Using Windows HPC Server Log Files


Applies To: Windows HPC Server 2008

This topic contains information about the location and contents of log files in Windows® HPC Server 2008. You can use the information in the log files to help troubleshoot the services and operations in Windows HPC Server 2008.

In this topic:


To verify that the necessary Windows HPC Server 2008 and other services are running and configured properly on specific nodes in the cluster, you can run the All Services Running and the Service Configurations Report diagnostics on the nodes. For more information, see Run Diagnostic Tests [Help link].

Deployment log files for Microsoft HPC Pack 2008

During deployment of Microsoft HPC Pack 2008 on a node, the following log files are generated in the HPCSetupLogs folder on each node:




Generated by Windows Installer for the HPC Pack package


Generated by HPC Pack Setup

The HPCSetupLogs folder is in the following folder:

  • On the head node: %Temp%\

  • On each compute node: %SystemRoot%\temp\

You can open these files to help troubleshoot issues with the installation of HPC Pack on a node.


If you install a service pack for Microsoft HPC Pack 2008 on the head node, the log file %Temp%\HPC_Patch.log is generated during the installation of the service pack.

Logs in HPC Cluster Manager

You can view the following logs in HPC Cluster Manager to track recent changes to your cluster and to monitor the progress of ongoing operations.



Operations log

Generated to record configuration, node management, job management, and diagnostics operations in the cluster.

You can view the operations log for the entire cluster, filter it to view only specific nodes, or pivot to operations to correlate log entries with specific nodes or diagnostic tests.

For information about viewing the operations log, see Read the Operations Log.

Provisioning log

Generated to track the progress of provisioning a compute node.

To view the provisioning log for a compute node, in Node Management, in the views pane, double-click the node. Then, click the Provisioning Log tab.


The provisioning log restarts each time a node is provisioned. Entries are also recorded to the operations log.

Activity log

Generated to track the progress of a cluster job.

To view the activity log for a job, in Job Management, in the views pane, double-click the job. Then, click the Activity Log tab.


You can also use the Get-HpcOperationLog PowerShell cmdlet to get the log for a specific operation in HPC Cluster Manager. For more information, see Get-HpcOperationLog.

Service log files

The Windows HPC Server 2008 management services that are running on the nodes generate the following log files:

Log file




Generated by the HPC Management Service, which manages the configuration information for the cluster.

Head node, compute nodes


Generated by the HPC SDM Store Service, which stores configuration information for the cluster.

Head node


Generated by the HPC Node Manager Service, which manages processes for applications that run on the cluster.

Head node, compute nodes

On each node, the location of the log files and the current logging level are determined by the following registry key settings in HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\HPC:

Registry key

Default value






The TraceLevel registry key setting on the head node also determines the logging level of the HPC Basic Profile Web Service log files.

How to change the logging level in the service log files

By default, the Windows HPC Server 2008 service log files show error messages and warning messages only. You can configure the logging level in the Windows HPC Server 2008 service log files to display more or fewer messages. For example, you might want to enable the logging of trace messages to troubleshoot the operations on a cluster node.

To change the logging level, you configure the value of the TraceLevel registry key. The valid values of TraceLevel are in the following table:

TraceLevel value



No logging


Log error messages only


Log error and warning messages only


Log error, warning, and information messages only


Log error, warning, information, and trace messages


Setting TraceLevel to 4 can greatly increase the size of the service log files and can affect the performance of the head node. It is recommended that you use this setting only temporarily to help troubleshoot an issue.


The following procedure includes steps to modify the registry. Incorrectly editing the registry may severely damage your system. Before making changes to the registry, you should back up any valued data on the computer.

To configure the TraceLevel registry key

  1. On the head node, click Start, and then click Run.

  2. In the Run dialog box, in the Open box, type regedit, and then click OK to open Registry Editor.

  3. In Registry Editor, navigate to HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\HPC.

  4. Set the TraceLevel key to an integer between 0 and 4, and then click OK.

HPC Basic Profile Web Service log files

The HPC Basic Profile Web Service on the head node provides access to the HPC Job Scheduler service using a web service protocol. If the HPC Basic Profile Web Service is activated, the following log files for the service are generated by default in the %CCP_HOME%\Bin\ folder:

Log file



Generated by the HPC Basic Profile Web Service.

A log of errors and information about how the HPC Basic Profile Web Service is working.


Generated by the WCF hosting environment.

A log of message traffic going in and out of the HPC Basic Profile Web Service.


The log files for the HPC Basic Profile Web Service are recorded in a text-based XML format that can be browsed using the Service Trace Viewer Tool (SvcTraceViewer.exe) that is part of the .NET 3.0 (and later) redistributable package. Alternatively, you can use any text file editor to view the messages. For information about the Service Trace Viewer Tool, see Service Trace Viewer Tool (SvcTraceViewer.exe) ( To download the Microsoft .NET Framework, see Microsoft .NET Framework (

For more information about log files for the HPC Basic Profile Web Service, see HPC Server Basic Profile Web Service Operations Guide.

Event logs

The following event logs for Windows HPC Server 2008 can be viewed in Event Viewer on the head node:



Applications and Services Logs\Windows HPC Server

Generated by multiple sources, including: HPC Management Service, HPC SDM Store Service, HPC Node Manager Service, HPC Job Scheduler Service

Applications and Services Logs\Microsoft\Windows\HPCServer\Scheduler/Operational

Generated by the HPC Job Scheduler Service.

A log of state transition information and internal scheduler data.

Applications and Services Logs\Microsoft\Windows\HPCServer\Scheduler/Debug

Generated by the HPC Job Scheduler Service.

A debug-level log of state transition information and internal scheduler data.


Debug-level logging is not enabled by default. See the procedure below to enable a debug log for the HPC Job Scheduler Service.


You can also open Event Viewer from HPC Cluster Manager to review events specifically for a node in your cluster. For more information, see View Node Events.

To enable debug-level logging for the HPC Job Scheduler Service

  1. Log on to your head node as a user with Administrator privileges.

  2. Open Event Viewer:

    Click Start, point to Administrative Tools, and then click Event Viewer.

  3. In the navigation pane, expand the Applications and Services Logs node, then expand the Microsoft node, then expand the Windows node, and then expand the HPCServer node.

  4. In the View menu, click Show Analytic and Debug Logs. The Scheduler/Debug log should now appear under the HPCServer node.

  5. Right-click Scheduler/Debug, and then click Enable Log.

MS-MPI trace logs

You can enable the creation of event trace logs for your Microsoft-Message Passing Interface (MS-MPI) applications.

For more information, see Tracing the Execution of MPI Applications with Windows HPC Server 2008 (

Additional references