JP4667312B2 - External storage device access control system, and controller, host system or external storage device performing access control system - Google Patents

Description

  The present invention relates to a control method for a computer system equipped with an external storage device, and in particular, the access control of an external storage device in which the error recovery processing of the external storage device itself works effectively and the operation rate of the external storage device is increased. The present invention relates to a method, an external storage device controller that executes this access control method, an external storage device that incorporates this controller, and a host system that has the function of this controller.

  In a computer system equipped with an external storage device such as a magnetic disk storage device (hereinafter referred to as HDD), the HDD and a host system using the HDD are directly connected to each other via a host interface such as a SCSI (Small Computer System Interface) bus. Or a controller for converting the interface or a RAID (Redundant Arrays of Independent Disks) controller.

  If an error occurs when the host or controller accesses the HDD and no response is returned from the HDD for a certain period of time, the controller issues a reset instruction to the HDD upon detection of a timeout (for example, error recovery by the HDD) The process was canceled), and then an error was reported to the host system.

  When an error occurs in the HDD, error recovery processing is performed in a predetermined procedure in the HDD. In recent years, the error recovery function of this HDD has been greatly enhanced, and it is configured to execute a huge number of recovery processes in a predetermined order. Therefore, even if an error occurs in the HDD in response to an access request from the host system, there is a high possibility that the error will be remedied by this enhanced error recovery processing.

  However, depending on the content of the error, the error recovery process may take time, and a timeout may be detected during the recovery process, and the reset process may be performed. When the reset process is performed, the recovery process is not completed, and as a result, the HDD remains in an error state. When the host system receives an HDD error report, it requests re-access to the HDD. However, in this state, an error occurs at the same location even if re-access is made. Even if the HDD is usable when the processing is completed, there is a situation in which it cannot be recovered.

For this situation, as described in Patent Document 1, for example, when error recovery processing is interrupted due to timeout, the access request content is stored, and when the same access request is received next time, monitoring until timeout is performed. A technique for extending the time (hereinafter referred to as prior art) has been proposed.
JP 2002-358170 A

  However, in the above prior art, the error recovery processing time is secured by extending the time-out time (time until an error is reported to the host), so when a recent new HDD is connected to a conventional old computer system, etc. However, in the case where some other inconvenience is caused by extending the time-out period, there is a possibility that the technique of the cited document cannot be used.

  An object of the present invention is to solve the above-described problems, and provides a controller access control method that enables error recovery processing of the HDD itself without changing the timeout time of a conventional host. There is.

In order to solve the above problems, the present invention mainly adopts the following configuration.
A controller having a processing unit that realizes various functions by a control program via each interface unit for accessing a host system and an external storage device,
When the processing unit accesses the external storage device based on an access request from the host system, if there is no normal response including completion of error recovery processing from the external storage device within a timeout period, the external storage device An error report was sent to the host system without giving a reset instruction to the storage device, and a reset instruction was given to the accessed external storage device based on a re-access request from the host system. Later, the process of accessing again is performed.

It also has a processing unit that implements various functions including mirroring by a control program via a host system interface unit for accessing the host system and an external storage device interface unit for accessing a plurality of external storage devices. Controller,
When the processing unit accesses one external storage device based on an access request from the host system, there is no normal response including completion of error recovery processing from the one external storage device within a timeout period The other external storage device is re-accessed without giving a reset instruction to the one external storage device, and the error recovery process from the other external storage device is performed within the time-out time for the other external storage device. When there is no normal response including completion, without performing a reset instruction to the other external storage device, an error report is given to the host system, and based on a re-access request from the host system, After a reset instruction is given to the one external storage device, a process of accessing again is performed.

A host system interface unit that communicates with the host system; an external storage device interface unit that communicates with the external storage device; a processing unit that performs arithmetic processing based on the communication; and a result of the arithmetic processing A controller for controlling communication between the host system and the external storage device,
The storage unit stores a determination flag for determining whether or not the controller has reset the error recovery processing of the external storage device,
The processing unit checks the determination flag when receiving an access request from the host system, and accesses the external storage device without resetting if the determination flag is ON (reset). If the determination flag is OFF (not reset), a process for executing access after resetting the external storage device, and after executing the access, the external storage device is predetermined. If no response is received within the time period, the external storage device is not reset, and after the flag is turned OFF, an error reporting process to the host system is executed.

In the controller, the interface unit for external storage device has a function of accessing a plurality of external storage devices, and the storage unit has reset the error recovery processing of one external storage device by the controller. A determination flag for determining whether or not, and a determination flag for determining whether or not a reset for error recovery processing of another external storage device has been performed,
The processing unit, when receiving an access request from the host system, a process of selecting the one or other external storage device to be accessed, a process of checking a determination flag of the selected one external storage device, If the determination flag is ON (reset), the process of executing access without resetting the selected one external storage device, and if the determination flag is OFF (not reset) A process of executing access after resetting the selected one external storage device, and when the selected one external storage device does not respond within a predetermined time after executing the access, The selected one external storage device is not reset, the flag is turned off, and the other external storage device is processed in the same manner as the one external storage device. And configured to execute cormorants and processing, the.

  Older generation HDDs also have simple error recovery processing and did not require much time. However, current new generation HDDs have advanced and complicated error recovery processing, which requires time. is there. For this reason, when a new generation HDD is applied to a host system using an old generation HDD, the time-out time is shorter than the time required for the recovery process, and the error recovery process of the new generation HDD can be used effectively. could not.

  According to the present invention, it is possible to secure a time for error recovery processing even in a host system with a short timeout time as described above, and to make the processing work effectively.

  A controller access control system according to an embodiment of the present invention will be described in detail below with reference to the drawings. The first embodiment of the present invention will be described with reference to FIGS. 1, 2 and 3, and the second embodiment of the present invention will be described with reference to FIGS. 4, 5-1 to 5-4 and FIG. 6 will be described. In the drawing, 10 is a host system, 20 is a controller, 21 is a host access interface, 22 is a processing unit, 23 is an external storage device access interface, 24 is a storage unit, 30 is an external storage device (HDD), and 30A is an external device. Storage devices 1 (HDD1) and 30B represent external storage devices 2 (HDD2), respectively.

“First Embodiment”
FIG. 1 is a block diagram showing the overall configuration of a computer system including a controller according to the first embodiment of the present invention.

  As shown in FIG. 1, the computer system according to the first embodiment includes a host system (host computer) 10 and an external storage device 30, such as an HDD, used as the external storage device of the host system 10. (Magnetic disk device), a host system 10, and a controller 20 that mediates the HDD 30. Here, the controller 20 refers to an interface conversion controller that mediates between the host interface and the HDD interface and a RAID controller for connecting two or more HDDs.

  The controller 20 includes a processing unit 22 for realizing various functions, a storage unit 24 in which a control program is stored in advance, an interface 21 for accessing the host system 10, and an interface 23 for accessing an external storage device. And have.

  Next, processing performed by the processing unit of the controller according to the first embodiment of the present invention will be described with reference to FIG. 2 (refer to FIG. 1 as appropriate). FIG. 2 is a flowchart showing processing of the controller according to the first embodiment of the present invention.

  First, the controller 20 receives an access request to the HDD from the host system 10 (S101). The processing unit 22 of the controller checks whether the HDD 30 is busy (S102). If “HDD is busy”, the process proceeds to S201. If “HDD is not busy”, the process proceeds to S103.

  When the process proceeds to S201, it is determined whether the busy state is canceled after a lapse of a specified time (S201). If “the busy state is released after a specified time”, the process proceeds to S103. If “the busy state is not released even after a specified time”, the process proceeds to S108. When the process proceeds to S103, the processing unit 22 determines the state of the reset flag in the storage unit 24 (S103). Here, the reset flag is a flag for determining whether or not the controller 20 has reset the HDD 30.

  If the “reset flag is ON (reset is performed)”, the process proceeds to S104, and the controller accesses the HDD (S104). When the “reset flag is OFF (not reset)”, the process proceeds to S301, where the controller 20 resets the HDD 30 and then accesses the HDD (S301). At this time, the reset flag of the storage unit is turned ON (S302). In either case of S104 or S302, the process proceeds to S105.

  As a result of accessing the HDD, if “an error response has been received from the HDD”, the process proceeds to S401. After resetting to the HDD, an error report is sent to the host system, and the process ends (S401). If “no error response from HDD”, the process proceeds to S106. It is determined whether no response is obtained from the HDD and a timeout is detected (S106). If “HDD time-out is detected”, the process proceeds to S107. If “HDD time-out is not detected”, the process proceeds to S501.

  When the process proceeds to S107, the processing unit turns off the reset flag of the storage unit (S107). Thereafter, an error report is made to the host system without resetting the HDD, and the process is terminated (S108). When the process proceeds to S501, a normal response is received from the HDD (S501). Thereafter, a normal response is transmitted to the host, and the process is terminated (S502).

  Next, the processing of the computer system according to the first embodiment of the present invention will be described with reference to FIG. 3 (refer to FIGS. 1 and 2 as appropriate). FIG. 3 is a schematic diagram for explaining processing of the computer system according to the first embodiment of the present invention. FIG. 3A shows a case where no error occurs when the controller accesses the HDD. FIG. 3B is a schematic diagram showing a case where error recovery processing is completed within the time-out time when the controller accesses the HDD, and FIG. 3C is a diagram showing the controller accessing the HDD. FIG. 10 is a schematic diagram showing a case where error recovery processing has not been completed within a timeout time when performing.

  Referring to FIG. 3A, first, the host system issues an access request to the controller (U110). The controller that has received the access request accesses the HDD (U120). The HDD performs normal operation (U130) and returns a normal response to the controller (U131). The controller that has received a normal response from the HDD returns a normal response to the host system (U121).

  Referring to FIG. 3B, first, the host system issues an access request to the controller (U210). The controller that has received the access request accesses the HDD (U220). If an error occurs here (when the read sector at a certain address has a temporary medium defect in the corresponding sector and cannot be read), the HDD starts error recovery processing (U230).

  At this time, if the recovery process is completed by the timeout time U221, the HDD returns a normal response to the controller (U231). The controller that has received a normal response from the HDD returns a normal response to the host system (U222).

  Referring to FIG. 3C, first, the host system issues an access request to the controller (U310). The controller that has received the access request accesses the HDD (U320). If an error occurs here, the HDD starts error recovery processing (U330).

  At this time, if the recovery process does not end before the timeout time U321, the controller normally tries to reset the HDD, but suppresses this (U322). (Here, if the recovery process does not end within the timeout period, the meaning of resetting is that if a new request different from the access request to be reset is made next time, the reset will be reset. This is to stop the recovery process because there is a case where normal operation can be performed for the next request without executing the HDD recovery process for the access request. Reports an error to the host system (U323). During this time, the recovery process continues as shown in the figure.

  In response to this, the host system makes a re-access request to the controller (for example, an access request to read the contents of the same address as the previous time) (U311). The controller that has received the re-access request resets the HDD (U324). Therefore, the error recovery process U330 can continue the HDD recovery process until the reset U324 is received.

  After the reset is completed, the controller accesses the HDD (U325). Since the error recovery processing has been completed, the HDD performs a normal operation (U331) and sends a normal response to the controller (U332). The controller that has received a normal response from the HDD returns a normal response to the host system (U326).

“Second Embodiment”
FIG. 4 is a block diagram showing an overall configuration of a computer system including a controller according to the second embodiment of the present invention. The second embodiment of the present invention assumes a case where a plurality of HDDs are connected to a controller and mirroring is performed. The mirroring described here refers to a mechanism in which the same information is recorded in a plurality of HDDs, and information is read from another HDD when an error occurs in the first HDD during information reading. In addition, about the part which is common in 1st Embodiment in the whole structure of a computer system, the same code | symbol is attached | subjected and description is abbreviate | omitted.

  First, the configuration of a computer system according to the second embodiment of the present invention will be described with reference to FIG. As shown in FIG. 4, in the computer system, the host system 10 and the controller 20 have the same configuration as in FIG. Here, the only difference from FIG. 1 is that the target connected to the external storage device access interface 23 of the controller is a plurality of external storage devices 1 (30A) and 2 (30B).

  Next, processing performed by the processing unit of the controller according to the second embodiment of the present invention will be described with reference to FIGS. 5-1 to 5-4 (refer to FIG. 4 as appropriate). FIGS. 5-1 to 5-4 are flowcharts showing a part of the processing of the controller according to the second embodiment of the present invention.

  FIG. 5A will be described. First, the controller 20 receives an access request to the HDD from the host system 10 (T111). The processing unit 22 of the controller selects the HDD to be accessed from the HDD 1 (30A) or the HDD 2 (30B) (T112). The selection method here is not limited, and a near-less seek in which the HDD whose head is close to the data to be accessed may be selected, or another method may be used. If “the selected HDD is HDD1,” the process proceeds to T113. If “the selected HDD is HDD2,” the process proceeds to B in FIG.

  When the process proceeds to T113, the processing unit 22 checks whether the HDD 1 is busy (T113). If “HDD1 is busy”, the process proceeds to T121. If “HDD1 is not busy”, the process proceeds to T114. When the process proceeds to T121, it is determined whether the busy state is canceled after a lapse of a specified time (T121). If “the busy state is released after a specified time”, the process proceeds to T114. If “the busy state is not released even after a specified time”, the process proceeds to T119.

  When the process proceeds to T114, the processing unit 22 determines the state of the reset flag of the HDD 1 in the storage unit 24 (T114). If “the reset flag of HDD1 is ON”, the process proceeds to T115, and the controller 20 accesses the HDD1 (T115). If “the reset flag of HDD1 is OFF”, the process proceeds to T131, and the controller 20 resets the HDD1, and then accesses the HDD1 (T131). At this time, the reset flag of the HDD 1 in the storage unit 24 is turned ON (T132). In either case of T115 or T132, the process proceeds to T116.

  As a result of accessing the HDD 1, if “an error response is received from the HDD 1”, the process proceeds to T 141, the HDD 1 is reset (T 141), and then the process proceeds to T 119. If there is no error response from HDD1, the process proceeds to T117. It is determined whether no response is obtained from the HDD 1 and a timeout is detected (T117). If “HDD 1 timeout is detected”, the process proceeds to T118. If “HDD 1 timeout is not detected”, the process proceeds to T151.

  When the process proceeds to T118, the processing unit turns off the reset flag of the HDD 1 in the storage unit (T118). In either case of T118 or T141, the HDD to be accessed is subsequently changed to HDD2 (T119), and the process proceeds to A in FIG. When the process proceeds to T151, a normal response is received from the HDD 1 (T151). Thereafter, a normal response is transmitted to the host, and the process is terminated (T152).

  Next, FIG. 5-2 will be described. When the process proceeds from T119 to T211 in FIG. 5A described above, the processing unit 22 checks whether the HDD 2 is busy (T211). If “HDD2 is busy”, the process proceeds to T221. If “HDD2 is not busy”, the process proceeds to T212. When the process proceeds to T221, it is determined whether the busy state is canceled after a lapse of a specified time (T221). If “the busy state is released after a specified time”, the process proceeds to T212. If “the busy state is not released even after a specified time”, the process proceeds to T217.

  When the process proceeds to T212, the processing unit determines the state of the reset flag of the HDD 2 of the storage unit (T212). If “the reset flag of the HDD 2 is ON”, the process proceeds to T213, and the controller 20 accesses the HDD 2 (T213). If “the reset flag of the HDD 2 is OFF”, the process proceeds to T231, and the controller 20 resets the HDD 2 and then accesses the HDD 2 (T231). At this time, the reset flag of the HDD 2 in the storage unit 24 is turned ON (T232). In either case of T213 or T232, the process proceeds to T214.

  As a result of accessing the HDD 2, if “an error response is received from the HDD 2”, the process proceeds to T 241, after resetting to the HDD 2, an error report is sent to the host system 10 and the process is terminated (T 241). If there is no error response from the HDD 2, the process proceeds to T215. It is determined whether no response is obtained from the HDD 2 and a timeout is detected (T215). If “HDD 2 timeout is detected”, the process proceeds to T216. If “HDD 2 timeout is not detected”, the process proceeds to T251.

  When the process proceeds to T216, the processing unit 22 turns off the reset flag of the HDD 2 in the storage unit 24 (T216). Thereafter, an error report is made to the host system without resetting to the HDD 2, and the process is terminated (T217). When the process proceeds to T251, a normal response is received from the HDD 2 (T251). Thereafter, a normal response is transmitted to the host, and the process is terminated (T252).

  FIG. 5-3 will be described. When the process proceeds from T112 to T311 in FIG. 5A described above, the processing unit 22 checks whether the HDD 2 is busy (T311). If “HDD2 is busy”, the process proceeds to T321. If “HDD2 is not busy”, the process proceeds to T312. When the process proceeds to T321, it is determined whether the busy state is canceled after a lapse of a specified time (T321). If “the busy state is released after a specified time”, the process proceeds to T312. If “the busy state is not released even after the specified time has elapsed”, the process proceeds to T317.

  When the process proceeds to T312, the processing unit 22 determines the state of the reset flag of the HDD 2 in the storage unit 24 (T312). If “the reset flag of HDD2 is ON”, the process proceeds to T313, and the controller accesses the HDD2 (T313). If “the reset flag of the HDD 2 is OFF”, the process proceeds to T331, and the controller resets the HDD 2 and then accesses the HDD 2 (T331). At this time, the reset flag of the HDD 2 in the storage unit 24 is turned ON (T332). In either case of T313 or T332, the process proceeds to T314.

  As a result of accessing the HDD 2, if “an error response is received from the HDD 2”, the process proceeds to T 341, the HDD 2 is reset (T 341), and then the process proceeds to T 317. If “no error response from HDD 2”, the process proceeds to T315. It is determined whether no response is obtained from the HDD 2 and a timeout is detected (T315). If “HDD 2 timeout is detected”, the process proceeds to T316. If “HDD 2 timeout is not detected”, the process proceeds to T351.

  When the process proceeds to T316, the processing unit 22 turns off the reset flag of the HDD 2 in the storage unit 24 (T316). In either case of T316 or T341, the HDD to be accessed is subsequently changed to HDD1 (T317), and the process proceeds to C in FIG. When the process proceeds to T351, a normal response is received from the HDD 2 (T351). Thereafter, a normal response is transmitted to the host, and the process ends (T352).

  FIG. 5-4 will be described. When the process proceeds from T317 in FIG. 5C to T411, the processing unit 22 checks whether the HDD 1 is busy (T411). If “HDD1 is busy”, the process proceeds to T421. If “HDD1 is not busy”, the process proceeds to T412. When the process proceeds to T421, it is determined whether the busy state is canceled after a lapse of a specified time (T421). If “the busy state is released after a specified time”, the process proceeds to T412. If “the busy state is not released even after the specified time has elapsed”, the process proceeds to T417.

  When the process proceeds to T412, the processing unit 22 determines the state of the reset flag of the HDD 1 in the storage unit 24 (T412). When “the reset flag of HDD1 is ON”, the process proceeds to T413, and the controller 20 accesses the HDD1 (T413). If “the reset flag of HDD1 is OFF”, the process proceeds to T431, where the controller resets HDD1 and then accesses HDD1 (T431). At this time, the reset flag of the HDD 1 of the storage unit is turned ON (T432). In either case of T413 or T432, the process proceeds to T414.

  As a result of accessing HDD 1, if “an error response is received from HDD 1”, the process proceeds to T 441, after resetting to HDD 1, an error report is sent to the host system 10, and the process is terminated (T 441). If there is no error response from HDD1, the process proceeds to T415. It is determined whether no response is obtained from the HDD 1 and a timeout is detected (T415). If "HDD 1 timeout is detected", the process proceeds to T416. If "HDD 1 timeout is not detected", the process proceeds to T451.

  When the process proceeds to T416, the processing unit 22 turns off the reset flag of the HDD 1 in the storage unit 24 (T416). Thereafter, the HDD 1 is not reset and an error is reported to the host system, and the process is terminated (T417). When the process proceeds to T451, a normal response is received from the HDD 2 (T451). Thereafter, a normal response is transmitted to the host, and the process ends (T452).

  Next, the processing of the computer system according to the second embodiment of the present invention will be described with reference to FIG. 6 (refer to FIGS. 4, 5-1 to 5-4 as appropriate). FIG. 6 is a schematic diagram for explaining processing of the computer system according to the second embodiment of the present invention. FIG. 6A shows a case where no error occurs when the controller accesses the HDD. FIG. 6B is a schematic diagram showing a case where error recovery processing is completed within the timeout time when the controller accesses the HDD, and FIG. 6C is a diagram showing the first controller. FIG. 6D is a schematic diagram showing a case where the error recovery processing is not completed within the time-out time when the HDD 2 is accessed and the access is switched to another HDD 1 and no error occurs in this HDD 1. When the controller accesses the first HDD 2, the error recovery processing is not completed within the timeout time, and the access is switched to another HDD 1. It is a schematic diagram showing a case where error correction processing is not completed within the out time.

  Referring to FIG. 6A, first, the host system 10 issues an access request to the controller 20 (V110). Receiving the access request, the controller 20 selects the HDD to be accessed and accesses the selected HDD. FIG. 6A shows a case where the HDD 2 is selected (V120).

  The HDD 2 performs a normal operation (V130) and returns a normal response to the controller (V131). The controller that has received a normal response from the HDD 2 returns a normal response to the host system (V121).

  Referring to FIG. 6B, first, the host system 10 issues an access request to the controller 20 (V210). Receiving the access request, the controller 20 selects the HDD to be accessed and accesses the selected HDD. FIG. 6B shows a case where the HDD 1 is selected (V220).

  If an error occurs here, the HDD 1 starts error recovery processing (V230). At this time, if the recovery process is completed by the timeout time V221, the HDD 1 returns a normal response to the controller (V231). The controller 20 that has received a normal response from the HDD 1 returns a normal response to the host system 10 (V222).

  Referring to FIG. 6C, first, the host system 10 issues an access request to the controller 20 (V310). The controller that has received the access request selects the HDD to be accessed, and accesses the selected HDD. FIG. 6C shows a case where the HDD 2 is selected (V320).

  If an error occurs here, the HDD 2 starts error recovery processing (V330). At this time, if the recovery process is not completed before the timeout time V321, the controller 20 normally tries to reset the HDD 2, but suppresses this (V322).

  The controller switches the HDD to be accessed to HDD 1 and performs re-access (V323). The HDD 1 performs a normal operation, and the HDD 1 returns a normal response to the controller 20 (V332). The controller that has received a normal response from the HDD 1 returns a normal response to the host system (V324). Meanwhile, the error recovery process V330 completes the process.

  Referring to FIG. 6D, first, the host system 10 issues an access request to the controller 20 (V410). The controller that has received the access request selects the HDD to be accessed, and accesses the selected HDD. FIG. 6D shows a case where the HDD 2 is selected (V420).

  If an error occurs here, the HDD 2 starts error recovery processing (V430). At this time, if the recovery process does not end before the timeout time V421, the controller 20 normally tries to reset the HDD 2, but suppresses this (V422). The controller 20 switches the HDD to be accessed to the HDD 1 and performs re-access (V423). If an error also occurs in the HDD 1, the HDD 1 starts error recovery processing (V431).

  At this time, if the recovery process is not completed before the timeout time V424, the controller 20 normally tries to reset the HDD 1 but suppresses this (V425). Thereafter, the controller reports an error to the host system (V426).

  In response to this, the host system 10 issues a re-access request to the controller 20 (V412). The controller 20 that has received the re-access request selects the HDD to be accessed, and resets the selected HDD 2 (V427). After the reset is completed, the controller 20 accesses the HDD 2 (V428).

  Since the HDD 2 has completed the error recovery process (assuming that the error recovery process V430 of the HDD 2 shown in the figure has completed the recovery process before the reset V427), the HDD 2 performs a normal operation (V432), A normal response is made to the controller (V433). The controller that has received a normal response from the HDD 2 returns a normal response to the host system (V429).

  As described above, the embodiment of the present invention is mainly characterized by having the following configuration and providing operational effects. That is, when the host system makes an access request, even if an error occurs in the HDD and a timeout occurs during the error recovery process, the controller does not reset the HDD. The reset is performed immediately before the next access to the HDD. As a result, when there is free time between the first access and the second access, the time can be used for error recovery processing, and the HDD itself can be increased by extending the substantial recovery processing time. It is possible to increase the possibility that the error recovery process can be completed without interruption.

  Furthermore, if the feature of the present embodiment is applied to a RAID controller (mirror controller), the time when an error occurs in the first HDD and the process is switched to another HDD can be allocated to the error recovery process. It is possible to increase the possibility that error recovery processing has been completed before an access request is made to the first HDD again. In the above description, the controller has the functions shown in FIGS. 2 and 5-1 to 5-4. However, the present invention is not limited to this, and the same function may be built in the host system. Also, it may be provided in an external storage device.

1 is a block diagram showing an overall configuration of a computer system including a controller according to a first embodiment of the present invention. It is a flowchart which shows the process of the controller which concerns on the 1st Embodiment of this invention. It is a mimetic diagram explaining processing of a computer system including a controller concerning a 1st embodiment of the present invention. It is a block diagram which shows the whole structure of the computer system containing the controller which concerns on the 2nd Embodiment of this invention. It is a flowchart which shows a part of process of the controller which concerns on the 2nd Embodiment of this invention. It is a flowchart which shows a part of process of the controller which concerns on the 2nd Embodiment of this invention. It is a flowchart which shows a part of process of the controller which concerns on the 2nd Embodiment of this invention. It is a flowchart which shows a part of process of the controller which concerns on the 2nd Embodiment of this invention. It is a mimetic diagram explaining processing of a computer system including a controller concerning a 2nd embodiment of the present invention.

Explanation of symbols

10 Host System 20 Controller 21 Host Access Interface 22 Processing Unit 23 External Storage Device Access Interface 24 Storage Unit 30 External Storage Device (HDD)
30A External storage device 1 (HDD 1)
30B External storage device 2 (HDD 2)

Claims (7)

A controller having a processing unit that realizes various functions by a control program via each interface unit for accessing a host system and an external storage device,
When the processing unit accesses the external storage device based on an access request from the host system, if there is no normal response including completion of error recovery processing from the external storage device within a timeout period, the external storage device Without giving a reset instruction to the storage device, reporting an error to the host system,
A controller that performs a process of accessing again after giving a reset instruction to the accessed external storage device based on a re-access request from the host system. A controller having a processing unit for realizing various functions including mirroring by a control program via a host system interface unit for accessing the host system and an external storage device interface unit for accessing a plurality of external storage devices Because
When the processing unit accesses one external storage device based on an access request from the host system, there is no normal response including completion of error recovery processing from the one external storage device within a timeout period The other external storage device is re-accessed without giving a reset instruction to the one external storage device, and the error recovery process from the other external storage device is performed within the time-out time for the other external storage device. When there is no normal response including completion, an error report is sent to the host system without instructing the reset to the other external storage device,
A controller that performs a process of accessing again after instructing a reset to the one external storage device based on a re-access request from the host system. A host system interface unit that communicates with a host system, an external storage device interface unit that communicates with an external storage device, a processing unit that performs arithmetic processing based on the communication, and a result of the arithmetic processing are stored A controller that controls communication between the host system and the external storage device,
The storage unit stores a determination flag (ON when resetting, OFF when not resetting) for determining whether or not the controller has reset the error recovery processing of the external storage device,
The processing unit checks the determination flag when receiving an access request from the host system, and accesses the external storage device without resetting if the determination flag is ON (reset). If the determination flag is OFF (not reset), a process for executing access after resetting the external storage device, and after executing the access, the external storage device is predetermined. If the response is not received within a predetermined time, the external storage device is not reset, and after the flag is turned OFF, an error is reported to the host system. Controller. In claim 3,
The external storage device interface unit has a function of accessing a plurality of external storage devices,
The storage unit determines whether or not the controller has reset the error recovery process of one external storage device and whether or not the error recovery process of another external storage device has been reset. A determination flag for determining, and
The processing unit, when receiving an access request from the host system, a process of selecting the one or other external storage device to be accessed, a process of checking a determination flag of the selected one external storage device, If the determination flag is ON (reset), the process of executing access without resetting the selected one external storage device, and if the determination flag is OFF (not reset) A process of executing access after resetting the selected one external storage device, and when the selected one external storage device does not respond within a predetermined time after executing the access, The selected one external storage device is not reset, the flag is turned off, and the other external storage device is processed in the same manner as the one external storage device. Controller and executes cormorants and processing, the.   5. A host system incorporating the function of the controller according to any one of claims 1 to 4.   An external storage device incorporating the function of the controller according to any one of claims 1 to 4.   A computer system comprising the controller according to any one of claims 1 to 4, a host system, and an external storage device.

Images