Load Balancer Logs / Log Levels

I’m trying to create the follow setup:

Load Balancer -> nginx -> PHP fast CGI

If I go straight to nginx everything works fine.

However, if I go through the load balancer I get:

502 Bad Gateway

The server returned an invalid or incomplete response.

In the nginx container I see the following:

11/02/2017 10:03:562017/02/11 10:03:56 [info] 8#8: *28 epoll_wait() reported that client prematurely closed connection, so upstream connection is closed too (104: Connection reset by peer) while reading upstream, client: 10.42.61.208, server: , request: "GET / HTTP/1.0", upstream: "fastcgi://10.42.201.15:9000", host: "test.example.com", referrer: "http://test.example.com/"
11/02/2017 10:03:5610.42.61.208 - - [11/Feb/2017:10:03:56 +0000] "GET / HTTP/1.0" 200 7997 "http://test.example.com/" "Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 5.1; Trident/4.0; .NET CLR 2.0.50727; .NET CLR 3.0.04506.648; .NET CLR 3.5.21022; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729; InfoPath.2)"

I’d like to enable info/debug logging in the load balancer in order to get more information about what is wrong. Is this possible with the Rancher load balancer?

I’ve managed to reproduce this problem with a local compose stack using bare haproxy (and got logs working) so I guess I’m not going to need LB logging in Rancher to get to the bottom of this.

FYI for anyone whose interested I’ve posted a question here: