Two node virtual shared disk cluster recovery

• US 7,302,607 B2
• Filed: 08/29/2003
• Issued: 11/27/2007
• Est. Priority Date: 08/29/2003
• Status: Expired due to Fees

• Current Assignee: Maplebear Inc.
• Original Assignee: International Business Machines Corporation
• Inventors: Gunda, Kalyan C., Herr, Brian D.
• Primary Examiner(s): Maskulinski; Michael
• Application Number:
  US10/652,509
• Publication Number:
  US 20050050376A1
• Time in Patent Office: 1,551 Days
• Field of Search: 714/4, 714/9, 714/10
• US Class Current: 714/10
• CPC Class Codes: G06F 11/2033 switching over of hardware resources
• CPC Class Codes: G06F 11/2092 Techniques of failing over between control units

Abstract

A method for recovery in a two-node data processing system is provided wherein each node is a primary server for a first nonvolatile storage device and for which there is provided shared access to a second nonvolatile storage device for which the other node is a primary server and wherein each node also includes a direct connection to the shared nonvolatile storage device for which the other node is the primary server. Upon notification of failure, the method operates by first confirming continued access by each node to the nonvolatile storage device for which it is the primary server and then by attempting to access the shared nonvolatile storage device via the direct connection and by waiting for a time sufficient for the same process to be carried out by the other node. If access to the shared nonvolatile storage device is successful, the node takes control of both nonvolatile storage devices. If the access is not successful a comparison of node numbers is carried out to decide the issue of control. Whenever a node determines that it does not have access to the storage device for which it is the primary server, it shuts down recovery at the node.

Claims

1. A method for recovery in a two-node data processing system wherein each node is a primary server for a first disk drive and for which there is provided shared access to a second disk for which the other node is a primary server and wherein each node also includes a direct connection to the shared disk for which the other node is the primary server, said method comprising the steps of:

receiving, at a first node, notification of communication failure with said second node;

determining if said first node has access to the disk for which said first node is the primary server;

shutting down recovery at said first node if said access is not present, but if it is present, accessing the disk for which said second node is the primary server via said hardware connection and waiting for a period of time sufficient to assure that recovery processes at said other node have completed past the same point as said first node;

determining if said first node still has access to the disk for which it is the primary server and if said first node still has access, taking over control of the second node'"'"'s disk; and

if said first node doesn'"'"'t have said access per said immediately preceding determining step, comparing node numbers to decide which node controls the other node'"'"'s disks.

2. The method of claim 1 in which said comparing also determines that the node that does not control the other node'"'"'s disk shuts down its recovery process.

3. The method of claim 1 in which said nodes communicate via communication adapters at each node wherein said adapters connect the nodes through a switch.

4. The method of claim 3 in which said notification is transmitted through an adapter.

5. The method of claim 1 in which said period of time is greater than about:

(process swap time+CPU time slice+time taken to break reservation) *2.