Hello folks.

I'm facing the same problem with ossec-csyslogd daemon.
Every time I start the process, it crashes after a few minutes.

I've tried to get some gdb traces as asked and here is what I get:

Starting program: /var/ossec/bin/ossec-csyslogd
warning: no loadable sections found in added symbol-file system-supplied DSO
at 0x2aaaaaaab000
[New process 503]
Program received signal SIGSEGV, Segmentation fault.
[Switching to process 504]
0x000000000040219f in inet_addr ()


(gdb) backtrace
#0  0x000000000040219f in inet_addr ()
#1  0x00000000004024bd in inet_addr ()
#2  0x000000000040289f in inet_addr ()
#3  0x00000031c081d994 in __libc_start_main () from /lib64/libc.so.6
#4  0x0000000000401d79 in inet_addr ()
#5  0x00007fffffffea38 in ?? ()
#6  0x0000000000000000 in ?? ()

The version installed is :
Thanks,

Regards
-
Bruno

-----Message d'origine-----
De : [email protected] [mailto:[email protected]] De la
part de blacklight
Envoyé : mardi 7 juin 2011 23:43
À : ossec-list
Objet : [ossec-list] Re: Concern about the ossec-csyslogd daemon

If I were to put this daemon under gdb, I am concerned that I could be
accumulating debugger data this for weeks before this daemon crashes again.
Hopefully, this daemon crash is a once in a blue moon event. On the other
hand, once in a blue moon events are very hard to troubleshoot. If it's
indeed a once in a blue moon event, I'll live with that. BTW, I haven't
found anything in the /var/log/messages that even hint at a crash. And from
reading the /var/ossec/logs/ossec.log at the time of the crash, you'd think
that the OSSEC service was the picture of health.

On Jun 7, 2:16 pm, Daniel Cid <[email protected]> wrote:
> It shouldn't segfault even during a package update... If any of you 
> can run it under gdb, it would be awesome :)
>
> thanks,
>
> On Tue, Jun 7, 2011 at 1:44 PM, Jefferson, Shawn
>
>
>
>
>
>
>
> <[email protected]> wrote:
> > I looked back through my logs and here is the alert:
>
> > ossec-alerts-06.log:Jun  6 10:12:55 bcfossec kernel: [501421.634671] 
> > ossec-csyslogd[3014]: segfault at 0 ip b7775821 sp bfc4ffbc error 4 
> > in libc-2.11.1.so[b7702000+153000]
>
> > To the original poster: what OS are you running your OSSEC server on?
I'm on Ubuntu 10.04.2 LTS.  I wonder if the segfault was caused by some
package being updated/upgraded?
>
> > -----Original Message-----
> > From: [email protected] 
> > [mailto:[email protected]] On Behalf Of Daniel Cid
> > Sent: Monday, June 06, 2011 6:48 PM
> > To: [email protected]
> > Subject: Re: [ossec-list] Concern about the ossec-csyslogd daemon
>
> > At least OSSEC is reporting it :) And yes, try to run it under gdb 
> > so we can see where it is crashing. Or try the latest snapshot to 
> > see if it works there.
>
> > Thanks,
>
> > On Mon, Jun 6, 2011 at 6:58 PM, dan (ddp) <[email protected]> wrote:
> >> Please try running it under gdb:
>
> >> gdb ossec-csyslogd
>
> >> (gdb) set follow-fork-mode child
> >> (gdb) run
>
> >> On Mon, Jun 6, 2011 at 5:50 PM, Jefferson, Shawn 
> >> <[email protected]> wrote:
> >>> Hey, I had the same crash too!
>
> >>> -----Original Message-----
> >>> From: [email protected] 
> >>> [mailto:[email protected]] On Behalf Of blacklight
> >>> Sent: Monday, June 06, 2011 2:36 PM
> >>> To: ossec-list
> >>> Subject: [ossec-list] Concern about the ossec-csyslogd daemon
>
> >>> Hello Folks,
>
> >>> I have a concern about the csyslogd demon:
>
> >>> 2011 Jun 04 13:51:03 Rule Id: 151601 level: 7
> >>> Location: ossec-server->/var/log/messages Grouping of kernel error 
> >>> rules.
> >>> Jun 4 13:51:02 ossec-server kernel: ossec-csyslogd[21507]: 
> >>> segfault at
> >>> 0000000000000000 rip 0000003dd8479a30 rsp 00007fff23ba3a88 error 4
>
> >>> The ossec-csyslogd daemon crashed over the weekend over a single 
> >>> segfault. I have no idea what caused this segfault. I am worried 
> >>> that this daemon is less than rock solid.
>
> >>> Regards,

Reply via email to