need help with the this.. apache keeps shutting down

at least once a month apache (Apache/1.3.26 (Unix) ) keeps shutting
down... the error logs just says this:

[Mon Jun 19 13:09:41 2006] [notice] caught SIGTERM, shutting down

We are running :

SunOS dmocic 5.8 Generic_117350-06

Any help or ideas will be greatly appreciated..

thanks

On 7 Jul 2006 13:23:19 -0700, iftikhar.barrie@gmail.com wrote:
> at least once a month apache (Apache/1.3.26 (Unix) ) keeps shutting
> down... the error logs just says this:
>
> [Mon Jun 19 13:09:41 2006] [notice] caught SIGTERM, shutting down
>
> We are running :
>
> SunOS dmocic 5.8 Generic_117350-06
>
> Any help or ideas will be greatly appreciated..

Upgrade?

Solaris 8 and apache 1.3.26 are ancient. Even cavemen had more advanced
web servers :-)

1.3.36 is available if you absolutely must stay with 1.3

--
* John Oliver http://www.john-oliver.net/ *

thanks.. unfortunatley it (apache) is bundled with another piece of
software supplied by one vendor..

John Oliver wrote:
> On 7 Jul 2006 13:23:19 -0700, iftikhar.barrie@gmail.com wrote:
> > at least once a month apache (Apache/1.3.26 (Unix) ) keeps shutting
> > down... the error logs just says this:
> >
> > [Mon Jun 19 13:09:41 2006] [notice] caught SIGTERM, shutting down
> >
> > We are running :
> >
> > SunOS dmocic 5.8 Generic_117350-06
> >
> > Any help or ideas will be greatly appreciated..
>
> Upgrade?
>
> Solaris 8 and apache 1.3.26 are ancient. Even cavemen had more advanced
> web servers :-)
>
> 1.3.36 is available if you absolutely must stay with 1.3
>
> --
> * John Oliver http://www.john-oliver.net/ *

On 7 Jul 2006 13:44:18 -0700, IIB wrote:
> thanks.. unfortunatley it (apache) is bundled with another piece of
> software supplied by one vendor..

Go to them for support? Maybe they'll get tired of trying to find
answers for such ancient software and decide to upgrade.

Other than that... I suppose you could strace apache, dump the output to
a log file, and rotate it at reasonable intervals. I imagine they'd
grow pretty big pretty fast.

You might also set up MRTG to monitor load, RAM, throughput and visitors
for Apache, etc. Maybe you'll notice some kind of pattern before a
crash.

Maybe there's an additional debug option you could use?

Oh, and you could also maybe start apache from supervise
http://cr.yp.to/daemontools.html instead of the runlevel scripts. When
it died, it'd be restarted again automatically. Hopefully, whatever is
causing the crash is a one-time sort of event, and not a condition that
might get into a race with supervise.

--
* John Oliver http://www.john-oliver.net/ *