Failure recovery for process relationships in a single system image environment

• US 6,115,830 A
• Filed: 03/28/1998
• Issued: 09/05/2000
• Est. Priority Date: 11/11/1997
• Status: Expired due to Term

• Current Assignee: Hewlett Packard Enterprise Development LP (Hewlett-Packard Enterprise Company)
• Original Assignee: Compaq Computer Corporation (HP Inc.)
• Inventors: Zabarsky, Jeffrey A., Walker, Bruce J.
• Primary Examiner(s): Wright, Norman M.
• Application Number:
  US09/050,226
• Time in Patent Office: 892 Days
• Field of Search: 714/15, 714/18, 714/20, 714/4, 714/23, 707/2, 707/102, 705/54
• US Class Current: 714/15
• CPC Class Codes: G06F 11/1482 by means of middleware or OS functionality
• CPC Class Codes: G06F 11/203 using migration
• CPC Class Codes: G06F 11/2035 without idle spare hardware
• CPC Class Codes: Y10S 707/99932 Access augmentation or optimizing

Abstract

A system for recovery of process relationships following node failure within a computer cluster is provided. For relationship recovery, each node maintains set of care relationships. Each relationship is of the form carer cares about care target. Care relationships describe process relations such as parent-child or group leader-group member. Care relationships are stored at the origin node of their care targets. Following node failure, a surrogate origin node is selected. The surviving nodes then cooperate to rebuild vproc structures and care relationships for the processes that originated at the failed node at the surrogate origin node. The surviving nodes then determine which of their own care targets were terminated by the node failure. For each terminated care targets, notifications are sent to the appropriate carers. This allows surviving processes to correctly recover from severed process relationships.

Claims

1. A method for recovery of process relationships after failure of a node within a computer cluster, the method comprising the steps of:

selecting, by a cluster management service process, a surrogate origin node;

generating, by a slave daemon, a list of care relationships, each care relationship involving a process that originated at the failed node;

sending by the slave daemon, the list of care relationships to the surrogate origin node;

receiving by the surrogate origin node, the list of care relationships; and

reconstructing, by the surrogate origin node, a complete set of care relationships for processes that originated at the failed node.

2. A method as recited in claim 1 wherein the step of generating, by the slave daemon, further comprises the steps, performed by the slave daemon, of:

detecting processes that originated at the failed node; and

detecting processes having care relationships to processes that originated at the failed node.

3. A method as recited in claim 1 further comprising the steps of:

detecting, by the slave daemon, processes that were terminated by the failure of the failed node; and

sending, by the slave daemon, messages to the processes and nodes having care relationships to the processes that were terminated by the failure of the failed node.

4. A method as recited in claim 1 further comprising the step, performed by the surrogate origin node, of creating a vproc structure for each process that originated at the failed node.

5. A method as recited in claim 4 further comprising the step, performed by the surrogate origin node, of marking vproc structures created for processes terminated by the failure of the failed node as ghosts.

6. A system for recovery of process relationships after failure of a node within a computer cluster, the system comprising the steps of:

a cluster management service process configured to select a surrogate origin node to substitute for the failed node;

a slave daemon configured to generate a list of care relationships, each care relationship involving a process that originated at the failed node, the slave daemon also configured to send the list of care relationships to the surrogate origin node; and

a rebuilding process configure to receive the list of care relationships and to reconstruct a complete set of care relationships for processes that originated at the failed node.

7. A system as recited in claim 6 wherein the slave daemon is further configured to detect processes that originated at the failed node and detect processes having care relationships to processes that originated at the failed node.

8. A system as recited in claim 6 wherein the slave daemon is further configured to detect processes that were terminated by the failure of the failed node, and send messages to the processes and nodes having care relationships to the processes that were terminated by the failure of the failed node.

9. A system as recited in claim 6 wherein the rebuilding process is further configured to create a vproc structure for each process that originated at the failed node.

10. A system as recited in claim 6 wherein the rebuilding process is further configured to mark vproc structures created for processes terminated by the failure of the failed node as ghosts.

11. A computer program product comprising:

a computer usable medium having computer readable code embodied therein for recovery of process relationships after failure of a node within a computer cluster, the computer program product comprising;

first computer readable program code devices configured to cause a computer system to select a surrogate origin node to substitute for the failed node;

second computer readable program code devices configured to cause a computer system to generate a list of care relationships, each care relationship involving a process that originated at the failed node;

third computer readable program code devices configured to cause a computer system to send the list of care relationships to the surrogate origin node;

fourth computer readable program code devices configured to cause a computer system to receive the list of care relationships; and

fifth computer readable program code devices configured to cause a computer system to reconstruct a complete set of care relationships for processes that originated at the failed node.

12. A computer program product as recited in claim 11 wherein the second computer readable program code devices are configured to cause a computer system to detect processes that originated at the failed node and detect processes having care relationships to processes that originated at the failed node.

13. A computer program product as recited in claim 11 wherein the second computer readable program code devices are configured to cause a computer system to detect processes that were terminated by the failure of the failed node, and send messages to the processes and nodes having care relationships to the processes that were terminated by the failure of the failed node.

14. A computer program product as recited in claim 11 wherein the fifth computer readable program code devices are configured to cause a computer system to create a vproc structure for each process that originated at the failed node.

15. A computer program product as recited in claim 14 wherein the fifth computer readable program code devices are configured to cause a computer system to mark vproc structures created for processes terminated by the failure of the failed node as ghosts.