Hello!

On Wed, Dec 17, 2014 at 08:02:47PM -0500, tatroc wrote:

> I folks, wondering if I could get your input.
>
> I have a nginx 1.7.3 configured as a reverse proxy. The system seems to work
> fine when not under heavy load. However, there are periods of time when we
> have extreme load because of ecom promotions.
>
> The $request_time gets up to 15 seconds. Normally the $request_time is only
> a few seconds. The client machine is making a web service call to the Nginx
> reverse proxy and the client web service making the call is designed to
> abort the connection if it takes longer than 15 seconds. As a result that
> is why you see all the $request_time's stop at 15 seconds.
>
> The $upstream_addr field is populated with “-“. What would cause the
> $request_time to take longer than 15 seconds?
> It seems the nginx reverse proxy never passes the connection on to the
> upstream server.
>
> 10.8.165.116 - - [13/Dec/2014:07:58:12 -0600] "POST
> /Common/service/GateService/GateService.svc HTTP/1.1" 499 0 "-" "JAX-WS RI
> 2.1.6 in JDK 6" "-" LB_req_Time: 15.000 WebSrv_rspTime: - Req_size: 8292
> HTTP_content_size: 7875

[...]

As requests are POSTs, likely the problem is with sending request
body. Probably due to network problems.

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

_______________________________________________
nginx mailing list
nginx@nginx.org
http://mailman.nginx.org/mailman/listinfo/nginx