Troubleshooting Jobs and Tasks

Updated: May 2011

Applies To: Windows HPC Server 2008, Windows HPC Server 2008 R2

This section contains information to help you troubleshoot and resolve common issues with compute jobs and tasks in your Windows HPC Server 2008 R2 or Windows HPC Server 2008 cluster.

In this section

This section includes the following topics:

Title Description

Investigating Job Failures

This topic describes how to view job and task error messages and provides a starting point for investigating job and task failures.

Job Failed to Start because of Logon Failure

This topic provides steps to diagnose and resolve logon failure errors that result from a user not having local logon permission on compute nodes.

Tasks That Complete Successfully Are Marked As Failed

This topic describes how to modify the command line of a task that returns a non-zero exit code for success so that upon completion, the task is marked as Finished.

Blue Screen Appears on a Node Running a GPGPU Job

This topic describes a registry setting that can be modified or disabled to prevent the interruption of a long-running GPGPU job on a compute node.

Job Is Scheduled Using Stale Node Hardware Configuration Information

This topic provides steps to ensure that a change to the hardware configuration of a compute node is discovered and is used by compute jobs in the cluster.

See Also

Windows HPC Server 2008 R2: Troubleshooting