Skip to content

NGINX Errors

400 Bad Request

Nginx has strict rules around the headers allowed on requests. If you encounter a 400 Bad Request response in your environment when accessing /boards it is likely caused by incorrect headers set in the upsteam proxy(s) before Boards.

To debug the cause, please views the logs for the webfront pods (as of build 20210924). You will see logs like:

  setting core: https://devconn7.internal.isw.net.au/api-boards
  setting buildId: 198
  setting product info url: https://huddo.com/boards
  setting force polling: true
  setting html base: /boards
  2021/09/24 01:10:49 [notice] 1#1: using the "epoll" event method
  2021/09/24 01:10:49 [notice] 1#1: nginx/1.21.3
  2021/09/24 01:10:49 [notice] 1#1: built by gcc 10.3.1 20210424 (Alpine 10.3.1_git20210424)
  2021/09/24 01:10:49 [notice] 1#1: OS: Linux 3.10.0-1160.15.2.el7.x86_64
  2021/09/24 01:10:49 [notice] 1#1: getrlimit(RLIMIT_NOFILE): 1048576:1048576
  2021/09/24 01:10:49 [notice] 1#1: start worker processes
  2021/09/24 01:10:49 [notice] 1#1: start worker process 20
  2021/09/24 01:10:49 [notice] 1#1: start worker process 21
  2021/09/24 01:10:49 [notice] 1#1: start worker process 22
  2021/09/24 01:10:49 [notice] 1#1: start worker process 23
  2021/09/24 01:15:37 [info] 20#20: *1 client 10.244.115.83 closed keepalive connection
  2021/09/24 01:15:38 [info] 31#31: *118 client sent invalid host header while reading client request headers, client: 172.20.0.1, server: boards.company.com, request: "GET / HTTP/1.1", host: "boards.company.com, boards.company.com"

In this example, the client sent invalid host header while reading client request headers. You can see the host is included twice. This can occur if the host is set twice, or in some instances when the X-Forwarded-Host is also set.

Please read this error carefully and make sure your environment complies with the latest NGINX specification.