[peruser] Patch for busy servers
marcelo at tpn.com.br
Tue Jan 4 11:29:51 MST 2011
I'm trying to use ab to reproduce the problem with:
ab -c 40 -n 1000000 http://server/
without success. What parameters are you using?
marcelo at mco2.com.br
On Jan 4, 2011, at 4:07 PM, Janno Sannik wrote:
> to clarify things how to reproduce this - the full recipie is to use the provided configuration and run ab on server and then make graceful restart in the meantime.
> On 4.01.2011 18:36, Marcelo Coelho wrote:
>> 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.
>>> Dave Steinberg
>>> Peruser mailing list
>>> Peruser at telana.com
>> Peruser mailing list
>> Peruser at telana.com
> Peruser mailing list
> Peruser at telana.com
More information about the Peruser