On Wed, Nov 3, 2010 at 4:28 PM, Stefan Fritsch <[email protected]> wrote:
> On Wednesday 03 November 2010, Stefan Fritsch wrote:
>> some of the recent changes in the proxy code introduced some
>> regressions
>
> Strange, I have these problems only with prefork, not with event.
> But with event I get a segfault in the reslist cleanup code.
>
> Can somebody cross-check this?
look familiar?
httpd: misc/apr_reslist.c:161: reslist_cleanup: Assertion `rl->ntotal
== 0' failed.
[Tue Nov 02 07:59:32.863718 2010] [core:notice] [pid 26494:tid
3078747840] child pid 26495 exit signal Aborted (6), possible coredump
in /home/trawick/ApacheCon/23inst
(gdb) where
#0 0x008c1422 in __kernel_vsyscall ()
#1 0x0013a651 in *__GI_raise (sig=6) at
../nptl/sysdeps/unix/sysv/linux/raise.c:64
#2 0x0013da82 in *__GI_abort () at abort.c:92
#3 0x00133718 in *__GI___assert_fail (assertion=0x31c7a2 "rl->ntotal == 0",
file=0x31c780 "misc/apr_reslist.c", line=161, function=0x31c7b2
"reslist_cleanup")
at assert.c:81
#4 0x00315b4c in reslist_cleanup (data_=0x96ec250) at misc/apr_reslist.c:161
#5 0x00dedb5e in run_cleanups (pool=0x96ec210) at memory/unix/apr_pools.c:2308
#6 apr_pool_destroy (pool=0x96ec210) at memory/unix/apr_pools.c:779
#7 0x00dedb40 in apr_pool_destroy (pool=0x96ea208) at
memory/unix/apr_pools.c:776
#8 0x00559959 in clean_child_exit (child_num_arg=<value optimized
out>) at worker.c:379
This is with HEAD of apr-util 1.4.
