httpd-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Rainer M. Canavan" <>
Subject Re: [users@httpd] Segmentation fault in logs
Date Fri, 23 May 2014 10:40:31 GMT

On May 23, 2014, at 11:42 , Hugo Gomes <> wrote:

> Hi Guys, 
> 	I have httpd-2.2.3-85.el5.centos in CentOS release 5.10, i and see all
> days, more than 10 times, the error in httpd logs:
> [Thu May 22 01:20:15 2014] [notice] child pid 561 exit signal
> Segmentation fault (11)
> 	I've investigated logs and don't see nothing wrong with any scripts,
> this happens in random times and in random hours each day!
> 	I've google it and found something with some library
> ( ), does
> anybody can help me?
> 	Thanks in advance,
> 	Hugo Gomes

You'll have to save core dumps and analyze them with gdb, since the 
process that segfaulted most likely didn't log the request that caused
the crash. There's documentation here:

The important first steps are setting CoreDumpDirectory and ulimit -c unlimited.
As soon as you have a core dump, grab a .gdbinit from the appropriate apache
source (or package), install the debug info for apache and modules (in redhat-like
Linux distributions debuginfo-install may help), and check the request record. You
may want to print r->the_request and dump_table (r->headers_in) to see some info
that may help you reproduce the problem.

Segfaults may indicate a security relevant problem. If you've found that the 
problem is indeed in apache httpd and could exploited remotely, e.g. to crash
the server, you may want to read and

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message