Support for single-node quorum in a two-node nodeset for a shared disk parallel file system

• US 6,915,391 B2
• Filed: 12/15/2000
• Issued: 07/05/2005
• Est. Priority Date: 12/15/2000
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Curran, Robert J., Wang-Knop, Kuei-Yu, Kandadai, Radha R.
• Primary Examiner(s): MCLEAN MAYO, KIMBERLY N
• Application Number:
  US09/737,393
• Publication Number:
  US 20020078312A1
• Time in Patent Office: 1,663 Days
• Field of Search: 711/147, 711/148, 711/150, 711/151, 711/152, 711/153, 711/154, 711/161, 711/162, 713/200, 713/201, 713/100, 713/153, 713/202, 714/1, 714/2, 714/3, 714/7, 714/8, 714/5, 714/100, 710/240
• US Class Current: 711/150
• CPC Class Codes: G06F 11/1425 by reconfiguration of node membership
• CPC Class Codes: G06F 16/10 File systems; File servers

Abstract

A method is provided for continuing node operations in quorum-based protocols. This is particularly valuable in the situation in which a group of nodes includes only two nodes and in which communications to a node is lost. Provision is made for adding additional status indications to each node indicating whether or not it is enabled for single-node quorum operation. Provision is also made for identifying one node as a primary node and another node as a secondary node for time-out response purposes.

Claims

1. A method for controlling access to data in a shared disk parallel data file, said method comprising the steps of:

determining, in a two-node system, that communications to one node has failed;

determining whether or not the node that has not failed has been enabled for single-node quorum operation; and

upon detection of single node enablement status, fencing data access to and from said failed node.

2. The method of claim 1 in which, between said determination steps, a file status indicator for said non-affected node is switched so as to indicate single-node operation.

3. The method of claim 1 in which said two nodes are separately indicated as being the primary and secondary node, and wherein when said secondary node detects that it has been isolated, said secondary node waits for a predetermined period of time and then sets a data bit indicating that it has been fenced.

4. The method of claim 3 in which, upon setting of said data bit which indicates that said secondary node has been fenced, said secondary node then unmounts all file systems and waits for a signal indicating that two-node operation is possible.