We've determined that if MaxSpareProcessors is set, it will start giving
the long lost child came home errors.
<IfModule peruser.c>
ServerLimit 700
MaxClients 700
MinSpareProcessors 1
MaxSpareProcessors 20
MinProcessors 0
MaxProcessors 80
MaxRequestsPerChild 1000
ExpireTimeout 7200
IdleTimeout 120
MinMultiplexers 3
MaxMultiplexers 40
MultiplexerIdleTimeout 120
ProcessorWaitTimeout 5
</IfModule>
We've been using graceful for some time (at least a year). Haven't got
any major problems. of course we have quite mini system compared to your
setup.
Janno
On 4.01.2011 18:36, Marcelo Coelho wrote:
Hi,
I was using RC2 and graceful restart never worked well for me.
I have more than 1,000 vhosts and "apachectl restart" is faster and more stable than
"apachectl graceful".
Can someone tell me how to reproduce the "long lost child came home" error?
--
Marcelo Coelho
marcelo at mco2.com.br
On Jan 4, 2011, at 2:11 PM, Dave Steinberg wrote:
I'm seeing this as well, with the V3 patch on FreeBSD 7. I am transitioning to
using mpm-peruser still, I have around around 5 server environments, one with
~1000 vhosts (handled by mod_vhost_alias), and ~4 vhosts in their own
individual server environments.
I'm also seeing some segfaults, but I haven't dug in enough yet to see where
they're coming from. Does anybody have any advice on tracking that down?
PS - Great work on the patches, despite these problems, things appear to be
generally working well.
Regards,
--
Dave Steinberg
http://www.geekisp.com/
http://www.steinbergcomputing.com/
http://www.redterror.net/
_______________________________________________
Peruser mailing list
[email protected]
http://www.telana.com/mailman/listinfo/peruser
_______________________________________________
Peruser mailing list
[email protected]
http://www.telana.com/mailman/listinfo/peruser
_______________________________________________
Peruser mailing list
[email protected]
http://www.telana.com/mailman/listinfo/peruser
