Computer system failure management with topology-based failure impact determinations

• US 7,711,980 B1
• Filed: 05/22/2007
• Issued: 05/04/2010
• Est. Priority Date: 05/22/2007
• Status: Active Grant

• Current Assignee: Hewlett Packard Enterprise Development LP (Hewlett-Packard Enterprise Company)
• Original Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Inventors: Colnick, Erik C., Scarpello, Samuel L. Jr.
• Primary Examiner(s): Maskulinski; Michael C
• Application Number:
  US11/751,789
• Time in Patent Office: 1,078 Days
• Field of Search: None
• US Class Current: 714/4.1
• CPC Class Codes: G06F 11/0709 in a distributed system consisting of a plurality of standalone computer nodes, e.g. clusters, client-server systems
• CPC Class Codes: G06F 11/0772 Means for error signaling, e.g. using interrupts, exception flags, dedicated error registers
• CPC Class Codes: G06F 11/2007 using redundant communication media
• CPC Class Codes: G06F 2209/506 Constraint
• CPC Class Codes: G06F 9/5027 the resource being a machine, e.g. CPUs, Servers, Terminals
• CPC Class Codes: H04L 41/069 using logs of notifications; Post-processing of notifications
• CPC Class Codes: H04L 41/12 Discovery or management of network topologies
• CPC Class Codes: H04L 67/1001 for accessing one among a plurality of replicated servers
• CPC Class Codes: H04L 67/125 involving control of end-device applications over a network

Abstract

The present invention provides for indicating devices that are impacted by the failures of another device. Then when allocating workloads to devices, non-impacted devices are given priority over impacted devices as allocation targets for workloads.

Claims

1. A method comprising:

detecting a failure of a first device in a computer system;

identifying impacted and non-impacted devices of said computer system, said impacted devices being impacted by said failure, said non-impacted devices being less impacted as a group by said failure than said impacted devices as a group; and

assigning workloads to devices giving priority to said non-impacted devices over said impacted devices.

2. A method as recited in claim 1 wherein said failure impairs the manageability of one of said impacted devices.

3. A method as recited in claim 2 wherein said first device is a terminal server for managing resource servers, said terminal server being coupled to each of said resource serves via a switch, said resource servers being identified as impacted as they cannot be managed by said terminal server given its failed status.

4. A method as recited in claim 1 further comprising generating a dependency topology from a change database describing said system, said identifying involving tracing dependency relationships to identify devices that are indirectly impacted by said failure.

5. A method as recited in claim 1 wherein said identifying involves a determination that an impacted device is connected to said first device or to another device that is impacted by the failure of said first device.

6. A method as recited in claim 1 wherein said identifying involves a determination that an impacted device includes said first device.

7. A method as recited in claim 1 wherein said identifying involves a determination that an impacted device is functionally dependent on said first device.

8. A computer system comprising:

resource servers for rum user workloads;

management servers for managing said resource servers;

network infrastructure devices for managing communications among said management and resource servers;

a change database in computer-readable storage media indicating direct dependency relationships of said resource servers on said network infrastructure devices and said management servers;

a computer executable failure monitor in computer-readable media for detecting a failure of one of said infrastructure devices or said management devices; and

a failure marker for updating said database to indicate which of said resource servers is impacted by said failure.

9. A computer system as recited in claim 8 wherein at least one of said resource servers is impacted in that it is no longer functional.

10. A computer system as recited in claim 8 wherein at least one of said resource servers is impacted in that its manageability is impaired or lost due to said failure.

11. A computer system as recited in claim 8 wherein a set of said management servers allocates workloads to said resource servers, said management servers giving priority to servers not impacted by said failure over servers that are impacted by said failure.

12. Computer-readable storage media comprising:

a computer-executable failure monitor for detecting a failure of a device in a computer system;

a computer-executable database for indicating direct dependency relationships between devices of said computer system so that devices directly impacted by said failure can be identified;

a computer-executable topology generator for generating a dependency topology of said system from said database to identify devices indirectly impacted by said failure; and

a computer executable failure marker for updating said database to indicate which of said devices are impacted by said failure.

13. Computer-readable storage media as recited in claim 12 wherein said direct dependency relationship includes inclusion relationships, connection relationships, and functional relationships.

14. Computer-readable storage media as recited in claim 12 further comprising an allocation planner for planning allocation of workloads to said devices, said allocation planner giving priority to devices not impacted by said failure over devices impacted by said failure.

15. Computer-readable storage media as recited in claim 12 wherein said topology indicates that a resource server is impacted by a failure of a management server.

16. Computer-readable storage media as recited in claim 12 wherein said database distinguishes between parent, child, and sibling dependency relationships.