1. Field of the Invention
The present invention generally relates to identification information diagnosis methods and input and output apparatuses, and more particularly to an identification information diagnosis method for diagnosing identification information of a storage unit that is implemented in an input and output apparatus, and to an input and output apparatus that employs such an identification information diagnosis method.
2. Description of the Related Art
Conventionally, there is an input and output apparatus that has a plurality of device enclosures mounted on a rack, where a plurality of Hard Disk Drives (HDDs) are implemented in each device enclosure. In such an input and output apparatus, the HDDs are inserted into a plurality of slots that are provided in each device enclosure, and a different slot value is set for each HDD. An Arbitrated Loop Physical Address (AL_PA) map is created based on an ID of the device enclosure and the slot value that is read from the HDD when the power is turned ON, for example. The ID of the device enclosure is identification information unique to and indicating each device enclosure. In addition, the slot value is identification information unique to and indicating the position of each slot within each device enclosure. When an access request is received from a host unit, an access is made to a corresponding HDD by referring to the AL_PA map. By using the AL_PA map, it is possible to know the slot and the device enclosure where the corresponding HDD requested by the access request exists.
Within one device enclosure, the slot values of the HDDs that are inserted into the slots are unique. However, among different device enclosures, the slot values of the HDDs that are inserted into the slots may not necessarily be unique. For this reason, every time the power is turned ON, for example, the AL_PA map is created by acquiring slot IDs from the ID of the device enclosure and the slot values of the HDDs that are implemented in this device enclosure. When an access request is received from the host unit, an access is made to a corresponding HDD, that is, a unique address (physical location) corresponding to the HDD, by referring to the AL_PA map.
For example, a Japanese Laid-Open Patent Application No.2004-220265 proposes a type of the input and output apparatus described above.
From a back panel of the device enclosure, a slot position signal corresponding to the slot value is supplied via a connector to the HDD that is inserted into the corresponding slot. When creating the AL_PA map described above, the slot values are read from the HDDs. For this reason, if foreign particles and the like are adhered on the connector and at least a portion (for example, 1 bit) of the slot position signal is blocked by the foreign particles and the like and not supplied to the HDD, a correct slot value cannot be read from the HDD and it is impossible to create a correct AL_PA map that is originally intended. When the correct AL_PA map cannot be created, there was a problem in that it is impossible to correctly access the HDD that is originally intended.
In addition, if the correct slot value cannot be read from the HDD, there is a possibility that the same slot value will be read from two different HDDs that are implemented within one device enclosure, for example. In this case, the HDD that is inserted into the slot where the abnormality is actually generated should be disconnected from the device enclosure, but it is impossible to judge the HDD from which the erroneous slot value is read. Consequently, both the HDDs from which the same slot value is read must be disconnected from the device enclosure, and there was a problem in that it is difficult to efficiently utilize the resources (storage units such as the HDDs).
Accordingly, it is a general object of the present invention to provide a novel and useful identification information diagnosis method and input and output apparatus, in which the problems described above are suppressed.
Another and more specific object of the present invention is to provide an identification information diagnosis method and an input and output apparatus, which can specify a storage unit, such as an HDD, from which an erroneous identification information is read, so as to enable a normal access to the storage unit.
Still another object of the present invention is to provide an identification information diagnosis method for an input and output apparatus in which first identification information unique to each of storage units is written in each of the storage units that is inserted into a corresponding one of slots of each of device enclosures, and a physical address of each of the storage units is obtained from second identification information unique to each of the device enclosures and the first identification information that is read from each of the storage units of a corresponding one of the device enclosures and stored in the corresponding one of the storage units, comprising the steps of (a) obtaining a regular physical address from first and second identification information prestored in a memory; (b) reading the first identification information from one of the storage units that is accessed by the regular physical address; and (c) diagnosing the first identification information read from the one of the storage units that is accessed by the regular physical address, based on a diagnosis table that stores the regular physical address and a corresponding regular first identification information. According to the identification information diagnosis method of the present invention, it is possible to specify the storage unit, such as an HDD, from which an erroneous identification information is read, so as to enable a normal access to the storage unit.
A further object of the present invention is to provide an input and output apparatus in which first identification information unique to each of storage units is written in each of the storage units that is inserted into a corresponding one of slots of each of device enclosures, and a physical address of each of the storage units is obtained from second identification information unique to each of the device enclosures and the first identification information that is read from each of the storage units of a corresponding one of the device enclosures and stored in the corresponding one of the storage units, comprising an obtaining part configured to obtain a regular physical address from first and second identification information prestored in a memory; a reading part configured to read the first identification information from one of the storage units that is accessed by the regular physical address; and a diagnosing part configured to diagnose the first identification information read from the one of the storage units that is accessed by the regular physical address, based on a diagnosis table that stores the regular physical address and a corresponding regular first identification information. According to the input and output apparatus of the present invention, it is possible to specify the storage unit, such as an HDD, from which an erroneous identification information is read, so as to enable a normal access to the storage unit.
Other objects and further features of the present invention will be apparent from the following detailed description when read in conjunction with the accompanying drawings.
A description will be given of embodiments of an identification information diagnosis method and an input and output apparatus according to the present invention, by referring to the drawings.
An input and output apparatus 1 includes a RAID controller 2 and a plurality of device enclosures 3 that are mounted on a rack (not shown), but for the sake of convenience, only one device enclosure 3 is shown in
The RAID controller 2 carries out a known RAID control, and centrally controls each of the device enclosures 3. The RAID controller 2 has a known structure including a Channel Adaptor (CA), a Centralized Module (CM), a Device Adaptor (DA), a Router (RT) and a Service Controller (SVC). The CA is formed by a module that controls an external interface with respect to a host unit. The CM is formed by a module that manages all operations within the corresponding device enclosures 3, and is provided with a cache memory. The DA is formed by a module that controls an external interface with respect to a subordinate apparatuses, that is, the physical devices (HDDs). The module forming the DA is provided with a 2-port Fiber Channel Arbitrated Loop (FCAL) interface. The RT is formed by a module that forms a communication interface with respect to other RAID controllers.
The HDD controller 31 is formed by a Port Bypass Circuit (PBC) that is made of a switch for Fiber Channel drive connection. The HDD controller 31 includes a MPU 301 that is provided with a memory, register or the like 302 (hereinafter simply referred to as a memory 302) for storing tables and the like which will be described later, a Fiber Channel Controller (FCC) 303, and a Switch-Chip (SW-CHIP) 304. The MPU 301 controls the entire operation of the corresponding device enclosure 3. The FCC 303 monitors the HDDs 41-1 through 41-N within the corresponding device enclosure 3, and is capable of executing a Fiber Channel (FC) command with respect to each of the HDDs 41-1 through 41-N. The SW-CHIP 304 is formed by a IC chip having a Fiber Channel Arbitrated Loop (FCAL) switching function and an AL_PA value acquiring function, and for example, chips SOC312, 320 and 422 manufactured by Emulex and a chip PM8375 manufactured by PMC-SIERRA may be used for the SW-CHIP 304.
Registers 34-1 through 34-N are provided in the pack panel 32 in correspondence with each of the slots. The registers 34-1 through 34-N respectively store slot values (first identification information) that form slot IDs of the corresponding HDDs 41-1 through 41-N. The slot values are fixed with respect to the HDDs 41-1 through 41-N, and each form lower 4 bits of the slot ID, for example. Device enclosure IDs (second identification information) are unique to each of the device enclosures 3. The device enclosure ID is prestored in the memory 302 within the MPU 301 of the HDD controller 31, together with the slot values of the HDDs 41-1 through 41-N that are preset in the registers 34-1 through 34-N of the back panel 32. The device enclosure ID forms upper 3 bits of the slot ID, for example.
The RAID controller 2 and the MPU 301 within the HDD controller 31 are connected via a control interface. The RAID controller 2 and the SW-CHIP 304 within the HDD controller 31 are connected via a Fiber Channel Interface (FCI). Within the HDD controller 31, the MPU 301 and the FCC are connected via a control interface, and the MPU 301 and the SW-CHIP 304 are connected via a control interface. The FCC 303 and the SW-CHIP 304 within the HDD controller 31 are connected via a Fiber Channel Interface (FCI). In addition, the SW-CHIP 304 within the HDD controller 31 and the HDDs 41-1 through 41-N are connected via Fiber Channel Interfaces (FCIs), and the slot values can be read out to the SW-CHIP 304 via the FCIs. Moreover, the MPU 301 and the HDDs 41-1 through 41-N are connected via device enclosure ID setting interfaces, and the device enclosure ID is written to the HDDs 41-1 through 41-N via the device enclosure ID setting interfaces when the power of the device enclosure 3 is turned ON, for example.
By using the upper 3 bits within the 7 bits representing the slot ID of each HDD 41 to set the device enclosure ID that is unique to each device enclosure 3 when the power is turned ON, for example, and storing the device enclosure ID in the memory 302 within the MPU 301, it becomes possible to thereafter read the 7-bit slot ID of each HDD 41, obtain the AL_PA values by referring to tables which will be described later to set the AL_PA values in a fiber channel map FC-MAP, and access the HDD 41 requested by the access request that is received from the host unit via the RAID controller 2 based on the fiber channel map FC-MAP.
DE00→000xxxx
DE01→001xxxx
. . .
A step ST4 creates a table of fiber channel maps FC-MAP (hereinafter simply referred to as an FC-MAP table) shown in
Accordingly, after the value that is unique for each device enclosure 3 is set in the upper 3 bits of the 7-bit slot ID of each of the HDDs 41 within each device enclosure 3 when the power of the input and output apparatus 1 is turned ON, for example, it becomes possible to create the FC-MAP table shown in
A step ST13 uses the AL_PA value acquiring function of the SW-CHIP 304, and obtains (or calculates) the corresponding AL_PA value by referring to the AL_PA table shown in
A step ST14 decides whether or not an overlap of the AL_PA value exists. In other words, the step ST14 decides whether or not the AL_PA value obtained in the step ST13 overlaps (or coincides) with the AL_PA value that is obtained with respect to another HDD 41 within the same device enclosure 3. If the decision result in the step ST14 is YES, there is an overlap of the AL_PA value, and thus, a step ST15 changes the AL_PA value to an AL_PA value unique to the corresponding HDD 41, and the process advances to a step ST16. On the other hand, if the decision result in the step ST14 is NO, the process advances directly to the step ST16.
The step ST16 decides whether or not the steps ST11 through ST15 have been carried out with respect to the slot IDs of all of the HDDs 41 within the device enclosure 3 (DE00). If the decision result in the step ST16 is YES, a step ST17 creates the FC-MAP table shown in
Therefore, it is possible to obtain (or calculate) the AL_PA value from the cell table and the AL_PA table shown in
Accordingly, when the RAID controller 2 receives from the host unit the access request specifying the slot SLOT00 of the device enclosure DE00 or the like, it is possible to access the requested HDD 41 by referring to the FC-MAP table shown in
Next, a description will be given of a slot ID diagnosis operation. The slot ID diagnosis operation may be realized by a self-diagnosis carried out in each device enclosure 3 or, by an apparatus diagnosis carried out by the RAID controller 2.
The diagnosis operation shown in
In a step S4, the MPU 301 obtains (or calculates) the regular AL_PA value of each of the HDDs 41-1 through 41-N that are implemented in the corresponding device enclosure 3 by using the AL_PA acquiring function of the SW-CHIP 304, based on the regular slot IDs obtained in the step S3. The regular AL_PA value may be calculated based on a predetermined algorithm or, obtained using the tables described above in conjunction with
In a step S6, the MPU 301 issues a command to each of the HDDs 41-1 through 41-N via the SW-CHIP 304 and the Fiber Channel Interface (FCI) using a SCSI Enclosure Service (SES) command, and reads the slot IDs from each of the HDDs 41-1 through 41-N that are accessed by each of the regular AL_PA values stored in the memory 302. In a step S7, the MPU 301 refers to a diagnosis table similar to the FC-MAP table shown in
In a step S9, the HDD 41 that has been confirmed by the MPU 301 as not being inserted in a normal manner into the corresponding slot is disconnected from the corresponding device enclosure 3, and the MPU 301 notifies the abnormality specified in the step S8 to the RAID controller 2 via the control interface. In a step S10, the MPU 301 decides whether or not the diagnosis has been completed with respect to all of the HDDs 41-1 through 41-N within the corresponding device enclosure 3. The process returns to the step S3 if the decision result in the step S10 is NO. But if the decision result in the step S10 is YES, the external port that was bypassed last is restored and the process ends.
The diagnosis operation shown in
In a step S22, the RAID controller 2 causes the MPU 301 within the corresponding device enclosure 3 to obtain (or calculate) the regular AL_PA value of each of the HDDs 41-1 through 41-N that are implemented in the corresponding device enclosure 3 by using the AL_PA acquiring function of the SW-CHIP 304, based on the regular slot IDs obtained in the step S21. The regular AL_PA value may be calculated based on a predetermined algorithm or, obtained using the tables described above in conjunction with
In a step S24, the RAID controller 2 causes the MPU 301 to issue a command to each of the HDDs 41-1 through 41-N via the SW-CHIP 304 and the Fiber Channel Interface (FCI) using a SCSI Enclosure Service (SES) command, and to read the slot IDs from each of the HDDs 41-1 through 41-N that are accessed by each of the regular AL_PA values stored in the memory within the RAID controller 2. In a step S25, the RAID controller 2 refers to a diagnosis table similar to the FC-MAP table shown in
In a step S27, the HDD 41 that has been confirmed by the RAID controller 2 as not being inserted in a normal manner into the corresponding slot is disconnected from the corresponding device enclosure 3. In a step S28, the RAID controller 2 decides whether or not the diagnosis has been completed with respect to all of the HDDs 41-1 through 41-N within the corresponding device enclosure 3 and the diagnosis has been completed with respect to all of the device enclosures 3 within the input and output apparatus 1. The process returns to the step S21 if the decision result in the step S28 is NO. But if the decision result in the step S28 is YES, the process ends.
Therefore, when an abnormal slot ID is detected by the diagnosis operation of the device enclosure 3 or the RAID controller 2, the HDD 41 written with the abnormal slot ID is disconnected from the slot in the back panel 32 of the corresponding device enclosure 3, and the starting of the input and output apparatus 1 is completed thereafter. Since the HDD 41 implemented in each device enclosure 3 and written with the abnormal slot ID is disconnected from the corresponding device enclosure 3 before the starting of the input and output apparatus 1 is completed, the abnormality of the slot ID is not visible from the RAID controller 2.
Of course, at least a portion of the various kinds of tables, the slot values, the device enclosure IDs and the AL_PA values may be stored in a memory within the SW-CHIP 304, instead of within the memory 302 within the MPU 301.
Therefore, the present invention is suited for use in an input and output apparatus which uses a large number of storage units such as HDDs, when specifying the storage unit such as the HDD from which an erroneous slot value is read.
This application claims the benefit of a Japanese Patent Application No.2005-039469 filed Feb. 16, 2005, in the Japanese Patent Office, the disclosure of which is hereby incorporated by reference.
Further, the present invention is not limited to these embodiments, but various variations and modifications may be made without departing from the scope of the present invention.
Number | Date | Country | Kind |
---|---|---|---|
2005-039469 | Feb 2005 | JP | national |