Noel J. Bergman wrote: > Stefano Bagnara wrote: > >> Noel J. Bergman wrote: >>> An increase in traffic would not result in a consistent ~2MB per day >>> decrease in available heap space after a GC. By definition, memory >>> that is consumed and not released is a memory leak. > >> No, I don't agree with this definition of leak. >> Even Wikipedia have a better definition: > >> "In computer science, a memory leak is a particular kind of >> unintentional memory consumption by a computer program where >> the program fails to release memory when no longer needed" > > Exactly. And I don't believe that any of us are going to say that JAMES is > intentionally holding on to an extra 2MB per day. So it is unintentional > and no longer needed. :-) Either that or we've identified something where > we have not properly controlled an upper limit, e.g., the DNS cache should > be much lower or the heap size much higher (I don't believe this to be the > case).
Right, this last sentence simply describe one of our alternative to leaking: we may simply make misuse/abuse of the memory because of non-weak caches with bad defaults or similar things... This is why I still say we're looking to confirm this is a leak. I'm not ignoring this, I just think that we need some more concrete proof in order to call it "confirmed leak". I guess that 50% the problem is in james or bundled libraries, 30% it is in connector/j/mm-mysql, 20% it is not a leak. This 50% is what is keeping me from ignoring the bug and stop reading this thread. It has nothing to do with my "personal desires" ;-) Stefano --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
