(im not a subscriber to this list, so please CC to my address) Joe Orton wrote: > On Mon, Nov 13, 2006 at 01:40:46PM +0100, Peter Steiner wrote: >> I have observed and opened a bug [1] in allocator_free() in apr_pools.c, >> that was reported first by Brian Pane on [EMAIL PROTECTED] [2]. Brian >> drafted a >> patch: > > The issues Brian reference look real but are entirely unrelated to the > symptoms you have described - they cannot cause a "memory leak" in your > app. If the memory use in your app is increasing *indefinitely* then > that's a pool memory misuse problem. > > Check that e.g. all pools you create are also destroyed and that no > memory is being allocated out of long-lived pools by whatever operation > it is you see consuming memory.
Thanks to Joe for directing my search in the right direction. I found that my application (or rather log4cxx) didn't set the threshold with apr_allocator_max_free_set() AND had some small allocation in its long-lived pool. I have proposed a patch for the original issue (the underflow of current_free_index), see http://issues.apache.org/bugzilla/show_bug.cgi?id=40955. While hunting down my memory problems I did use the APR_POOL_DEBUG switch and had to fix two small problems first. I'm proposing another patch for these bugs, see http://issues.apache.org/bugzilla/show_bug.cgi?id=41063 Regards, Peter -- _ _ Peter Steiner <[EMAIL PROTECTED]> / /_/ / Hug-Witschi AG <http://www.hugwi.ch/> / _ / Electronic Engineering /_/ /_/ _ _ Auriedstrasse 10 / / / / / / CH-3178 Boesingen / /_/ /_/ / Tel +41 31 740 44 44 /_ _ _ _ _/ Fax +41 31 740 44 45
