On Fri, 10/03/03 at 19:15:52 +0100, Toad wrote: > On Sun, Sep 28, 2003 at 07:58:48AM -0700, Brandon Low wrote: > > Modified Files: > > OpenConnectionManager.java > > Log Message: > > Fix deadlock in the changes moving sync(lru) outside of KillSurplusConnections. > > No more deadlocks, and since we now remove from lru _before_ actually going to > > terminate the connection this still ensures that we don't kill too many > > Does this cause us to go over the limit more than we would otherwise, > or is it reasonably safe? Do we have any evidence?
Still can't go over the limit any more than we already could, no thread will exit the OCM.put() until after any excess connections are closed (this is the same as the behaviour we had before) in order to ensure that we are _never_ over the limit we would have to get down to max-1 open before we add, and basically do something like QThreadFactory does (WRT synched counts, and pool of maxConnections size)... --Brandon > > -- > Matthew J Toseland - [EMAIL PROTECTED] > Freenet Project Official Codemonkey - http://freenetproject.org/ > ICTHUS - Nothing is impossible. Our Boss says so. _______________________________________________ Devl mailing list [EMAIL PROTECTED] http://dodo.freenetproject.org/cgi-bin/mailman/listinfo/devl
