NASIG 2017

When the University of North Texas Libraries first implemented Summon as their web-scale discovery service, the staff encountered many anecdotal reports of problems with the service, but often were unable to replicate the issues themselves due to lack of concrete details. With the complexity involved with diagnosing problems, most patrons were unable to provide the level of detail needed to determine what the root cause of an issue might be. Realizing that they needed a more effetive way to gather information about these phantom problems, in 2013 they added an error reporting link to their results pages which harvested metadata about specific citations and links, enabling them to begin effectively troubleshooting the errors. Since that time, they have recieved over 7,000 error reports for review and have developed their workflows to resolve these issues as best they can with limited staff time.  This presentation will walk through the basic troubleshooting workflows, provide a breakdown of the types of errors encountered, and share some analysis of the most frequently encountered errors and their causes.