Method and apparatus for using a serial cable as a cluster quorum device

• US 7,412,500 B2
• Filed: 01/13/2005
• Issued: 08/12/2008
• Est. Priority Date: 08/23/2004
• Status: Active Grant

• Current Assignee: Oracle America, Inc. (Oracle Corporation)
• Original Assignee: Sun Microsystems Incorporated (Oracle Corporation)
• Inventors: Mazzilli, Jean-Pascal, Mckinty, Stephen J.
• Primary Examiner(s): Myers; Paul R.
• Assistant Examiner(s): Spittle; Matthew D
• Application Number:
  US11/034,694
• Publication Number:
  US 20060041779A1
• Time in Patent Office: 1,307 Days
• Field of Search: 710/105, 710/107, 710/108, 710/240, 710/110, 709/221
• US Class Current: 709/221
• CPC Class Codes: G06F 11/18 using passive fault-masking of the redundant circuits error detection by comparing the output of redundant processing systems with continued operation after detection of the error G06F11/165
• CPC Class Codes: G06F 9/5072 Grid computing
• CPC Class Codes: H04L 69/40 for recovering from a failure of a protocol instance or entity, e.g. service redundancy protocols, protocol state redundancy or protocol service redirection management of faults, events, alarms or notifications in data switching networks H04L41/06

Abstract

A method for obtaining a quorum vote by a first node using a quorum cable, wherein the quorum cable comprises a first end connected to the first node and a second end connected to a second node, including determining whether the quorum cable is reserved by the second node using a remote reservation input (RRI) on the first end, if the quorum cable is not reserved by the second node asserting a local reservation input (LRI) on the first end by the first node, monitoring a local reservation output (LRO) on the first end by the first node to determine whether the LRO on the first end is asserted in response to asserting the LRI on the first end, and obtaining the quorum vote by the first node, if the LRO on the first end is asserted.

Claims

1. A method for obtaining a quorum vote by a first node using a quorum cable, comprising:

determining that a network connection between the first node and a second node is inactive, wherein the network connection is associated with a local area network;

determining whether the quorum cable is reserved by the second node using a remote reservation input (RRI) on a first end;

if the quorum cable is not reserved by the second node;

asserting a local reservation input (LRI) on the first end by the first node;

monitoring a local reservation output (LRO) on the first end by the first node to determine whether the LRO on the first end is asserted in response to asserting the LRI on the first end; and

obtaining the quorum vote by the first node, if the LRO on the first end is asserted,

wherein the quorum vote is obtained without direct communication between the first node and the second node,

wherein the quorum cable comprises the first end directly connected to the first node and the second end directly connected to a second node.

2. The method of claim 1, further comprising, prior to obtaining the quorum vote:

waiting for a collision detection timeframe to elapse, if the first LRO was asserted;

monitoring the RRI on the first end by the first node to determine whether the RRI on the first end is asserted after the collision detection timeframe has elapsed, if the LRO on the first end was asserted; and

executing a collision detection algorithm if the RRI on the first end is asserted after the collision detection timeframe has elapsed.

3. The method of claim 2, wherein executing the collision detection algorithm, comprises:

stopping assertion of the LRI on the first end by the first node;

stopping assertion of a LRI on the second end by the second node;

reasserting the LRI on the first end by the first node after a first pre-determined amount of time; and

reasserting the LRI located on the second end by the second node after a second pre-determined amount of time.

4. The method of claim 3, wherein the first pre-determined amount of time and the second pre-determined amount of time are randomly determined.

5. The method of claim 3, wherein the first pre-determined amount of time is determined using information associated with the first node and the second pre-determined amount of time is determined using information associated with the second node.

6. The method of claim 2, wherein executing the collision detection algorithm, comprises:

obtaining a first priority information associated with the first node;

obtaining a second priority information associated with the second node;

determining whether the first node or the second node has priority to reserve the quorum cable using the first priority information and the second priority information;

stopping assertion of a LRI on the second end by the second node, if the first node has priority; and

stopping assertion of the LRI on the first end by the first node, if the second node has priority.

7. The method of claim 1, using the RRI on the first end to determine whether the quorum cable is reserved, wherein determining whether the quorum cable is reserved comprises determining whether the RRI on the first end is asserted.

8. A cluster comprising:

a quorum cable having a first end and a second end;

a first node directly connected to the first end; and

a second node directly connected to the second end,

wherein the first node and the second node comprise functionality to reserve the quorum cable, thereby obtaining a quorum vote, wherein the quorum vote is obtained without direct communication between the first node and the second node,

wherein the first end comprises a first local reservation input (LRI), a first local reservation output (LRO), and a first remote reservation input (RRI), and the second end comprises a second LRI, a second LRO, and a second RRI,

wherein the first node and the second node are connected via a network connection associated with a local area network.

9. The cluster of claim 8, wherein

the first LRO is connected to the second RRI,

the first RRI is connected to the second LRI,

the second LRO is connected to the first RRI, and

the second RRI is connected to the first LRI.

10. The cluster of claim 8, wherein functionality to reserve the quorum cable by the first node comprises functionality to:

determine whether the quorum cable is reserved using the first RRI;

if the quorum cable is not reserved by the second node;

assert the first LRI by the first node;

monitor the first LRO by the first node to determine whether the first LRO is asserted in response to asserting the first LRI; and

obtain the quorum vote by the first node, if the first LRO is asserted.

11. The cluster of claim 10, further comprising functionality to:

prior to obtaining the quorum vote;

wait for a collision detection timeframe to elapse, if the first LRO was asserted;

monitor the first RRI by the first node to determine whether the first RRI is asserted after the collision detection timeframe has elapsed, if the first LRO was asserted; and

execute a collision detection algorithm if the first RRI is asserted after the collision detection timeframe has elapsed.

12. The cluster of claim 11, wherein executing the collision detection algorithm, comprises functionality to:

stop assertion of the first LRI by the first node;

stop assertion of the second LRI byte second node;

reassert the first LRI by the first node after a first pre-determined amount of time; and

reassert the second LRI by the second node after a second pre-pre-determined amount of time.