Tracing the cause of errors in SharePoint can leave you wondering where to start. In this article I provide some tips on preparing your SharePoint farm with error tracing in mind.

In this example I will assume we have a SharePoint farm with two web front ends (with network load balancing) and an application server. The same method can be used on larger farms.

This method is useful for identifying problems that occur on one server in your farm but not others. The are usually intermittent issues that occur randomly for some users and not others (due to different Web Front Ends servicing the requests).

Configuring SharePoint and IIS 

Alternative Access Mappings

• In Central Admin, configure alternative access maps for each server in the farm using the servers actual name e.g. SP_WFE1, SP_WFE2, SP_APP1
• Test by connecting to your SharePoint servers using HTTP://servername

Configured IIS

• Logon to each server and modify the IIS Bindings to include a host header with the servers actual name (matching the alternative access mapping)

Download ULSviewer from MSDN and put a copy on each web front end.

Trapping errors

The trick to catching errors is to reproduce the error on a known host and capture the error with ULSviewer.

• Choose a server e.g. SP_WFE1
• Logon to the server and run ULSviewer
• Reproduce the error using SP_WFE1 rather than the normal SharePoint URL
• Copy the error correlation code from the error message
• Create a filter in ULSviewer
• Filter by Correlation and paste the code copied in the step above
• If multiple errors are returned in the log, look for the words error or exception (usually)
• Highlight the error and full details will appear in the top of ULSviewer pane.