Fault recovery method in a system having a plurality of storage systems

• US 7,337,353 B2
• Filed: 06/29/2004
• Issued: 02/26/2008
• Est. Priority Date: 05/12/2004
• Status: Expired due to Fees

• Current Assignee: Hitachi, Ltd.
• Original Assignee: Hitachi, Ltd.
• Inventors: Yamamoto, Yasutomo, Eguchi, Yoshiaki
• Primary Examiner(s): Chu; Gabriel
• Application Number:
  US10/878,440
• Publication Number:
  US 20050268147A1
• Time in Patent Office: 1,337 Days
• Field of Search: 714/7, 714/47
• US Class Current: 714/47.2
• CPC Class Codes: G06F 11/0727 in a storage system, e.g. in a DASD or network based storage system drivers for digital recording or reproducing units G06F3/06; circuits for error detection or correction within digital recording or reproducing units G11B20/18; for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS], H04L67/1097
• CPC Class Codes: G06F 11/0766 Error or fault reporting or storing
• CPC Class Codes: G06F 11/2094 Redundant storage or storage space G06F11/2056 takes precedence
• CPC Class Codes: G06F 3/0617 in relation to availability
• CPC Class Codes: G06F 3/0635 by changing the path, e.g. traffic rerouting, path reconfiguration
• CPC Class Codes: G06F 3/0647 Migration mechanisms
• CPC Class Codes: G06F 3/067 Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]

Abstract

System availability is improved in a second storage system, connected to a first storage system, and having means for virtualizing devices within the first storage system as its own devices. When the virtual storage system or a storage management server detects a fault in the virtual storage system, the management server investigates the range affected by the fault, identifies a device for which measures must be taken, determines a transfer target device which accommodates the performance, reliability, and other attributes of the affected device, and issues a device transfer instruction for the virtual storage system. In the virtual storage system, the data of the device specified by the instruction within the virtual storage system is transferred to a device, specified by the management server, within the system itself, or to a device within another virtual storage system.

Claims

1. A management server, managing a second storage system which provides a host computer with access to both a second physical device in the second storage system and a first logical device configuring a parity group in a first storage system coupled to the second storage system, comprising:

transfer source decision means for identifying, based on information received from said second storage system prognosticating a fault or detecting a fault in said first logical device, a storage area affected by said fault as a transfer source; and

data transfer instruction means for selecting, based on data capacity and evaluation of performance and reliability levels determined in advance of transfer of data of said transfer source, a storage area of a transfer target from among storage areas of said first storage system and said second storage system managed by said second storage system, and for issuing to said second storage system an instruction to configure said parity group with said transfer target and said first logical device except said transfer source and to transfer the data of said transfer source to said transfer target,

wherein if said data transfer instruction means selects a storage area of a transfer target, said first logical device configures said parity group with said transfer target in said second storage system and said first logical device except said transfer source in said first storage system,

wherein said data transfer instruction means selects a first device, satisfying both the conditions of having the data capacity of said transfer source and of being evaluated as meeting said predetermined performance and reliability levels, from among the storage areas of said first storage system or of said second storage system managed by said second storage system, and determines said first device to be the transfer target storage area; and

when no first device exists which satisfies both of said conditions within the storage areas of said first storage system or said second storage system managed by said second storage system, a second device which at least satisfies said data capacity condition is selected, and said second device is determined to be the transfer target storage area.

2. The management server according to claim 1,

wherein when no second device exists which satisfies at least said data capacity condition within storage areas of said first storage system or said second storage system managed by said second storage system, a third device is selected which satisfies both said data capacity condition and said performance and reliability level evaluation condition from among the storage area of another storage system being managed by said management server, said third device is determined as the transfer target storage area, and an instruction is issued to said second storage system to register the third device selected as the transfer target as an external device providing access to the host computer;

when no third device exists which satisfies both of said conditions within the storage areas of said other storage systems, a fourth device which satisfies at least said data capacity condition is selected from among the storage areas of said another storage systems, the fourth device is determined as the transfer target, and an Instruction is issued to said second storage system to register the fourth device selected as the transfer target as an external device of said second storage system; and

,

when no fourth device exists which satisfies at least said data capacity condition within the storage areas of said another storage systems, information indicating that data transfer is not possible is output.

3. A storage system, comprising:

physical devices installed within said storage system;

managing as logical devices both external devices in an external storage system coupled to and separate front said storage system and said physical devices in said storage system,

logical device management means for allocating logical devices to said physical devices and said external devices and managing both said physical devices and said external devices with respect to device configuration information in association wit said logical devices provided to a host;

input/output processing means for performing processing, when an input/output processing request for a logical device is received from said host to convert the received input/output request into input/output processing for said physical device or said external device based on said device configuration information, and to perform input/output processing according to said conversion result;

anomaly detection means for judging the occurrence, during input/output processing for an accessed one of said external devices by said input/output processing means, of an access fault or performance decline forte accessed external device;

notification means for issuing, when an access fault or decline in performance has been detected by said anomaly detection means, notification of the detected anomaly and the accessed external device to a management server which manages said storage system and said external storage system; and

,

data transfer means for performing, upon receipt of an instruction from said management server for data transfer specifying the transfer source and transfer target, the transfer of the data of said transfer source to said transfer target according to the instruction; and

wherein said logical device management means updates said device configuration information after the completion of said data transfer.

4. The storage system according to claim 3, further comprising external device monitoring means for determining said external devices to be accessed at predetermined periods, and for executing input/output processing for said external devices.

5. A storage system, comprising:

a first storage system having physical devices; and

a second storage system having physical devices, coupled to and separate from said first storage system and further coupled to a computer,

wherein said first storage system comprises external device provision means for providing said physical devices in said first storage system to said second storage system as external devices with respect to said first storage system; and

, said second storage system comprises;

logical device management means for allocating logical devices to said physical devices in said second storage system and said external devices provided by said external device provision means, and managing both said physical devices installed within said second storage system and said external devices with respect to device configuration information associated with said logical devices accessed by a host computer;

input/output processing means for performing processing, when an input/output request is received from the host computer, to convert the received input/output request into input/output processing for one of said physical devices in said second storage system or one of said external device based on said device configuration information, and to perform input/output processing according to said conversion result;

anomaly detection means for judging the occurrence, during input/output processing for an accessed one of said external devices by said input/output processing means, of an access fault or performance decline for the accessed external device;

notification means for issuing, when an access fault or decline in performance has been detected by said anomaly detection means, notification of the detected anomaly and the accessed external device to a management server which manages said second storage system and said first storage system; and

,

data transfer means for performing, upon receipt of an instruction from said management server for data transfer specifying the transfer source and transfer target, the transfer of the data of said transfer source to said transfer target according to the instruction; and

wherein said logical device management means updates said device configuration information after the completion of said data transfer.

6. A computer system, comprising:

at least one first storage system;

a second storage system having physical devices, coupled to and separate from said at least one first storage system; and

a management server coupled to said at least one first storage system and said second storage system, wherein said at least one first storage system comprises external device provision means for providing a physical device in said at least one first storage system to said second storage system as an external device with respect to said second storage system;

wherein said second storage system comprises;

logical device management means for allocating logical devices to said physical devices in said second storage system and said external devices provided by said external device provision means, and managing both said physical devices installed within said second storage system and said external devices with respect to device configuration information in association with said logical devices accessed by a host computer;

monitoring means for monitoring said external devices;

notification means for notifying said management server when prognostication of a fault in one of said external devices is detected by said monitoring means; and

data transfer means for transferring data according to a data transfer instruction received from said management server, and for updating said device configuration information according to the data transfer; and

wherein said management server comprises data transfer instruction means for selecting a data transfer range as a transfer source and for selecting a transfer target based on notification from said notification means, and for issuing a data transfer instruction to said second storage system, specifying the transfer range and transfer target and instructing transfer of the data of said data transfer range to said transfer target.

7. The computer system according to claim 6, wherein said second storage system further comprises:

input/output instruction means for performing processing, when an input/output request for a logical device is received from the host computer, to convert the received input/output request into input/output processing for one of said physical devices or one said external device based on said device configuration information, and to perform input/output processing according to said conversion result; and

anomaly detection means for judging the occurrence, during input/output processing for an accessed external device by said input/output processing means, of an access fault or performance decline forte accessed external device;

wherein said notification means issues, when an access fault or decline in performance has been detected by said anomaly detection means, notification of the detected anomaly and the accessed external device to the management server;

wherein said management server further comprises transfer source decision means for identifying, based on information received from said second storage system for said detected anomaly and said accessed external device, storage areas in which faults may occur due to the anomaly as the transfer source; and

,

wherein said data transfer instruction means selects the transfer target from among the storage areas of said at least one first storage system or said second storage system, based on the data capacity of said transfer source determined by said transfer source decision means and on evaluation of predetermined performance and reliability levels, and issues a data transfer instruction to said second storage system, specifying said transfer source and said transfer target.

8. The computer system according to claim 7,

wherein said monitoring means determines said external devices for accessing at predetermined intervals, and causes said input/output instruction means to execute input/output processing for said external devices.

9. The computer system according to claim 6,

wherein said monitoring means monitors the states of said external devices, and upon detecting a prognostication of a fault occurrence, said notification means notifies said management server of the details of the detected prognostication and the external device being monitored;

said management server further comprises analysis means for analyzing, upon receiving said notification from said notification means, the received notification details, and for determining as the transfer source the external device which would be affected upon occurrence of the fault contained in the notification; and

,

said data transfer instruction means confirms the capacity of the transfer source external device determined by said analysis means, selects a device as said transfer target from among the devices being managed by said second storage system, and other than the external device which was accessed when said fault prognostication occurred.

10. The computer system according to claim 9,

wherein external devices provided by said external device provision means are provided from RAID groups and said RAID groups are configured by physical devices of said at least one first storage system;

said notification means issues notifications of the occurrence of faults in addition to notification of said prognostications of fault occurrence; and

,

said analysis means, upon receiving notification of fault occurrence from said notification means, analyzes the received notification, identifies the RAID group to which the external device in which the fault has occurred belongs, and recovers the data of said external device in which the fault has occurred using data held by other devices constituting said identified RAID group, while also setting the recovered data as said transfer source.

11. The computer system according to claim 9, wherein said data transfer instruction means, upon determining the transfer target, confirms the performance and reliability level of the transfer source determined by said analysis means, and selects a transfer target which, in addition to satisfying said capacity condition, satisfies said performance and reliability level condition from among the devices being managed by said second storage system, and other than the device for which said fault prognostication occurred.

12. The computer system according to claim 11, further comprising a third storage system coupled to said management server,

wherein, if when determining the transfer target there is no transfer target device being managed by said second storage system which satisfies said capacity condition or said performance and reliability level condition, said data transfer instruction means selects a device from among said third storage system which satisfies at least said capacity condition as the transfer target, and issues an instruction to said second storage system to register as an external device the device of the third storage system as said transfer target in the second storage system.

13. A management server for managing a first storage system and a second storage system coupled to and separate from said first storage system, said first storage system having physical devices that are external devices with respect to the second storage system and, said second storage system having physical devices and receiving an access request for a logical device from a host computer and accessing one of said physical devices in said second storage system and an external device in said first system according to a received access request, said management server comprising:

a processor; and

a memory storing a program executed by said processor,

wherein, according to the program stored in said memory, said processor performs;

upon receiving notification from said second storage system indicating prognostication of a fault occurring in said external device, transfer sauce decision processing to identify as a transfer source a storage area affected by said fault based on said notification; and

data transfer instruction transmission processing to select as a data transfer target storage areas of said first storage system or said second storage system managed by said management server, based on the data capacity of said transfer source and predetermined performance and reliability level of the transfer source, and to transmit an instruction to said second storage system to transfer the data of said transfer source to said transfer target; and

wherein said data transfer instruction processing selects a first device, satisfying both the conditions of having the data capacity of said transfer source and meeting said predetermined performance and reliability levels, from among the storage areas of said first storage system or said second storage system and determines said first device to be the transfer target storage area; and

when no first device exists which satisfies both of said conditions within the storage areas of said first storage system or said second storage system, a second device which at least satisfies said data capacity condition is selected, and said second device is determined to be the transfer target.

14. A storage system, coupled to and separate from another storage system having physical devices that are external devices with respect to said storage system, which receives an access request to a logical device from a host computer and accesses one of said external devices in said another storage system associated with said logical device according to said received access request, comprising:

a plurality of physical devices, each associated with a logical device accessed by a host computer;

a processor coupled to said plurality of physical devices; and

a memory storing a program executed by said processor and device configuration information indicating an associative relation between said plurality of physical devices and said external devices with the logical devices accessed by a host computer,

wherein, in accordance with said program stored in said memory, said processor executes;

input/output processing, upon receiving an input/output request from the host computer, to convert the received input/output request into an input/output request for said physical device or said external device based on said device configuration information, and to perform input/output processing according to said conversion result;

anomaly detection processing to judge the occurrence of access faults or declines in performance for one of the external devices during input/output processing of said external device;

notification processing, when an access fault or decline in performance is detected in the one external device during said anomaly detection processing, to notify said storage system and a management server which manages said another storage system of the detected anomaly and the one external device being accessed;

data transfer processing, when a data transfer instruction specifying as a transfer source the one external device and a transfer target from among other external devices is received from said management server in response to a notification of detection of an access fault or decline in performance in the one external device, to perform data transfer according to said instruction; and

,

configuration information update processing to update said device configuration information after the completion of said data transfer with respect to said, physical devices, and said logical devices.

15. The storage system according to claim 14, wherein said processor further performs input/output processing for said external devices at predetermined intervals.

16. A fault avoidance and recovery method, in a computer system comprising at least one first storage system, a second storage system, and a management server coupled to said first storage system and to said second storage system, wherein;

said second storage system is coupled to and separate from said first storage system and said first storage system has physical devices that are external devices with respect to said second storage system;

wherein said second storage system has physical devices and comprises logical device management means allocating logical devices to said physical devices and said external devices and manages both said physical devices and said external devices with respect to device configuration information in association with logical devices provided to a host;

said method comprising the steps of;

said second storage system, upon receipt of an input/output request from said host, converting the received input/output request into an input/output request for said physical devices or said external devices, based on said device configuration information;

said second storage system performing input/output processing according to the conversion result;

when the device for said input/output processing is one of said external devices, said second storage system judging whether an access fault or decline in performance has occurred for said one external device during said input/output processing;

when, as a result of said judgment, said access fault or decline in performance is detected in said one external device, said second storage system notifies said management server of the detected access fault or decline in performance and the one external device being accessed;

said management server receiving said notification from said second storage system;

said management server specifying storage areas in which a fault may occur due to said access fault or decline in performance as a transfer source, based on information contained in the received notification;

said management server selecting a transfer target from among the storage areas of said first storage system or said second storage system connected to said management server, based on the data capacity of said transfer source and predetermined performance and reliability levels;

said management server issuing an instruction to said second storage system to transfer the data of said transfer source to said transfer target;

said second storage system, upon receiving the data transfer instruction from said management server specifying the transfer source and transfer target, performing data transfer to transfer the data of said transfer source to said transfer target according to said instruction; and

,

after the completion of said data transfer, said second storage system updating said device configuration information with respect to the association of said physical devices and said external devices with said logical devices.

17. The fault avoidance and recovery method according to claim 16,

wherein said external devices are part of RAID groups configured by external devices of said first storage system;

when an access fault or decline in performance is detected, said second storage system issues notification of the occurrence of the fault and the one external device in which the fault occurred to said management server; and

,

upon receiving notification from said second storage system of the occurrence of the fault and of the external device in which said fault has occurred, said management server recovers the data of the external device in which said fault has occurred, using the data held by other devices of the RAID group to which the external device in which said fault has occurred belongs, and selects the recovered data as said transfer source.