On Wed, 8 Sep 2004 13:32:09 -0400, Jeff Trawick <[EMAIL PROTECTED]> wrote: > On Wed, 08 Sep 2004 09:14:56 -0600, Jean-Jacques Clar <[EMAIL PROTECTED]> wrote:
> > > > Sorry but when running the same test on SLES9 with 2.0.51rc1, > > my error_log is full of threads segfaulting. > > Next I'll try to replicate with 2.0.51 rc2 on the same box. yep chunk_free() free() cleanup_cache_object() decrement_refcount() run_cleanups() apr_pool_destroy() ap_process_http_connection() ap_run_process_connection() process_socket() worker_thread() interestingly, I'm not getting the "exit signal" message in httpd.conf; the crashing thread has left its core.PID but it is a zombie so far after applying your patch I haven't hit a problem, though I'm still waiting on segfaults just in case I find the possible need for the patch very disturbing; it implies to me a problem in the atomics code
