Quantcast

upstream timed out

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

upstream timed out

c0nw0nk
I am a newbie.

I'm running Ubuntu, Nginx and 5 Thin servers - that serve up a Ruby on Rails
app.

The system hangs about once a day.

The Nginx log has this error:
2013/07/11 10:06:46 [error] 21344#0: *201 upstream timed out (110:
Connection timed out) while reading response header from upstream, client:
192.168.10.11, server: ,

and

2013/07/11 10:05:11 [warn] 21345#0: *225 an upstream response is buffered to
a temporary file /opt/bitnami/nginx/tmp/proxy/2/00/0000000002 while reading
upstream, client: 192.168.10.11, server: , request: "GET
/assets/application-b7b1695e978934a03ade49d20bf63139.js HTTP/1.1",
upstream:

Is this telling me there is a problem with Thin or Ruby on Rails?
I'm not seeing error in logs for those systems.

Thanks,
Reddirt

Posted at Nginx Forum: http://forum.nginx.org/read.php?2,240762,240762#msg-240762

_______________________________________________
nginx mailing list
[hidden email]
http://mailman.nginx.org/mailman/listinfo/nginx
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: upstream timed out

Francis Daly
On Thu, Jul 11, 2013 at 12:43:06PM -0400, Reddirt wrote:

Hi there,

> The Nginx log has this error:
> 2013/07/11 10:06:46 [error] 21344#0: *201 upstream timed out (110:
> Connection timed out) while reading response header from upstream, client:
> 192.168.10.11, server: ,

That says that as far as nginx is concerned, its upstream (presumably
a Thin server) took too long before returning useful content. That
suggests a problem on that Thin server, or with the "timeout" values
that nginx and Thin have being different.

> 2013/07/11 10:05:11 [warn] 21345#0: *225 an upstream response is buffered to
> a temporary file /opt/bitnami/nginx/tmp/proxy/2/00/0000000002 while reading
> upstream, client: 192.168.10.11, server: , request: "GET
> /assets/application-b7b1695e978934a03ade49d20bf63139.js HTTP/1.1",
> upstream:

That just says that the response was too big to fit in nginx's memory
buffers, and so was written to disk. Not a problem, unless you don't
expect the response to be that big.

        f
--
Francis Daly        [hidden email]

_______________________________________________
nginx mailing list
[hidden email]
http://mailman.nginx.org/mailman/listinfo/nginx
Loading...