nginx doesn't seem to register configuration changes

[Maxim Dounin]

Hello!

On Tue, Apr 01, 2014 at 12:34:21PM +0200, Nicholas Wieland wrote:

> Hi *, I’m using nginx as reverse proxy for some puma backends 
> via unix socket.
> The problem I’m having right now is that even after several 
> reloads nginx doesn’t seem to use the changes I did to the 
> configuration. I honestly have no idea what to try as this is 
> definitely very weird. The problem appears to be the upstream 
> directive, nginx keeps using the old url to the old socket, even 
> though I changed it.
> 
> deployer at demo:~$ uname -a
> Linux demo.ec.thefool.it 3.2.0-37-virtual #58-Ubuntu SMP Thu Jan 24 15:48:03 UTC 2013 x86_64 x86_64 x86_64 GNU/Linux
> deployer at demo:~$ sudo nginx -v
> nginx version: nginx/1.4.1
> 
> My conf:
> https://gist.github.com/ngw/91312b5602816cfb2632
> 
> The error:
> 014/04/01 12:10:43 [crit] 30954#0: *1 connect() to unix:///home/deployer/apps/conversationflow/puma.sock failed (2: No such file or directory) while connecting to upstream, client: 93.51.167.60, server: demo.ec.thefool.it, request: "GET / HTTP/1.1", upstream: "http://unix:///home/deployer/apps/conversationflow/puma.sock:/", host: “demo.ec.thefool.it"
> 
> This is the only error I get.
> As you can see puma.sock is in the wrong place, the correct one 
> is the one I configured (obviously).
> I’ve also tried to change the socket path to something I made 
> up, and nginx registers the change and behaves accordingly. If I 
> change the socket path to the real one, here we go and it 
> doesn’t use it…
> Any suggestion?

Take a look at global error log as defined in your nginx.conf, likely 
it has an explanation.

-- 
Maxim Dounin
http://nginx.org/