System for online SCSI drive repair utilizing detachable secondary I/O buses pigtailed to primary I/O bus wherein each secondary I/O bus has a length in excess of 100mm

• US 5,870,630 A
• Filed: 07/31/1995
• Issued: 02/09/1999
• Est. Priority Date: 07/31/1995
• Status: Expired due to Term

• Current Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Original Assignee: Hewlett-Packard Company (HP Inc.)
• Inventors: Kato, Rick A., Jones, David P., Reasoner, Kelly J.
• Primary Examiner(s): Lee, Thomas C.
• Assistant Examiner(s): Ton, David
• Application Number:
  US08/509,553
• Time in Patent Office: 1,289 Days
• Field of Search: 395/856, 395/182.05, 395/182.1, 395/182.04, 395/894, 361/685, 711/114, 711/159, 364/281
• US Class Current: 710/74
• CPC Class Codes: G06F 13/409 Mechanical coupling back panels H05K7/1438
• CPC Class Codes: G11B 33/128 of the plurality of recording/reproducing devices, e.g. disk drives, onto a chassis
• CPC Class Codes: G11B 33/142 by air cooling

Abstract

An online repairable multiple drive library computer system, and method of online drive repair for repairing same, wherein the method comprises the steps of detecting that a drive has failed, placing the failed drive in an offline state, removing power to the failed drive, disconnecting any cabling attached to the failed drive, replacing the failed drive, reconnecting cabling to the replaced drive, restoring power to the replaced drive, and placing the replaced drive in an online state. All of the steps occur while one or more good drives in the multiple drive library computer system remain online, and accessible to a host.

Claims

1. A multiple drive library computer system, comprising:

a) an I/O bus comprising;

i) a primary I/O bus; and

ii) a plurality of detachable secondary I/O buses pigtailed to said primary I/O bus;

wherein each of said plurality of secondary I/O buses has a length in excess of 100 mm, and the I/O bus as a whole, but for its length, is configured as a SCSI single-ended cable;

b) a plurality of SCSI storage drives, connected to said plurality of secondary I/O buses; and

c) a library controller, connected to said primary I/O bus, said controller comprising;

i) means to communicate with said SCSI storage drives; and

ii) means to communicate with a host computer via said I/O bus.

2. A library computer system as in claim 1, wherein said storage SCSI storage drives further comprise:

a) media inserting faces; and

b) electrical connection faces, opposite said media inserting faces, said SCSI storage drives being mounted within said library computer system such that their media inserting faces abut a robotic media inserter, and said plurality of secondary I/O buses are connected to said plurality of SCSI storage drives via said electrical connection faces.

3. A library computer system as in claim 1, wherein said SCSI storage drives are optical drives.

4. A library computer system as in claim 1, further comprising:

a) a plurality of power cables, connected to said SCSI storage drives;

b) means for enabling or disabling power to said SCSI storage drives; and

c) means for changing the states of said SCSI storage drives.

5. A library computer system as in claim 4, wherein said power enabling/disabling means and said state changing means comprise firmware elements.

6. A library computer system as in claim 4, wherein said power enabling/disabling means comprises one or more switches.

7. A library computer system as in claim 1, further comprising means for detecting when a SCSI storage drive has failed.

8. A library computer system as in claim 7, wherein each of said SCSI storage drives is associated with an LED, and said LEDs are electrically connected to said drive failure detecting means.

9. A library computer system as in claim 1, further comprising one or more modular housings in which one or more of said SCSI storage drives are mounted, each modular housing, and the SCSI storage drives mounted therein, being removable from said library computer system as a unit.

10. A library computer system as in claim 9, wherein each modular housing further comprises a cooling fan.

11. A library computer system as in claim 9, wherein one or two SCSI storage drives are mounted within each of said modular housings.

12. A library computer system as in claim 9, wherein the SCSI storage drives mounted within any modular housing in which two or more SCSI storage drives are mounted, are daisy-chained to a singular secondary I/O bus.

13. A library computer system as in claim 1, wherein each of said SCSI storage drives is connected to its own secondary I/O bus.

14. A library computer system as in claim 1, wherein said plurality of SCSI storage drives comprises at least one subset of two or more SCSI storage drives, said SCSI storage drives of said subset being daisy-chained to a singular secondary I/O bus.

15. A method of replacing one or more failed SCSI storage drives in a multiple drive library computer system, wherein each of the SCSI storage drives comprises an electrical connection face opposite a media-inserting face, wherein the SCSI storage drives of said library computer system are mounted in one or more modular housings which in turn are mounted within said library computer system such that the media-inserting faces of said SCSI storage drives abut a robotic media inserter, and the electrical connection faces of said SCSI storage drives are connected to pigtails of an I/O bus which are greater than 100 mm in length, wherein said I/O bus, but for its length, is configured as a SCSI single-ended cable, wherein said I/O bus is accessible to a service technician while the library computer system is online, and wherein said replacement is accomplished while one or more good SCSI storage drives in said library computer system remain in an Online_-- good state so that a host, with the aid of a library controller, has continual access to information stored within said library computer system, said method comprising the steps of:

a) detecting that a SCSI storage drive has failed;

b) placing said failed SCSI storage drive in an Offline_-- failed state, wherein further access to said failed SCSI storage drive is not allowed;

c) placing any good SCSI storage drives mounted within the same modular housing as said failed SCSI storage drive in an Offline_-- good_-- pending state;

d) removing any media remaining in said SCSI storage drives in an Offline_-- good_-- pending state;

e) placing said Offline_-- good_-- pending SCSI storage drives in an Offline_-- good state;

f) removing power to any SCSI storage drive in an Offline_-- failed or Offline_-- good state;

g) disconnecting from the library computer system any cabling which is attached to said electrical connection faces of said Offline_-- failed and Offline_-- good SCSI storage drives, including the pigtails of the I/O bus;

h) replacing said failed SCSI storage drive;

i) reconnecting to the electrical connection faces of said replaced SCSI storage drive, and any good SCSI storage drives mounted within the same modular housing as said replaced SCSI storage drive, any cabling disconnected from said SCSI storage drives, including the pigtails of the I/O bus;

j) placing said Offline_-- failed and Offline_-- good SCSI storage drives in an Online_-- pending state;

k) restoring power to said Online_-- pending SCSI storage drives;

l) placing said Online_-- pending SCSI storage drives in an Online_-- good state so that these SCSI storage drives may again be accessed by said host and library controller.

16. A method as in claim 15, wherein step h) further comprises the step of pulling said modular housing in which said failed SCSI storage drive is mounted away from said robotic media inserter.

17. A method as in claim 15, wherein before step d), the method further comprises the step of:

d'"'"') issuing one or more commands to remove any media remaining in said SCSI storage drives in an Offline_-- good_-- pending state.

18. A method as in claim 17, wherein steps a), b), c), d'"'"'), e) and l) are performed by said host.

19. A method as in claim 18, wherein each of steps b), c), e) and l) further comprises the step of said host issuing a SCSI Write Buffer command.

20. A method as in claim 15, wherein said library controller performs steps f) and k), and step a) comprises the step of:

a'"'"') said library controller informing said host that execution of a library command was unsuccessful; and

before step l), the method further comprises the step of;

l'"'"') said library controller informing said host of said SCSI storage drive state changes.

21. A method as in claim 15, wherein step a) comprises the further step of said host receiving a status code from said library controller that execution of a SCSI storage drive command was unsuccessful.

22. A method as in claim 15, wherein step j) comprises the further step of resetting one or more switches associated with said Online_-- pending SCSI storage drives.