I was called into Ecolab over in St. Paul today to fix this problem. Even though the SharePoint 2013 environment was functional, nothing really broke on that front, within the event log there were a whole bunch of 6398 errors happening.
There error text that was being bubbled up was:
The Execute method of job definition Microsoft.Office.Server.UserProfiles.LMTRepopulationJob (ID f6b8af32-0dbb-4459-8f0b-f9ed376e9324) threw an exception. More information is included below.
An update conflict has occurred, and you must re-try this action. The object LMTRepopulationJob Name=User Profile Service Application_LMTRepopulationJob was updated by EcoLab\SP_ServiceApps, in the w3wp (4952) process, on machine SPWFE01. View the tracing log for more information about the conflict.
As well, the following error was being thrown fairly frequently:
The Execute method of job definition Microsoft.Office.Server.UserProfiles.LMTRepopulationJob (ID d55dfdda-c513-415b-9809-77aba0cce788) threw an exception. More information is included below.
Unexpected exception in FeedCacheService.BulkLMTUpdate: Region not found..
Fixing this error is not trivial. Generally, this error will come up if the WFE is being frequently restarted, where the the AppFabric Caching Service is likewise restarted which will cause the cache region to be rebuilt. Following, no LMT region is present, and this error will present itself.
Now, this would normal beg the question, how can SharePoint assume that the LMT region will always be present? The only reason that an LMT region wouldn’t be present would be if relevant timer jobs are not configured correctly. As the timer job will run every 5 minutes, you will see this error if it is misconfigured every 5 minutes. So a whole lot of entries indeed?