First of all, to generate thread dumps and see where threads holding write locks might be stuck, see generating thread dumps wiki: http://www.jboss.org/community/wiki/StackTrace I must say that 100ms for lock timeout is pretty small. Something around 10-20 seconds is more usual. Rather than increasing the lock timeout to insane numbers, use thread dumps to figure out what's holding up the progress or if there're any deadlocks...etc There're even tools like TDA or Samurai that can help analyze them. WRT "Caused by: java.lang.IllegalStateException: Cache not in STARTED state!" message, that looks strange, did you check whether stop/destroy was called at some point? Finally, when you say that your cache usage is very write heavy, do you mean that you call attach() loads of times? Or that once you've attached a Pojo to the cache, you modify it via its set methods very often? View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4239697#... Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&a...