JP2008148339A - Storage device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To suppress reduction in performance or reliability to a minimum when a fault occurs in a system comprising a loop-like communication means such as FC-AL. <P>SOLUTION: A plurality of controllers 101, 102 and a plurality of drives 111-114 are connected via a plurality of FC loops 161, 171, wherein in addition to PBC 121-126 and PBC 131-136 for individually controlling bypasses to the FC loops 161, 171 of the controllers 101, 102 and the drives 111-114, PBC 141-143 and PBC 151-153 are provided for short-circuiting the FC loops 161, 171 in the middle of the path and the FC loops are bypassed at positions of the PBC 141-143 and the PBC 151-153, so that the FC loops are continuously used by partially disconnecting a fault portion in the FC loop and reduction in performance or reliability in fault occurrence is suppressed to a minimum. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a storage device and information processing technology, and more particularly to a technology effective when applied to a storage device, an information processing system, and the like having a configuration in which a plurality of components are connected by loop communication means such as a fiber channel loop.

  As one of ultra-high speed gigabit network technologies, Fiber Channel, which is being standardized by ANSI NCITST11 (formerly ANSI X3 T11), is known.

  By using such a fiber channel (FC), it is possible to connect a plurality of hard disks (drives) to construct one large storage device. In this FC loop connection method (FIBRE CHANNEL ARBITRATED LOOP (FC-AL)), the controller for controlling the drive of the storage device and each drive are connected in a loop. A connection portion between each drive and the FC loop is connected to a port bypass circuit (PBC) for disconnecting the drive from the loop, such as when a drive is replaced or failed.

  According to the FC loop standard, if the loop is disconnected even at one location, communication between the controller and each drive on the loop becomes impossible. Therefore, in the external storage device constructed using the FC loop, individual When disconnecting the drive, it is necessary to control the FC loop so that the connection is changed by the PBC and the loop is not broken.

As described in Japanese Patent Laid-Open No. 10-285198 (Patent Document 1), a drive with a low usage rate is used by controlling the PBC not only when an abnormality occurs but also during normal operation to improve performance. It is also possible to control the PBC to connect to the loop only for the time and to distribute the load of each loop.
Japanese Patent Laid-Open No. 10-285198

  In a single storage device constructed by connecting a plurality of drives using an FC loop, if a failure occurs in a specific drive, the PBC decouples the failed drive from the loop to another drive. Can continue to operate. However, there is a technical problem that all the drives connected to the loop become unusable when the connection of the loop is interrupted for some reason or when communication of the loop is interrupted.

  As a countermeasure for the above technical problem, in a storage device configured by connecting multiple drives via an FC loop, if the FC loop is set to be double and a failure occurs in one loop, the other A method of continuing communication using a loop can be considered.

  However, even when a countermeasure is taken to set such an FC loop in a double manner, the location of the fault is unknown in the loop where the fault has occurred. For this reason, the maintenance staff can identify the location of the failure by exchanging the individual components such as individual drives included in the failure loop one by one and confirming that they are operating normally. Another technical problem arises that it is necessary to specify complicated work.

  Further, since communication is performed using only one of the double loops, performance such as data transfer speed is degraded. Furthermore, even when the loop is set to be double, if a failure occurs in a common part of each loop of the drive, communication with the two loops becomes impossible, and the entire storage device becomes inoperable. There is a technical problem.

  An object of the present invention is to provide a storage device capable of minimizing degradation in performance and reliability when a failure occurs in a configuration including a loop communication means.

  Another object of the present invention is to provide a storage device capable of quickly, simply and accurately performing identification and recovery of a faulty part in a configuration including a loop communication means.

  Another object of the present invention is to provide a storage device capable of reliably performing recovery when a multiple failure occurs across a plurality of loop communication means in a configuration including multiple loop communication means. .

  An object of the present invention is to provide an information processing system capable of minimizing a decrease in performance and reliability when a failure occurs in a configuration including a loop communication unit.

  Another object of the present invention is to provide an information processing system capable of quickly, simply and accurately performing a failure site identification and recovery operation in a configuration including a loop communication means.

  Another object of the present invention is to provide an information processing system capable of reliably recovering from the occurrence of multiple faults extending to a plurality of loop communication means in a configuration having multiple loop communication means. is there.

  The present invention relates to at least one unit storage device, at least one controller for controlling the unit storage device, the unit storage device and the controller connected in a loop, and between the controller and the unit storage device. In a storage device including a loop communication means for exchanging information, a first bypass means for individually controlling connection and disconnection of the unit storage device and the controller to the loop communication means, and the loop communication means And a second bypass means for selectively disconnecting a part of the path of the loop communication means.

  The present invention also provides a plurality of unit components that each perform at least one of storage and processing of information and a plurality of the unit components connected in a loop to exchange information between the unit components. In the information processing system including the loop-shaped communication unit, the first bypass unit that individually controls connection and disconnection of the unit component to and from the loop-shaped communication unit, and the loop-shaped communication unit at an arbitrary position. And a second bypass means for selectively disconnecting a part of the path of the loop communication means.

  More specifically, in a storage device having a configuration in which a plurality of drives and a controller are connected by a loop such as FC_AL, in addition to the PBC (first bypass means) that separates the drive and the controller from the loop, the loop PBC (second bypass means) for short-circuiting the loop is installed at these locations.

  By controlling these PBCs, it is determined where in the loop the failure has occurred. Actually, the PBC is controlled to enable only a part of the loop, and the operation of confirming whether communication is possible is performed by changing the effective part of the loop and repeating the operation to identify the faulty part. If there is a part that can be used in the loop, the loop is used to switch only the unusable part to the other loop, thereby minimizing performance degradation.

  Also, instead of issuing an instruction to control the PBC from the loop, by providing a dedicated bus for controlling the PBC, even if both loops cannot communicate, the faulty part is separated. The other parts can return to the communicable state.

  According to the storage device of the present invention, it is possible to obtain an effect that it is possible to minimize a decrease in performance and reliability when a failure occurs in a configuration including a loop communication unit.

  According to the storage device of the present invention, in the configuration including the loop communication means, it is possible to obtain the effect that the faulty part can be identified and recovered quickly, simply and accurately.

  According to the storage device of the present invention, it is possible to reliably perform recovery when a multiple failure occurs across a plurality of loop communication means in a configuration including multiple loop communication means.

  According to the information processing system of the present invention, it is possible to obtain an effect that it is possible to minimize a decrease in performance and reliability when a failure occurs in a configuration including a loop communication unit.

  According to the information processing system of the present invention, in the configuration including the loop communication means, it is possible to obtain an effect that it is possible to quickly and easily and accurately perform the identification and recovery work of the faulty part.

  According to the information processing system of the present invention, it is possible to reliably perform recovery when a multiple failure occurs across a plurality of loop communication means in a configuration including multiple loop communication means.

  Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings.

  FIG. 1 is a conceptual diagram showing an example of a loop connection configuration of a storage device which is an embodiment of an information processing system and a storage device of the present invention. FIG. 2 is an overall view of the storage device of the present embodiment. It is a conceptual diagram which shows an example of a structure.

  In this embodiment, a case where FC_AL (FIBER CHANNEL ARBITRATED LOOP) (hereinafter referred to as an FC loop) is used as an example of a loop communication unit in a storage device will be described as an example.

  As illustrated in the overall configuration of FIG. 2, the storage device according to the present embodiment is configured to exchange data between a plurality of controllers 101, a controller 102, a cache memory 103 shared by them, and a host device (not shown). A plurality of host interfaces 104 and a host interface 105 are provided.

  Under the plurality of controllers 101 and 102, a plurality of drives 111, drives 112, drives 113, and drives 114 are shared via multiple FC loops 161, FC loops 171 and port bypass circuits (PBC). Connected so that.

  That is, as illustrated in FIG. 2, in the storage device of the present embodiment, two controllers 101 and 102 are connected to separate FC loops 161 and 162 through PBC 121, PBC 122, PBC 131, and PBC 132, respectively. Yes. The four drives 111, 112, 113, 114 are connected to one FC loop 161 through PBC123, PBC124, PBC125, PBC126, and connected to the other FC loop 171 through PBC133, PBC134, PBC135, PBC136. .

  In the present embodiment, PBC 141, PBC 142, and PBC 143 for short-circuiting the FC loop are connected to the FC loop 161, and PBC 151, PBC 152, and PBC 153 are connected to the FC loop 171.

  3 and 4 show an example of the configuration of the PBC in the present embodiment. That is, as illustrated in FIG. 3, each of PBC 121 to PBC 126 and PBC 131 to PBC 136 (hereinafter collectively referred to as PBC 600) that connects individual controllers and drives to the FC loop is a drive for the FC loop 161 and the FC loop 171. The selector 601 for switching the presence / absence of connection of the controller and the presence / absence of an operation abnormality on the drive / controller side are monitored. When an abnormality (failure) occurs, the bypass control signal 602 input to the selector 601 Lights up when the fault monitoring circuit 603 for performing the operation of bypassing the controller and disconnecting from the FC loop side (the state of the selector 601 in FIG. 3 indicates the bypass state) and the bypass control signal 602 are ON (bypass state). Make sure that it is bypassed externally. Display lamp 604 such as an LED that knowledge, and a like.

  The bypass control signal 602 can be input from the outside (for example, a control terminal (not shown) provided in the drive), and the bypass state can be controlled from the outside regardless of the operation of the failure monitoring circuit 603. It is.

  That is, in the storage device of this embodiment, as an example, SCSI-FCP (SCSI-3 Fiber Channel Protocol) is used as the data link layer of the fiber channel protocol in the FC loop 161 and the FC loop 171.

  When this SCSI-FCP is used, the storage device is a parameter list in SCSI Send Diagnostics of FCP commands issued from the controllers 101 and 102 which are SCSI-FCP initiators to each of the drives 111 to 114 as targets. By specifying, it is possible to instruct each of the drives 111 to 114 to output an ON / OFF command of the bypass control signal 602 and the bypass control signal 702.

  Thus, the controllers 101 and 102 control the bypass operation in each of the PBC 600 and the PBC 700 by outputting the bypass control signal 602 and the bypass control signal 702 via control terminals (not shown) of the drives 111 to 114.

  Similarly, the current presence / absence of bypass in the PBC 600 and PBC 700 (that is, the states of the bypass control signal 602 and the bypass control signal 702) is determined by setting each of the drives 111 to 114 in the SCSI Receive Diagnostics Results of the FCP command. Through this, the controllers 101 and 102 can know.

  On the other hand, as illustrated in FIG. 4, each of PBC 141 to 143 and PBC 151 to 153 (hereinafter collectively referred to as PBC 700) provided in the middle of the FC loop 161 and the FC loop 171 is the FC loop 161 and the FC loop 171. Lights when the selector 701 performs a short circuit (bypass) in the middle of the signal, the bypass control signal 702 input from outside for controlling the switching operation of the selector 701, and the bypass control signal 702 is ON (bypass state). And a display lamp 703 such as an LED for informing the outside of the bypass state.

  For example, in the configuration example of FIG. 1 and FIG. 2, bypass control signal 602 and bypass control signal 702 input to each of PBC 121 to PBC 126 and PBC 131 to PBC 136, PBC 141 to 143 and PBC 151 to 153 are transmitted from controllers 101 and 102, respectively. It is possible to input from the nearest drive via FC loops opposite to each other.

  Thereby, in the storage device of the present embodiment, when a failure occurs in one FC loop 161 (171), the bypass control signal 602 is sent to the target drive via the other FC loop 171 (161). By outputting the bypass control signal 702 to nearby PBCs, switching control of individual PBCs as shown in FIG. 9 described later becomes possible.

  In the storage device of this embodiment, the FC loops 161 and 171 are not limited to using a communication medium such as an optical fiber or a conductive wire, but may be a wiring pattern on a mounting board as illustrated in FIG. The configuration of the FC loops 161 and 171 is also included.

  That is, as illustrated in FIG. 6, in the storage device according to the present embodiment, a plurality of FC loops 161 and FC loops 171 are arranged as wiring patterns on the mounting board 10, and further, PBC 121-1 connected to these are arranged. 126, PBC 131 to 136, and further PBC 141 to 143 and PBC 151 to 153 are formed on the mounting board 10. The units of the plurality of controllers 101 and 102 and the plurality of drives 111 to 114 are detachably mounted on each of the plurality of FC loops 161 and 171 via the connector 11. In the case of such a configuration of FIG. 6, the display lamps of the individual PBCs are arranged on the mounting board 10, for example, and the bypass state of the individual PBCs can be visually recognized from the outside.

  Although not particularly shown, if necessary, a communication medium such as an external optical fiber or a conductor can be connected to the FC loop 161 and the FC loop 171 using a connector 11 instead of a part of the drives. The storage device of this embodiment may be connected to an external system.

  As a storage device having the configuration as shown in FIG. 6, for example, by distributing and storing data exchanged with a host device and redundant data generated from the data to a plurality of drives 111 to 114 A disk array system that can improve the reliability of stored data and improve the throughput by parallel access to a plurality of drives 111 to 114 can be considered.

  In normal operation, the controller 101 uses only the FC loop 161, and the controller 102 uses only the FC loop 171, so that each controller can communicate with each of the communication bands of the FC loops 161 and 171 provided in multiplex. Can be used exclusively. Further, the communication between each of the controllers 101 and 102 and the drives 111 to 114 is not affected by the operation of other controllers.

  As shown in Japanese Patent Laid-Open No. 10-285198, when a drive fails or when an unused drive is disconnected for performance improvement, the storage device of this embodiment uses PBC 123-126 or PBC 133-135. Use to disconnect the drive.

  As an example, when the drive 114 is disconnected from each FC loop, the storage device of this embodiment switches the PBC 126 to disconnect the drive 114 from the FC loop 161 and switches the PBC 136 to disconnect the drive 114 from the FC loop 171.

  The state of the PBC at this time is shown in FIG. In the PBC of the storage device of this embodiment, when a drive or a controller is connected, the connection state 201 illustrated on the left side of FIG. 7 is obtained. When the drive or controller is disconnected, the disconnection state 202 (bypass state) illustrated on the right side of FIG. 7 is set, and the drive or controller is disconnected from the FC loop.

  However, if a failure occurs in a part of the FC loop 161 or 171 instead of the drive or controller, the above method cannot be used for recovery. If a failure occurs in the FC loop 161, the FC loop 161 is used. The controller 101 that had been used avoids a failure by using the FC loop 171 as in the controller 102. In this case, since the controllers 101 and 102 use the same FC loop 171, the communication band is halved compared to the case where two FC loops are used, and the performance is degraded.

  Therefore, in the present embodiment, when an FC loop failure occurs, performance degradation is prevented by separating the failure portion of the FC loop using PBC 141 to 143 or PBC 151 to 153. Specific examples are shown below.

  FIG. 9A shows a state where one FC loop 161 becomes unable to communicate due to a loop failure point 181 (for simplification, only one FC loop 161 is shown). . When the failure of the FC loop occurs, the controller 101 or 102 first disconnects the drive 114 farthest from the controller from the FC loop by switching the PBC 126. This state is shown in FIG. However, even in the state shown in FIG. 9B, the loop failure location 181 exists on the FC loop 161, so that the FC loop 161 cannot communicate.

  Next, the controller 101 or 102 switches the PBC 143 to short-circuit the FC loop. The state of the PBC 143 at this time is shown in FIG. In the normal connection state, the PBC 143 is in the state of the connection state 301 on the left side of FIG. 8, but when the FC loop is short-circuited, the PBC 143 becomes the short-circuit state 302 (bypass state) on the right side of FIG. It is divided into a side closer to (FC loop 161a) and a side farther (FC loop 161b). FIG. 9C shows a state where the PBC 143 is switched to the short circuit state 302. In the state shown in FIG. 9C, the FC loop 161b including the loop fault location 181 is disconnected from the FC loop 161, so that the FC loop 161a can communicate. As a result, it can be determined that a failure of the FC loop has occurred at the loop failure location 181.

  That is, in each PBC of the present embodiment, the display lamp is lit in the bypass state (short circuit state 302). Therefore, where the failure of the FC loop or drive depends on whether or not the display lamp is lit in each PBC at the time of troubleshooting As a result, maintenance work such as replacement of parts becomes easy. For example, since a storage device such as a disk array has a configuration including a large number of drives, a particularly great effect can be expected in speeding up the identification of a failure location.

  In the state of FIG. 9C, the drive 114 belonging to the FC loop 161b including the loop fault location 181 cannot communicate from the FC loop 161. Therefore, it is necessary to perform communication from the other FC loop 171, but the drives 111 and 112 113 can communicate from the FC loop 161 (FC loop 161a), so that an increase in communication load of other FC loops 171 can be suppressed to a minimum.

  Even if a loop failure occurs elsewhere in the FC loop 161, if the controller 101 or 102 switches the PBC 126, 143, 125, 142, 124, 141, 123 in order of increasing distance from the controller of the FC loop, the loop It is possible to determine whether or not a failure has occurred, and by separating the loop failure portion, it is possible to minimize the performance degradation by using the communicable portion of the FC loop.

  FIG. 5 is a flowchart showing an example of the control operation when the controller automatically executes the operation for detecting the faulty part as described above. In this flowchart, IDs such as 0, 1, 2, 3,... Are assigned to each of a plurality of PBCs to be controlled in ascending order from the position farthest from the controller to be recognized / recognized by the failure detection program. An example is shown in which the controller 101 or 102 is automatically controlled to perform bypass / connection operation of each PBC by being controlled.

  That is, when attention is focused on the FC loop 161 (the FC loop 162 side is the same as the following), the controllers 101 and 102 first set the PBCs 123 to 126 and the PBC 141 to 143 other than the controller in the order of increasing distance from the controller. ID of ~ 6 is given. Specifically, PBC126: ID = “0”, PBC125: ID = “1”, PBC124: ID = “2”, PBC123: ID = “3”, PBC143: ID = “4”, PBC142: ID = “ 5 ”, PBC141: ID =“ 6 ”.

  Then, the controllers 101 and 102 monitor the occurrence of a fault in the loop (step 801), and if detected, set the ID of the PBC farthest from the controller (in this case, PBC 126) (step 802), and set the ID of the set ID. The PBC is set to a bypass state (step 803). This operation can be performed with the FCP command described above.

  Next, the controller 101 or 102 checks whether or not the failure has been recovered by the above-described bypass operation (step 804), and performs the same operation while incrementing the ID (up to ID = 6 in this case) until the failure is recovered. Repeat (step 810). Note that if the controller 101 or 102 cannot use a part of the loop (PBC 141: failure is not recovered even by bypass operation with ID = “6”) (step 809), the controller 101 or 102 is closer to the controller than the PBC 141. It is determined that there is an error on the loop of the current loop, and the abandonment of the FC loop 161 is displayed (step 813).

  When the failure is recovered in step 804, the controller 101 or 102 determines whether the bypassed PBC of the ID is the PBC 123-126 for bypassing the drive or the PBC 141-143 for bypassing the loop itself ( Step 805). That is, the controller divides the fault depending on whether the fault site is caused by a drive or a loop.

  When the controller 101 or 102 is found to have a drive failure (step 806), the bypassed PBC (drive) indicator lamp (LED, etc.) is turned on to indicate the bypass state to the outside (step 807). .

  Returning to step 805, if it is determined that the fault is a partial fault in the loop (step 811), the controller 101 or 102 turns on the display lamp 703 of the relevant PBC for the loop to indicate that the fault is a partial fault in the loop. A process of displaying outside (step 812) and shifting the drive belonging to the loop separated by bypass to the other (in this case, FC loop 171) side is performed (step 808).

  As described above, in the storage device of this embodiment, in the information processing system such as the storage device, the PBCs 121 to 126 are connected to the FC loop 161 and the FC loop 171 in which the controllers 101 and 102 and the drives 111 to 114 are provided in a multiplexed manner. In addition, each of the FC loops 161 and 171 is provided with PBC 141 to 143 and PBC 151 to 153 for short-circuiting the FC loop in the configuration connected via the PBC 131 to 136, and the failure of the bypass operation switching control by these PBCs The position is specified, the part is cut off, and the FC loop of the fiber channel is reconfigured.

  As a result, in the storage device according to the present embodiment, a healthy part (FC loop 161a) that can be communicated is used as it is even in the FC loop in which a failure occurs, and only the part that cannot communicate (FC loop 161b) is used for other FCs. By switching to the loop, it is possible to maintain the multiplex state of the FC loop partially, but to minimize degradation of performance and reliability.

  Further, in the storage device of the present embodiment, each controller can perform PBC switching control in the order of distance from the controller, so that not only the failure of the drive but also the failure position of the FC loop can be automatically detected. Maintenance management work such as recovery operations can be performed quickly and accurately.

  As described above, when a plurality of FC loops 161 and 171 are used and PBC switching control is performed from the other healthy FC loop side at the time of a loop failure, the two FC loops are simultaneously controlled. When a failure occurs, recovery becomes difficult.

  Therefore, as a modification of the storage device of the present embodiment, an example of means for recovering communication of the FC loop when two faults occur at the same time will be described below.

  In the storage device having the structure of FIG. 1 having two FC loops, a loop fault that occurs in only one of the FC loops by the method described above can be dealt with by switching to the other FC loop. However, the controller and the drive are connected to both FC loops, and a failure of the controller and the drive may cause both FC loops to fail.

  In this case, in the storage device configured as shown in FIG. 1, if one FC loop is communicable, it is possible to control the PBC of the failed FC loop from that FC loop to recover the communication. When the FC loop becomes incapable of communication, the PBC cannot be controlled, so that the communication inability cannot be recovered.

  Therefore, as a countermeasure for this technical problem, in the modification of the storage device of the present embodiment, the PBC is not controlled by communication of the FC loop, but each controller and each PBC are connected by another signal line. The PBC is controlled by signal line communication. FIG. 10 shows a configuration of a modification of the storage device according to the present embodiment.

  In the configuration of the modified example of the storage device of this embodiment in FIG. 10, a signal line 501 and a signal line 502 that connect each controller and each PBC are added to the configuration illustrated in FIG. 1 described above. A recovery method when a failure occurs in two FC loops simultaneously is shown below. As an example, consider a case where both FC loops cannot communicate due to a failure of the drive 113.

  In the configuration of the modification example of the storage device of this embodiment in FIG. 10, the recovery of the FC loop 161 is performed by the individual controllers 101 and 102 switching the individual PBC from the far side from the controller via the signal line 501. This is done by searching for a place where communication is possible.

  Actually, the individual controllers 101 and 102 perform an operation of switching the PBCs 126, 143, 125, 142, 124, 141, and 123 through the signal line 501 in order from the controller in the FC loop. In the case of the failure of the part of the drive 113 as described above, the communication of the FC loop 161 is restored when the PBC 125 is switched.

  Similarly, the recovery of the FC loop 171 is also performed by the individual controllers 101 and 102 switching the individual PBC from the far side from the controller via the signal line 502 and searching for a place where communication is possible.

  Actually, the individual controllers 101 and 102 perform an operation of switching the PBC 136, PBC 153, PBC 135, PBC 152, PBC 134, PBC 151, and PBC 133 in order from the controller in the FC loop via the signal line 502. As described above, the failure of the part of the drive 113 causes the FC loop communication to recover when the PBC 135 is switched.

  As described above, in the configuration of the modification example of the storage device of this embodiment illustrated in FIG. 10, the operations of the controller, the drive, and the plurality of PBCs that bypass and separate the FC loop are separated from the multiple FC loops. Since the controller controls the signal line 501 and the signal line 502 provided in the communication line, even if all the communication means on the FC loop become incapable of communication, it is possible to recover the failure by separating the faulty part. Is possible.

  As mentioned above, the invention made by the present inventor has been specifically described based on the embodiment. However, the present invention is not limited to the embodiment, and various modifications can be made without departing from the scope of the invention. Needless to say.

  For example, in the above description, the case where the present invention is applied to a storage device has been described as an example of the information processing system. However, the information processing system is widely applied to general information processing systems connected by loop communication means such as FC_AL. Can do.

  The fault location display method is not limited to a method using a display lamp or the like. For example, a system configuration diagram as shown in FIG. 1 is displayed on the screen of a control terminal that controls the controller from the outside, and is visualized on the system configuration diagram. The method of displaying the

  As mentioned above, the invention made by the present inventor has been specifically described based on the embodiments. However, the present invention is not limited to the above-described embodiments, and various modifications can be made without departing from the scope of the invention. Needless to say.

It is the conceptual diagram which showed an example of the structure of FC loop of the memory | storage device which is one Embodiment of the information processing system of this invention, and a memory | storage device. It is a conceptual diagram which shows an example of the whole structure of the memory | storage device which is one embodiment of this invention. It is a block diagram which shows an example of a structure of the port bypass circuit (PBC) used for the connection of an apparatus and FC loop in the memory | storage device which is one embodiment of this invention. It is a block diagram which shows an example of a structure of the port bypass circuit (PBC) used for the short circuit of FC loop in the memory | storage device which is one embodiment of this invention. It is a flowchart which shows an example of the detection operation | movement of the fault site | part in the memory | storage device which is one embodiment of this invention. It is a perspective view which shows the example of mounting of the memory | storage device which is one embodiment of this invention. It is a conceptual diagram which shows an example of isolation | separation operation | movement from the FC loop of the drive in the memory | storage device which is one embodiment of this invention, and a controller. It is a conceptual diagram which shows an example of the effect | action of partial separation operation | movement of FC loop in the memory | storage device which is one embodiment of this invention. (A)-(c) is a conceptual diagram which shows an example of the effect | action of partial disconnection operation | movement of the FC loop at the time of FC loop failure occurrence in the memory | storage device which is one embodiment of this invention. It is a conceptual diagram which shows the modification of the control method of the port bypass circuit (PBC) in the memory | storage device which is one embodiment of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 ... Mounting board, 11 ... Connector, 101 ... Controller (unit component), 102 ... Controller (unit component), 103 ... Cache memory, 104 ... Host interface, 105 ... Host interface, 111-114 ... Drive (unit storage) Device, unit component), 121-126 ... PBC (first bypass means), 131-136 ... PBC (first bypass means), 141-143 ... PBC (second bypass means), 151-153 ... PBC (second bypass means), 161 ... FC loop (loop communication means), 171 ... FC loop (loop communication means), 181 ... loop failure location, 201 ... connection state, 202 ... disconnection state, 301 ... connection State 302: Short circuit state 501 Signal line (control line) 502 ... Signal line (control line) 6 0 ... PBC, 601 ... selector, 602 ... bypass control signal, 603 ... fault monitoring circuit, 604 ... display lamp, 700 ... PBC, 701 ... selector, 702 ... bypass control signal, 703 ... display lamp.

Claims (5)

A storage device including a plurality of disk drives,
The storage device includes: a Fiber Channel loop connecting the plurality of disk drives; a port bypass circuit for bypassing the Fiber Channel loop; and a controller for controlling the plurality of disk drives and the port bypass circuit. Including
The controller monitors the occurrence of a failure in the storage device and, when detecting a failure in the Fiber Channel loop, performs switching control of the port bypass circuit, disconnects the failure position of the Fiber Channel loop, A storage device that controls the plurality of disk drives by using a portion in which no failure has occurred.   2. The storage device according to claim 1, wherein the failure monitoring is a failure caused by any of the plurality of disk drives or a failure caused by the Fiber Channel loop. The storage device according to claim 1, wherein the storage device is a process for determining whether or not. A storage device including a plurality of disk drives,
The storage device controls a fiber channel loop connecting the plurality of disk drives, a plurality of port bypass circuits for bypassing the fiber channel loop, and the plurality of disk drives and the plurality of port bypass circuits. Controller and
The controller monitors the occurrence of a failure in the storage device and, when detecting a failure in the Fiber Channel loop, performs a process of specifying the failure position of the Fiber Channel loop,
In accordance with the identified failure position, the switching control of the plurality of port bypass circuits is performed, the failure position of the Fiber Channel loop is disconnected, and the plurality of the failure points of the Fiber Channel loop are used. A storage device that controls a disk drive. The storage device according to claim 3,
The process of identifying the failure position of the Fiber Channel loop is a bypass operation in order from the port bypass circuit far from the controller among the plurality of port bypass circuits, and whether or not the failure has been recovered by the bypass operation. The storage device according to claim 3, wherein the storage device is checked. The storage device according to claim 3,
The failure monitoring is a process for determining whether a failure has occurred due to any of the plurality of disk drives or a failure has occurred due to the Fiber Channel loop. The storage device according to claim 3, wherein the storage device is provided.

Images