I look after www.lesscss.org. The source is here https://github.com/cloudhead/lesscss.org.
This is a thin web application and runs on heroku. Accessing the website in a browser is fine.
We have had a bug that curl -I lesscss.org gives a 400 request - https://github.com/cloudhead/less.js/issues/1232
and it does
$ curl -I lesscss.org
HTTP/1.1 400 Bad Request
Server: nginx
Date: Tue, 19 Mar 2013 01:15:50 GMT
Connection: close
I've done alot of searching and haven't found a reason why or how to rectify the above.. shouldn't it be a 302 redirect or a 200 ok?
[Edit]
with verbose option
$ curl -Iv http://www.lesscss.org
* About to connect() to www.lesscss.org port 80 (#0)
* Trying 107.21.106.77...
* 0x8001f140 is at send pipe head!
* STATE: CONNECT => WAITCONNECT handle 0x80057408; line 1032 (connection #0)
* Connected to www.lesscss.org (107.21.106.77) port 80 (#0)
* STATE: WAITCONNECT => DO handle 0x80057408; line 1151 (connection #0)
> HEAD / HTTP/1.1
> User-Agent: curl/7.29.0
> Host: www.lesscss.org
> Accept: */*
>
* STATE: DO => DO_DONE handle 0x80057408; line 1236 (connection #0)
* STATE: DO_DONE => WAITPERFORM handle 0x80057408; line 1352 (connection #0)
* STATE: WAITPERFORM => PERFORM handle 0x80057408; line 1363 (connection #0)
* HTTP 1.1 or later with persistent connection, pipelining supported
< HTTP/1.1 400 Bad Request
HTTP/1.1 400 Bad Request
< Server: nginx
Server: nginx
< Date: Wed, 20 Mar 2013 09:34:37 GMT
Date: Wed, 20 Mar 2013 09:34:37 GMT
< Connection: keep-alive
Connection: keep-alive
<
* STATE: PERFORM => DONE handle 0x80057408; line 1533 (connection #0)
* Connection #0 to host www.lesscss.org left intact
* Expire cleared
The toto engine that you're using has this line in it:
return [400, {}, []] unless @request.get?
So anything that is not a get will result in a 400. Toto could probably afford to allow head requests through (and combine with Rack::Head to drop request bodies on the floor if needed)