Alan P Brown

US Patent:

20070245313, Oct 18, 2007

Filed:

Apr 14, 2006

Appl. No.:

11/404679

Inventors:

Donald Hardy - Seattle WA, US
Alan Brown - Ellensburg WA, US
Curtis Anderson - Kent WA, US
Jared Wilson - Sammamish WA, US
Loren Merriman - Duvall WA, US
Richard Wright - Redmond WA, US
Xun Zhang - Sammamish WA, US

Assignee:

Microsoft Corporation - Redmond WA

International Classification:

G06F 9/44

US Classification:

717124000

Abstract:

Failure tagging may be provided. A plurality of tests may be initiated on a program module and output resulting from the plurality of tests may be received. The output resulting from the plurality of tests may be determined to comprise at least one failure. The at least one failure may comprise an unknown failure or a known failure. Results of the plurality of tests indicating the failure may be transmitted. In addition, a request to stop one or more of a plurality of computers on a failure may be received when one of the plurality of computers encounters the failure. In addition, the program module may be transmitted to the plurality of computers configured to run tests on the program module. The one of the plurality of computers may be stopped on the failure when the one of the plurality of computers encounters the failure identified by the request.

US Patent:

20080184079, Jul 31, 2008

Filed:

Jan 31, 2007

Appl. No.:

11/700992

Inventors:

Loren Merriman - Duvall WA, US
Gordon Hardy - Seattle WA, US
Curtis Anderson - Kent WA, US
Alan T.B. Brown - Ellensburg WA, US
Richard L. Wright - Redmond WA, US
Jared Wilson - Redmond WA, US
Xun Zhang - Sammamish WA, US

Assignee:

Microsoft Corporation - Redmond WA

International Classification:

G06F 11/00

US Classification:

714 48, 714E11001

Abstract:

Computing environments, each executing at least one software program, are monitored for failures occurring during execution of the software program. Information associated with the failure, such as an identification of the software program and a failure type describing the failure, is recorded. The failure information is quantified to report the number of times the software program has failed or the number of times a particular failure has occurred. The quantified data may provide help in prioritizing what program or what failures merit investigation and resolution. Reports may be received from failing computing systems stopped at a state following the occurrence of the failure. In response, hold information is checked to determine whether to instruct the failing computing system to hold a state existing upon the occurrence of the failure.