Method and apparatus for reliable failover involving incomplete raid disk writes in a clustering system

• US 20040019821A1
• Filed: 07/26/2002
• Published: 01/29/2004
• Est. Priority Date: 07/26/2002
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: King, Allen, Chu, Davis Qi-Yu
• Granted Patent: US 7,069,465 B2
• Time in Patent Office: Days
• US Class Current: 714/6
• CPC Class Codes: G06F 11/2092 Techniques of failing over between control units
• CPC Class Codes: G06F 2003/0697 device management, e.g. handlers, drivers, I/O schedulers
• CPC Class Codes: G06F 2201/82 Solving problems relating to consistency ensuring consistency in mirrored systems G06F11/2064
• CPC Class Codes: G06F 3/0601 Interfaces specially adapted for storage systems

Abstract

Disclosed is a system and a method for reliable failover involving incomplete redundant arrays of inexpensive disks (RAID) writes in clustering systems. The system comprises the restructuring of a mirror race table (MRT) and a storage location that may be a non-volatile random access memory (NVRAM) module in a shared disk enclosure to make the MRT accessible to all nodes in the system. In one embodiment, the NVRAM module comprises copies of MRTs from each node operating within the cluster system.

Claims

1. An apparatus for reliable failover capabilities, the apparatus comprising:

a plurality of RAID controllers configured to access a plurality of shared disks;

a mirror race table (MRT); and

a common MRT storage location accessible by each of the RAID controllers.

2. The apparatus of claim 1, wherein the common MRT storage location comprises a non-volatile random access memory (NVRAM) module.

3. The apparatus of claim 1, further comprising a shared disk enclosure comprising a SCSI accessed fault-tolerant enclosure (SAF-TE).

4. The apparatus of claim 1, further comprising an MRT search module configured to search the MRT for the first free entry.

5. The apparatus of claim 4, wherein the MRT search module is configured to search from the top to the bottom of the MRT.

6. The apparatus of claim 1, further comprising an MRT entry module configured to create an entry in the MRT by entering the logical block address (LBA) of the first cache line group to be written to the plurality of shared disks.

7. The apparatus of claim 1, further comprising an MRT pointer module configured to save an MRT pointer for the entry.

8. The apparatus of claim 1, further comprising an MRT retrieve module configured to find and retrieve the MRT pointer for the entry

9. The apparatus of claim 1, further comprising an MRT read module configured to locate the cache line group of data on disks for consistency restoration after a node bootup or cluster failover.

10. The apparatus of claim 1, further comprising an MRT clear module configured to find and clear the entry from the MRT.

11. The apparatus of claim 1, further comprising an MRT transfer module configured to transfer the MRT from the shared disk enclosure to at least one RAID controller.

12. A method for reliable failover capabilities, the method comprising:

accessing a plurality of shared disks with a plurality of RAID controllers;

updating a mirror race table (MRT); and

providing a common MRT storage location accessible by each of the RAID controllers.

13. The method of claim 12, wherein establishing a common MRT storage location further comprises establishing a common MRT storage location within a shared disk enclosure.

14. The method of claim 12, further comprising:

accessing the MRT;

updating the MRT from the common MRT storage location;

detecting a failure of at least one RAID controller; and

reliably distributing the work load of the failed RAID controller.

15. The method of claim 12, further comprising providing a shared disk enclosure comprising a SCSI accessed fault-tolerant enclosure (SAF-TE).

16. The method of claim 12, further comprising searching the MRT for the first free entry.

17. The method of claim 14, wherein searching the MRT further comprises searching from the top to the bottom of the MRT.

18. The method of claim 12, further comprising creating an entry in the MRT by entering the logical block address (LBA) of the first cache line group to be written to the plurality of shared disks.

19. The method of claim 12, further comprising saving an MRT pointer for the entry.

20. The method of claim 12, further comprising finding and retrieving the MRT pointer for the entry

21. The method of claim 12, further comprising reading the MRT entry to locate data for consistency restoration when necessary.

22. The method of claim 12, further comprising finding and clearing the entry from the MRT.

23. The method of claim 12, further comprising transferring the MRT from the shared disk enclosure to at least one RAID controller.

24. A method for reliable failover capabilities, the method comprising:

providing a plurality of shared disks;

providing a plurality of RAID controllers;

providing a mirror race table (MRT);

establishing a common MRT storage location accessible by each of the RAID controllers;

accessing the MRT;

updating the MRT;

detecting a failure of at least one RAID controller;

distributing the work load of the failed RAID controller;

searching the MRT for the first free entry;

creating an entry in the MRT by entering the logical block address (LBA) of the first cache line group to be written to the plurality of shared disks;

saving an MRT pointer for the entry;

finding and retrieving the MRT pointer for the entry;

reading the MRT entry to locate data to be made consistent;

finding and clearing the entry from the MRT;

transferring the MRT from the shared disk enclosure to at least one RAID controller; and

performing uncompleted write operations.

25. A system for reliable failover capabilities, the system comprising:

a plurality of shared disks,

a plurality of RAID controllers;

a mirror race table (MRT); and

a common MRT storage location accessible by each of the RAID controllers, the common MRT storage location comprising a non-volatile random access memory (NVRAM) module.

26. An apparatus for reliable failover capabilities, the apparatus comprising:

means for accessing a plurality of shared disks with a plurality of RAID controllers;

means for updating a mirror race table (MRT);

means for accessing the MRT; and

means for storing a common MRT such that the common MRT is accessible by each of the RAID controllers.