It's two days that one of our SQL Server(2008 R2 running on Windows 2008) instances become unavailable and useres start to recieve login failure.Both incidences have happened almost at the same time around 09:00AM.
when I checked the windows log I found the following event:
EventId : 17884
event Description: New queries assigned to process on Node 2 have not been picked up by a worker thread in the last 60 seconds. Blocking or long-running queries can contribute to this condition, and may degrade client response time. Use the "max worker threads" configuration option to increase number of allowable threads, or optimize current running queries. SQL Process Utilization: 80%%. System Idle: 14%%.
I analyzed the dump file and notices that most of threads have sqlservr!CTraceWriteRequest::InitForRowsetTrace(If most of threads are stuck while trying to write profiler events to the destination)
Does this mean a trace is being run against that server and that has caused the problem?