Recovery procedure for a dynamically reconfigured quorum group of processors in a distributed computing system

• US 6,487,678 B1
• Filed: 08/31/1999
• Issued: 11/26/2002
• Est. Priority Date: 08/31/1999
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Novaes, Marcos N., Briskey, Kenneth C.
• Primary Examiner(s): Beausoleil, Robert
• Assistant Examiner(s): DUNCAN, MARC M
• Application Number:
  US09/387,185
• Time in Patent Office: 1,183 Days
• Field of Search: 714/6, 714/12, 714/23, 714/10, 714/820, 714/4, 710/39, 710/200, 707/10, 707/203, 707/28, 712/30, 712/15, 712/10
• US Class Current: 714/4.1
• CPC Class Codes: G06F 11/142 Reconfiguring to eliminate the error group management mechanisms in a peer-to-peer network H04L67/1044

Abstract

Dynamic reconfiguration of a quorum group of processors and recovery procedure therefore are provided for a shared nothing distributed computing system. Dynamic reconfiguration proceeds notwithstanding unavailability of at least one processor of the quorum group of processors assuming that a quorum of the remaining processors exists. Recovery processing is implementing by the group of processors so that the at least one processor which was unavailable during the dynamic reconfiguration of the group is able to obtain current state information once becoming active. Each processor of the group of processors includes an incarnation number and a list of member processors which resulted from a commit process resulting in its incarnation number. The recovery processing includes exchanging the processors'"'"' incarnation numbers and lists of processors for propagation of the current state of the quorum group of processors to the at least one processor now becoming available.

Claims

1. A method for recovering a current state of a quorum group of processors dynamically reconfigured while at least one processor of said quorum group of processors was unavailable, said method comprising:

obtaining, for said at least one processor, said current state of said dynamically reconfigured quorum group of processors from one or more processors of said quorum group of processors; and

wherein each processor of said quorum group of processors includes an incarnation number and a list of member processors which resulted from a commit process;

wherein said commit process resulted in said incarnation number and required participation by a majority of processors of said quorum group of processors; and

wherein said obtaining comprises checking said one or more processors of said quorum group of processors for incarnation numbers and lists of member processors which resulted from the commit processes.

2. The method of claim 1, wherein said checking comprises updating said at least one processor'"'"'s incarnation number and list of member processors with a current incarnation number and a current list of member processors received from said one or more processors of said quorum group of processors.

3. The method of claim 1, wherein said obtaining comprises exchanging states of processors in said quorum group of processors in order that a current incarnation number and a current list of member processors of said quorum group of processors is propagated to said at least one processor.

4. The method of claim 1, wherein said checking comprises sending from said at least one processor a message to said one or more processors of said quorum group of processors requesting their incarnation numbers and lists of member processors.

5. The method of claim 1, wherein said obtaining comprises determining said current state using a current quorum number of processors obtained using a current list of member processors received from said one or more processors of said quorum group of processors.

6. The method of claims 5, wherein said lists of member processors maintained by said one or more processors comprises for each processor of said one or more processors a list of inactive processors of the group and a list of active processors of the group, and wherein said quorum comprises for a particular processor, less than a majority of defined processors in said quorum group of processors if the number of defined processors less the number of processors in its inactive list minus one is less than a majority of the number of processors in its active list.

15. The method of claim 2, wherein said current incarnation number and said current list of member processors are received from a majority of processors of said quorum group of processors.

7. A system for recovering a current state of a quorum group of processors dynamically reconfigured while at least one processor of said quorum group of processors was unavailable, said system comprising:

means for obtaining, for said at least one processor, said current state of said dynamically reconfigured quorum group of processors from one or more processors of said quorum group of processors; and

wherein each processor of said quorum group of processors includes an incarnation number and a list of member processors which resulted from a commit process;

wherein said commit process resulted in said incarnation number and required participation by a majority of processors of said quorum group of processors; and

wherein said means for obtaining comprises means for checking said one or more processors of said quorum group of processors for incarnation numbers and lists of member processors which resulted from the commit processes.

8. The system of claim 7, wherein said means for checking comprises means for updating said at least one processor'"'"'s incarnation number and list of member processors with a current incarnation number and a current list of member processors received from said one or more processors of said quorum group of processors.

9. The system of claim 7, wherein said means for obtaining comprises means for exchanging states of processors in said quorum group of processors in order that a current incarnation number and a current list of member processors of said quorum group of processors is propagated to said at least one processor.

10. The system of claim 7, wherein said means for checking comprises means for sending from said at least one processor a message to said one or more processors of said quorum group of processors requesting their incarnation numbers and lists of member processors.

11. The system of claim 7, wherein said means for obtaining comprises means for determining said current state using a current quorum number of processors obtained employing a current list of member processors received from said one or more processors of said quorum group of processors.

12. The system of claim 11, wherein said lists of member processors maintained by said one or more processors comprises for each processor of said one or more processors a list of inactive processors of the group and a list of active processors of the group, and wherein said quorum comprises for a particular processor, less than a majority of defined processors in said quorum group of processors if the number of defined processors less the number of processors in its inactive list minus one is less than a majority of the number of processors in its active list.

16. The system of claim 8, wherein said current incarnation number and said current list of member processors are received from a majority of processors of said quorum group of processors.

13. A distributed processing system comprising:

a group of processors, said group of processors being adapted to allow recovering of a current state for at least one processor thereof which was unavailable during a dynamic reconfiguration of said group of processors, said group of processors being adapted to;

obtain, for said at least one processor, said current state of said dynamically reconfigured quorum group of processors from one or more processors of said quorum group of processors; and

wherein each processor of the quorum group of processors includes an incarnation number and a list of member processors which resulted from a commit process;

wherein said commit process resulted in said incarnation number and required participation by a majority of processors of said quorum group of processors; and

wherein the group of processors is adapted to allow the at least one processor to check the one or more processors for incarnation numbers and lists of member processors which resulted from the commit processes.

14. At least one program storage device readable by a machine, tangibly embodying at least one program of instructions executable by the machine to perform a method for recovering a current state of a quorum group of processors dynamically reconfigured while at least one processor of said quorum group of processors was unavailable, comprising:

obtaining, for said at least one processor, said current state of said dynamically reconfigured quorum group of processors from one or more processors of said quorum group of processors; and

wherein each processor of said quorum group of processors includes an incarnation number and a list of member processors which resulted from a commit process;

wherein said commit process resulted in said incarnation number and required participation by a majority of processors of said quorum group of processors; and

wherein said obtaining comprises checking said one or more processors of said quorum group of processors for incarnation numbers and lists of member processors which resulted from the commit processes.