Storage system and communications path control method for storage system

• US 7,451,341 B2
• Filed: 11/23/2004
• Issued: 11/11/2008
• Est. Priority Date: 10/06/2004
• Status: Expired due to Fees

• Current Assignee: Hitachi, Ltd.
• Original Assignee: Hitachi, Ltd.
• Inventors: Tanabe, Masayuki, Okaki, Takuya, Dohi, Hironori, Nakamura, Hirofumi
• Primary Examiner(s): Chu; Gabriel L
• Application Number:
  US10/994,339
• Publication Number:
  US 20060075156A1
• Time in Patent Office: 1,449 Days
• Field of Search: 714/4
• US Class Current: 714/4.1
• CPC Class Codes: G06F 11/201 between storage system components

Abstract

A method for controlling communications paths in a storage system provided between a storage device and a plurality of host devices. The method includes: respectively monitoring communications statuses between each of the host devices and the storage device; detecting whether a prescribed temporary failure, which can be corrected without switching communications paths, has occurred, based upon the detected communications statuses; a switching instruction step of not outputting a switching request signal for requesting switching from a failed communications path to a to-be-used communications path, before a number of occurrence of the prescribed temporary failure thus detected reaches a threshold value which is equal to or larger than 2, and outputting the switching request signal only when the number of occurrence of the prescribed temporary failure reaches the threshold value; and selecting the to-be-used communications path, from among the plurality of communications paths, on the basis of the switching request signal.

Claims

1. A storage system comprising:

a plurality of host devices;

a storage device connectable respectively to the host devices by means of a plurality of communications paths;

a communications status monitoring section which respectively monitors communications statuses between each of said host devices and said storage device;

a failure detecting section which detects whether or not each of a plurality types of prescribed temporary failures, each of which is able to be corrected without switching communications paths, has occurred, on the basis of at least said respective communications statuses detected by said communications status monitoring section, different threshold values of 2 or more being set for said different types of temporary failures respectively;

a switching instruction section which does not output a switching request signal for requesting switching from a failed communications path to a to-be-used communications path before a number of occurrence of at least one type of said prescribed temporary failure detected by said failure detecting section reaches a respective threshold value, and outputs said switching request signal only when and as soon as determining said number of occurrence of said at least one type of said prescribed temporary failure reaches said respective threshold value; and

a path control section which selects the to-be-used communications path, on the basis of said switching request signal from said switching instruction section.

2. The storage system according to claim 1, wherein said host devices share communications ports of said storage device, and said failure detecting section detects whether or not said prescribed temporary failure has occurred with respect to communications ports that are shared by said host devices.

3. The storage system according to claim 1, wherein said prescribed temporary failure include at least a communications time-out status.

4. The storage system according to claim 1, wherein said failure detecting section judges that said prescribed temporary failure has occurred in cases where a prescribed command has been issued by one of said host devices, or cases where a prescribed status has been detected inside said storage device.

5. The storage system according to claim 1, wherein said failure detecting section judges that said prescribed temporary failure has occurred, in cases where at least one of a processing abort command and a reset command has been issued by one of said host devices, or cases where at least one of a communications time-out status and an abnormal status of the communications signal is detected in said storage device.

6. The storage system according to claim 1, wherein said switching instruction section outputs said switching request signal in response to access from one of said host devices.

7. The storage system according to claim 1, wherein said switching instruction section outputs said switching request signal in response to a write or read access from one of said host devices, and does not output said switching request signal in response to a status enquiry access from said one host device.

8. The storage system according to claim 1, wherein said switching request signal is a signal that only affects one of the host devices related to said prescribed temporary failure.

9. The storage system according to claim 1, wherein said switching request signal is a hardware error response.

10. The storage system according to claim 1, wherein said communications status monitoring section, said failure detecting section and said switching instruction section are provided within said storage device, and said path control section is provided within said host device.

11. The storage system according to claim 1, wherein said path control section is provided in a relay device disposed between said host device and said storage device.

12. The storage system according to claim 1, wherein said failure detecting section is provided in either said host device or said relay device.

13. The storage system according to claim 1, wherein threshold values are previously established respectively in association with each type of said prescribed temporary failure, and

said switching instruction section detects a number of occurrences respectively for each type of said prescribed temporary failure and outputs said switching request signal, if the number of occurrences for any one prescribed temporary failure of these types has reached a corresponding threshold value.

14. The storage system according to claim 1, wherein, when detecting said switching request signal, said path control section judges whether or not there exists an available communications path, and when judging that an available communications path exists, then said path control section selects that available communications path.

15. The storage system according to claim 1, wherein, when detecting said switching request signal, then said path control section judges whether or not there exists an original communications path from which communications were switched on a previous occasion that said switching request signal was detected, and when judging that there exists an original communications path from which communications were switched due to detection of said switching request signal, then said path control section selects that original communications path.

16. The storage system according to claim 1, wherein said types of temporary failure include an abort request issued by one of said host devices, a reset request issued by one of said host devices, and an abnormality in a communications signal.

17. The storage system according to claim 16, wherein said abnormality in a communications signal includes missing a frame number or error detection information therefrom.

18. A control method for controlling communications paths in a storage system provided with a storage device connectable respectively to a plurality of host devices by means of a plurality of communications paths, in which the communications ports of said storage device can be shared by said respective host devices, comprising:

a monitoring step of respectively monitoring communications statuses between each of said host devices and said storage device;

a failure detecting step of detecting whether or not each of a plurality types of prescribed temporary failures, each of which is to be corrected without switching communications paths, has occurred, on the basis of said respective communications statuses thus detected, different threshold values of 2 or more being set for said different types of temporary failures respectively;

a switching instruction step of not outputting a switching request signal for requesting switching from a failed communications path to a to-be-used communications path before a number of occurrence of at least one type of said prescribed temporary failure detected by said failure detecting section reaches a respective threshold value, and outputting said switching request signal only when and as soon as determining said number of occurrence of said at least one type of said prescribed temporary failure reaches said respective threshold value; and

a path selecting step of selecting the to-be-used communications path, from among said plurality of communications paths, on the basis of said switching request signal.

19. The communications path control method for a storage system according to claim 18, wherein, when the access from one of said host devices to said storage device is either a write access or a read access, the step of outputting said switching request signal outputs said switching request signal to said one host device originating said access.

20. The communications path control method for a storage system according to claim 18, wherein a relay device for relaying communications between said respective host devices and said storage device is further provided, and said path selecting step selects said to-be-used communications path either within one of said host devices, or within said relay device.