred during execution of the I/O command. If the result of the check shows that there has been no CRC error or the like, the disk device 200 is inquired as to whether the I/O command in question is being processed internally, and if it is notified that the I/O command is not being executed internally in the disk device 200, it is determined that there has been no I/O command response even after a certain period of time (B) has passed due to a failure in the serial interface 300, and fault processing is initiated.

In the second variation, when the second time monitoring mechanism 142 detects that "the second time monitoring has expired," that is, when there is no response to an I/O command from the disk device 200 after a certain period of time (B) has elapsed, the I/O processing device 110 reads the CRC error counter and compares it with the value read before the I/O command was started to check whether a CRC error or the like has occurred during execution of the I/O command. If the check shows that no CRC error or the like has occurred, the I/O processing device 110 queries the disk device 200 about the CRC error counter of the disk device 200 side port. If the counter has counted up, it is determined that there has been no response to an I/O command after a certain period of time (B) has elapsed due to the influence of the fault detection that caused the count to be incremented, and fault processing is initiated.

In each of the embodiments described above, when a minor fault such as a CRC error occurs in the host-disk device interface, it is possible to detect the abnormality and start fault processing after a certain period of time (B) has elapsed, thereby achieving the effect of speeding up the start of fault processing for the affected I/O command and preventing delays in the entire system.

This point will be explained using the sequence diagrams in Figures 3 and 4.

Figure 3 shows an example of operation when this embodiment is not applied. First, the host 100 notifies the disk device 200 of the start of an I/O command (step A501). Then, the disk device 200 notifies the host 100 of a response to the I/O command (step A502). As shown in Figure 3, this process is usually completed in less than one second.

If a CRC error occurs and the information is discarded, the CRC error counter is simply incremented (step A503).

Then, 20 seconds later, a timeout is detected and processing of the I/O failure is started (step A504). As a result, it takes about 20 seconds for failure processing to start.

Next, FIG. 4 shows an example of operation when this embodiment is applied. First, the host 100 notifies the disk device 200 of the start of an I/O command (step A501). Then, the disk device 200 notifies the host 100 of a response to the I/O command (step A502). Note that, as shown in FIG. 3, this process is usually completed in less than one second.

If a CRC error occurs and the information is discarded, the CRC error counter is simply incremented (step A503).

Then, one second later (fixed time B), a timeout is detected and the CRC error counter is checked. As the check shows that the counter is up, processing of the I/O error is started (step A505). As a result, it takes only about one second until error processing can start.

Furthermore, this embodiment has the advantage of being able to determine whether the cause of the failure is the host, the disk device, or the serial interface, making it possible to initiate more appropriate failure handling.

The host and disk device of the present invention can be realized by hardware, software, or a combination of these.

Although the above-mentioned embodiment is a preferred embodiment of the present invention, the scope of the present invention is not limited to the above-mentioned embodiment, and the present invention can be implemented in various modified forms without departing from the spirit of the present invention.

100 Host 110 Central Processing Unit 120 Main Memory 130 Input/Output Processing Unit 141 First Time Monitoring Mechanism 142 Second Time Monitoring Mechanism 150, 220 CRC Error Counter 160, 210 Transmitting/Receiving Circuit 200 Disk Unit 230 Disk 240 Disk Control Unit 300 Serial Interface

Claims (9)

An I/O command fault recovery circuit incorporated in a host that is connected to a disk device through an interface, comprising :
a counter for counting the number of occurrences of a failure when the failure is detected in the host;
a first time monitoring means for monitoring a response to an I/O command issued by the host to the disk device, and notifying an input/output processing means of the fact that no response is received within a first fixed time period;
a second time monitoring means for monitoring a response to an I/O command issued by the host to the disk device, and notifying the input/output processing means of the fact that no response is received within a second fixed time period that is shorter than the first fixed time period;
an input/output processing means for starting a fault processing when a notification is received from the first time monitoring means and when a notification is received from the second time monitoring means and the counter is counted up;
Equipped with
When the input/output processing means receives a notification from the second time monitoring means and the counter has not counted up, it inquires of the disk device as to whether the issued I/O processing is being performed internally, and if it receives a reply that the I/O processing is not being performed internally, it determines that a fault has occurred in the interface and performs the fault processing for the fault in the interface . An I/O command fault recovery circuit incorporated in a host that is connected to a disk device through an interface, comprising :
a first counter for counting the number of occurrences of a failure when the failure is detected in the host;
a first time monitoring means for monitoring a response to an I/O command issued by the host to the disk device, and notifying an input/output processing means of the fact that no response is received within a first fixed time period;
a second time monitoring means for monitoring a response to an I/O command issued by the host to the disk device, and notifying the input/output processing means of the fact that no response is received within a second fixed time period that is shorter than the first fixed time period;
an input/output processing means for starting a fault processing when a notification is received from the first time monitoring means and when a notification is received from the second time monitoring means and the first counter is counted up;
Equipped with
When the input/output processing means receives a notification from the second time monitoring means and the first counter provided in the I/O instruction fault recovery circuit has not counted up, it inquires whether a second counter provided in the disk device, which counts the number of times a fault is detected in the disk device, has counted up, and when it receives a reply indicating that a count up has occurred, it determines that the count up is due to the detection of the fault, and performs the fault processing for the fault at the count up . 3. The I/O instruction fault recovery circuit according to claim 1 , wherein the counter detects a CRC (Cyclic Redundancy Check) error as a fault. 1. An I/O command fault recovery method performed by an I/O command fault recovery circuit incorporated in a host that is connected to a disk device via an interface, comprising:
providing a counter for counting the number of occurrences of a failure when the failure is detected in the host;
a first time monitoring step for monitoring a response to an I/O command issued by the host to the disk device, and notifying an input/output processing step of the fact that no response is received within a first certain time period;
a second time monitoring step for monitoring a response to an I/O command issued by the host to the disk device, and notifying the input/output processing step of the fact that no response is received within a second certain time period that is shorter than the first certain time period;
an input/output processing step of starting a fault processing when a notification is received in the first time monitoring step and when a notification is received in the second time monitoring step and the counter is counted up;
Equipped with
In the input/output processing step, when a notification is received in the second time monitoring step and the counter has not counted up, an inquiry is made to the disk device as to whether or not the issued I/O processing is being performed internally, and when a reply is received indicating that the I/O processing is not being performed internally in the disk device, it is determined that a fault has occurred in the interface, and the fault processing is performed for the fault in the interface . 1. An I/O command fault recovery method performed by an I/O command fault recovery circuit incorporated in a host that is connected to a disk device via an interface, comprising:
providing a first counter for counting the number of occurrences of a failure when the failure is detected in the host;
a first time monitoring step for monitoring a response to an I/O command issued by the host to the disk device, and notifying an input/output processing step of the fact that no response is received within a first certain time period;
a second time monitoring step for monitoring a response to an I/O command issued by the host to the disk device, and notifying the input/output processing step of the fact that no response is received within a second certain time period that is shorter than the first certain time period;
an input/output processing step of starting a fault processing when a notification is received in the first time monitoring step and when a notification is received in the second time monitoring step and the first counter is counted up;
Equipped with
In the input/output processing step, when a notification is received in the second time monitoring step and the first counter provided in the I/O instruction fault recovery circuit has not counted up, an inquiry is made as to whether a second counter provided in the disk device, which counts the number of times a fault is detected in the disk device, has counted up, and when a reply is notified that a count up has occurred, it is determined that the count up is due to the detection of the fault, and the fault processing is performed for the fault in the count up . 6. The I/O command fault recovery method according to claim 4 , wherein the fault detected by said counter is a CRC (Cyclic Redundancy Check) error. An I/O command fault recovery program installed in an I/O command fault recovery circuit incorporated in a host that is connected to a disk device via an interface, comprising :
a counter for counting the number of occurrences of a failure when the failure is detected in the host;
a first time monitoring means for monitoring a response to an I/O command issued by the host to the disk device, and notifying an input/output processing means of the fact that no response is received within a first fixed time period;
a second time monitoring means for monitoring a response to an I/O command issued by the host to the disk device, and notifying the input/output processing means of the fact that no response is received within a second fixed time period that is shorter than the first fixed time period;
an input/output processing means for starting a fault processing when a notification is received from the first time monitoring means and when a notification is received from the second time monitoring means and the counter is counted up;
Equipped with
an input/output processing means for inquiring of the disk device as to whether the issued I/O processing is being performed internally when the second time monitoring means has not notified the input/output processing means and the counter has not counted up; and if a reply is received indicating that the processing is not being performed internally in the disk device, the input/output processing means determines that a fault has occurred in the interface, and causes a computer to function as an I/O instruction fault recovery circuit that performs the fault processing for the fault in the interface . An I/O command fault recovery program installed in an I/O command fault recovery circuit incorporated in a host that is connected to a disk device via an interface, comprising :
a first counter for counting the number of occurrences of a failure when the failure is detected in the host;
a first time monitoring means for monitoring a response to an I/O command issued by the host to the disk device, and notifying an input/output processing means of the fact that no response is received within a first fixed time period;
a second time monitoring means for monitoring a response to an I/O command issued by the host to the disk device, and notifying the input/output processing means of the fact that no response is received within a second fixed time period that is shorter than the first fixed time period;
an input/output processing means for starting a fault processing when a notification is received from the first time monitoring means and when a notification is received from the second time monitoring means and the first counter is counted up;
Equipped with
An I/O instruction fault recovery program characterized in that, when the input/output processing means receives a notification from the second time monitoring means and the first counter provided in the I/O instruction fault recovery circuit has not counted up, the input/output processing means inquires whether a second counter provided in the disk device, which counts the number of times a fault is detected in the disk device, has counted up, and if a reply is received indicating that a count up has occurred, the input/output processing means determines that the count up is due to the detection of a fault, and causes the computer to function as an I/O instruction fault recovery circuit that performs the fault processing for the fault in the count up. 9. The I/O instruction fault recovery program according to claim 7 , wherein the fault detected by said counter is a CRC (Cyclic Redundancy Check) error.

Images