Set Event Logging Level and Collect Trace Logs
In HPC Cluster Manager, the Services view lets you manage and edit the service registration files from a single location, and enable and configure service (session) tracing. When you enable tracing in the service configuration file, the trace information is logged to a file on the compute nodes. The log files trace steps from the service call and the intermediate results on cluster. You can collect and remove traces by using the Job Management view or the HPC PowerShell cmdlets. You can view the trace log files with the WCF Service Trace Viewer (SvcTraceViewer.exe).
Important
Starting with HPC Pack 2012 you can view detailed information about the progress of SOA jobs and sessions, and view message-level traces for SOA sessions that are running either on-premises or on Windows Azure nodes. You can also export the SOA traces and share them offline. For more information, see View and Export Message-Level Traces. HPC Pack 2008 R2 only support the collection of session-level trace logs on on-premises nodes.
Important
Event logging is not generally recommended for production environments. After collecting the trace logs, ensure that you delete them from the compute nodes to avoid consuming disk space and to prevent unauthorized access to sensitive information that might be stored in the logs. Trace logs are created in the %CCP_DATA%SoaTrace folder.
Note
Service tracing is not automatically enabled on workstation nodes or on unmanaged server nodes. (Unmanaged server nodes are supported in HPC Pack starting with HPC Pack 2008 R2 with Service Pack 3.) To enable tracing, you must create the %CCP_DATA%SoaTrace folder on the workstation nodes or unmanaged server nodes. You must grant cluster users write permissions to the SoaTrace folder. After collecting the trace logs, ensure that you delete them from the compute nodes to avoid consuming disk space and to prevent unauthorized access to sensitive information that might be stored in the logs. To disable tracing, delete the SoaTrace folder.
To configure event logging level for a service
In HPC Cluster Manager, click Configuration, and then click Services.
Right-click a service, then click Set Event Logging Level. In the dialog box, select the desired trace level and then click OK.
To collect event logs (trace logs) in HPC Pack 2008 R2
In HPC Cluster Manager, click Job Management, and then click All Jobs.
In the job list, find the job that is associated with the session that you are debugging. The job ID is the same as the session ID.
Right-click the job, and then click Collect Trace.
Note
Collect Trace is not available in HPC Pack 2012.
In the Collect Trace dialog box, specify the shared folder where you would like to collect the trace logs. The folder must be accessible from the compute nodes.
Verify that the trace logs appear in the specified folder.
Right-click the job, and then click Delete Trace to delete the trace logs from the compute nodes.