NetBSD-Users archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: Please help Apache Segmentation fault



Hi,

On Mon, 7 Jan 2013, Takahiro Kambe wrote:
In message <20130106153757.GA20183%kyyhky.embedtronics.fi@localhost>
        on Sun, 6 Jan 2013 17:37:57 +0200,
        Jukka Marin <jmarin%embedtronics.fi@localhost> wrote:
On Sun, Jan 06, 2013 at 10:30:27AM -0500, Al wrote:
Hello,

I am seeing "[notice] child pid 16566 exit signal Segmentation fault (11)"
in my apache error_log. There is something like 66790 in maybe 1 or 2
days.

Apache is crashing on our server as well (ever since we upgraded to NetBSD
6.0).  Haven't had the time to sort this out yet..
I've seen it, too.

Yes, since upgrading to NetBSD 6.0.

   NetBSD 6.0_STABLE (HPMICRO) #7: Wed Dec 12 11:29:58 JST 2012

It looks like apache24 from pkgsrc has the problem but apache22
dosen't have it.  In my case,

I am running aapche22.

* Crash happens on reloading configuration or log rotation.
* System V IPC, semaphores are left after crash and need to remove
 them manually with ipcrm(1).

I am not sure when I am getting the core dumps. I can say this, if I delete my apache error_log and then restart it, then within minutes I have hundreds of segmentation faults.

Maybe I don't understand what I am looking at, but I only see one semaphores:

ipcs -s
IPC status from <running system> as of Sun Jan  6 11:31:39 2013

Semaphores:
T        ID     KEY        MODE       OWNER    GROUP
s    131072          0 --rw-------  webuser webgroup

Would this be enough to overwhelm sokva?

Also, would this remove them:
"for i in $(ipcs -s | grep www | awk '{print $2}'); do ipcrm sem $i; done"

And will it effect pure_ftpd?

Thanks,
Al


Home | Main Index | Thread Index | Old Index