Server Fault is a question and answer site for system and network administrators. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I recently set up a NodeBalancer in front of a Rails 3.2.12 app. The app is served by nginx and Unicorn.

All seems to work fine, but I get a lot of errors such as these that I didn't get when I only had 1 server.

actionpack (3.2.12) lib/action_dispatch/middleware/remote_ip.rb:55:in `calculate_ip'

Here is my nginx configuration for the app.

upstream unicorn {
server unix:/tmp/unicorn.ahotu-calendars.sock fail_timeout=0;

server {
listen 80 default deferred;
root /home/deployer/apps/appdirectory/current/public;
if (-f $document_root/system/maintenance.html) {
return 503;
error_page 503 @maintenance;
location @maintenance {
rewrite ^(.*)$ /system/maintenance.html last;

location ^~ /assets/ {
gzip_static on;
expires max;
add_header Cache-Control public;

try_files $uri/index.html $uri @unicorn;
location @unicorn {
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
proxy_set_header Host $http_host;
proxy_redirect off;
proxy_pass http://unicorn;

error_page 500 502 503 504 /500.html;
client_max_body_size 4G;
keepalive_timeout 10;

Did I do something wrong in the config file?

Thank you.

share|improve this question
Do you recognize any of those IP addresses? – Michael Hampton Mar 15 '13 at 10:57
No. And they change on every error. – jlfenaux Mar 15 '13 at 13:37

To prevent clients from forging a proxy's headers, both X-Forwarded-For and Client-Ip have to equal to not raise this kind of error.

Simply set the Client-Ip header in your nginx config:

As for the IP addresses in your log:

  • probably is NodeBalancer's internal IP
    (examples in Linode Library show subnet as usual; that should be static - or at least keep inside the subnet)
  • is the real remote address you want to know
  • is some other, unknown IP address - maybe another proxy added it earlier
    (does it appear on each request or is it sometimes empty?)

I'm sorry, I somehow came to believe (your) nginx would be running on the NodeBalancer.
Use the HttpRealIpModule like this:

# trust connections from internal addresses
real_ip_header X-Forwarded-For;

Or, at least make sure to pass the original X-Forwarded-For header to Rails instead of nginx's modified one, and unset the Client-Ip (whereever it may come from):

proxy_set_header X-Forwarded-For $http_x_forwarded_for;
proxy_set_header Client-Ip "";
share|improve this answer
With this technique, I loose the IP of the client. In my logs, all I get is the NodeBalancer's internal IP. I ended up removing the ip_spoofing_check in the rails app. Thanks – jlfenaux Mar 22 '13 at 10:00
Edited my answer some days ago... maybe the question still matters for you. – Lukas Mar 27 '13 at 23:32

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.