Since I have been profiling for 3 days, there are thousands and thousands of instances of MergeThread (org.apache.lucene.index.ConcurrentMergeScheduler$MergeThread) which I presume correlates to the garbage collection events and re-using threads since there's probably a max number of active threads.
Since this is a TEST system, is there ANY reason to have this many threads launched, re-used, and garbage collected? I've run indexing and run searches and opened items but not thousands and thousands :-) With no users hitting the system, I would expect a dozen threads have been needed over the past 3 days but something is launching and launching... Our version of Tomcat is 8.0.23 and there is an upgrade available to 8.0.30. Is that false hope to upgrade and have a different experience? If DSpace code isn't launching these threads (is that what the screen capture <http://i.stack.imgur.com/Rg8Ba.png> reveals?) then can we focus on Tomcat? Bonus info...there are also hundreds of instances of java.lang.Thread recorded in the past 3 days which all reportedly lasted for 60,000 milliseconds (60 seconds). Although there is 1 java.lang.Thread has lasted for 258,752,817 ms or 71 hours(!) -- You received this message because you are subscribed to the Google Groups "DSpace Technical Support" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/dspace-tech. For more options, visit https://groups.google.com/d/optout.
