Upstream failures against Mongrels instances

[Ken]

actually, "proxy_next_upstream" works in despite of the error log. i used
apachebench to hammer my web server, and continually restarted a couple of
mongrel instances, but there is no request failed.

so, i think the error messages are just "notification".

On 8/14/07, Ken <2828628 at gmail.com> wrote:
>
> i meet the same problem too.
>
> On 8/14/07, Alexander Staubo <alex at purefiction.net> wrote:
> >
> > On 8/13/07, Ezra Zygmuntowicz <ezmobius at gmail.com> wrote:
> > >         When you say you kill and respawn the mongrels, do you mean
> > you kill
> > > -9 them and then restart?
> >
> > No, I mean SIGTERM. This signals a graceful stop, equivalent to
> > calling the stop method on all Mongrel listeners.
> >
> > So there's no reason why connections should be going down gracefully.
> > However, Nginx ought to behave correctly even when a process is
> > terminated with SIGKILL -- ie., it should follow whatever
> > proxy_next_upstream is set to.
> >
> > Thanks for the Monit script; I don't know if we will ever use Monit
> > (we outsource our sysadmin/management, so it's mostly out of our
> > hands), but it might come in handy.
> >
> > Alexander.
> >
> >
>
>
> --
> 君問歸期未有期，巴山夜雨漲秋池。何當共剪西窗燭，卻話巴山夜雨時。




-- 
君問歸期未有期，巴山夜雨漲秋池。何當共剪西窗燭，卻話巴山夜雨時。
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://nginx.org/pipermail/nginx/attachments/20070814/c13da1f7/attachment.html>