On Fri, Sep 11, 2009 at 08:56:13AM +0800, Jeffrey 'jf' Lim wrote:

> 2009/9/10 Igor Sysoev <is@rambler-co.ru>
>
> > On Thu, Sep 10, 2009 at 03:10:05PM +0400, Maxim Dounin wrote:
> >
> > > Hello!
> > >
> > > On Thu, Sep 10, 2009 at 06:29:18AM -0400, gloomen wrote:
> > >
> > > > Friends, help deal with nginx. I can not go to the site.
> > > >
> > > > 502 Bad Gateway
> > > > nginx/0.7.61
> > > >
> > > > ee /var/log/nginx-error.log
> > > >
> > > > 2009/09/10 14:58:19 734#0: *3 upstream prematurely closed connection
> > while reading response header from upstream, client: 10.10.10.10, server:
> > xxx.net, request: "GET / HTTP/1.1", upstream: "fastcgi://127.0.0.1:9000",
> > host: "xxx.net"
> > >
> > > So upstream probably just died. Likely to be php-related issue,
> > > try looking there.
> > >
> > > Anyway it isn't "bugs in nginx".
> >
> > There is one nginx fastcgi bug when fastcgi sends very large headers.
> > The attached patch fixes this. In other cases it's more like backend error.
> >
> >
> and I assume this problem is only for fastcgi? (ie., not happening for http
> proxies)

Yes, this is fastcgi only bug.


--
Igor Sysoev
http://sysoev.ru/en/