502 Bad Gateway

Raphael Marot rm at rmax.be
Mon Mar 26 12:01:17 UTC 2012

Le 26/03/2012 13:45, mailing at securitylabs.it a écrit :
> Il 26/03/2012 13:41, Raphael Marot ha scritto:
>> Hi all,
>> I have a simple setup with Nginx in front Apache, under debian wheezy 
>> (nginx for static content, apache for php)
>> Since the last Nginx update (1.1.17-2 via apt-get), every night after 
>> daily crons i have a 502 error (bad gateway) until i reload apache.
>> Everything was fine before.
>> I tried to reload apache after nginx in the logrotate daily task, but 
>> without effect.
>> In my logs i just have :
>> [error] 22332#0: *10804 connect() failed (111: Connection refused) 
>> while connecting to upstream
>> Where should i look at?
> Look at Apache's logs. It seems that apache does not restart correctly 
> after che log rotate, and a manual restart fix it. The error tells you 
> that nginx  can't communicate with apache.
> _______________________________________________
> nginx mailing list
> nginx at nginx.org
> http://mailman.nginx.org/mailman/listinfo/nginx

Indeed I have an error in my apache log.
seg fault or similar nasty error detected in the parent process

It is definitely php related and has nothing to do with Nginx.
I think i uninstalled a php module which is still called in the php.ini, 
and maybe this is the cause...

More information about the nginx mailing list