[ruby-core:63078] [ruby-trunk - Bug #9927] [Rejected] webrick does not unset content-length when responding to HEAD requests

Issue #9927 has been updated by Eric Hodel.

Status changed from Open to Rejected

This appears to be a bug in curl.

RFC 7230 says:

    3.3.  Message Body
=20=20=20=20
       The message body (if any) of an HTTP message is used to carry the
       payload body of that request or response.  The message body is
       identical to the payload body unless a transfer coding has been
       applied, as described in Section 3.3.1.
=20=20=20=20
         message-body =3D *OCTET
=20=20=20=20
       The rules for when a message body is allowed in a message differ for
       requests and responses.
=20=20=20=20
       The presence of a message body in a request is signaled by a
       Content-Length or Transfer-Encoding header field.  Request message
       framing is independent of method semantics, even if the method does
       not define any use for a message body.
=20=20=20=20
       The presence of a message body in a response depends on both the
       request method to which it is responding and the response status code
       (Section 3.1.2).  Responses to the HEAD request method (Section 4.3.2
       of [RFC7231]) never include a message body because the associated
       response header fields (e.g., Transfer-Encoding, Content-Length,
       etc.), if present, indicate only what their values would have been if
       the request method had been GET (Section 4.3.1 of [RFC7231]). 2xx
       (Successful) responses to a CONNECT request method (Section 4.3.6 of
       [RFC7231]) switch to tunnel mode instead of having a message body.
       All 1xx (Informational), 204 (No Content), and 304 (Not Modified)
       responses do not include a message body.  All other responses do
       include a message body, although the body might be of zero length.

RFC 2616 says the same using different words in 4.4:

       1.Any response message which "MUST NOT" include a message-body (such
         as the 1xx, 204, and 304 responses and any response to a HEAD
         request) is always terminated by the first empty line after the
         header fields, regardless of the entity-header fields present in
         the message.


----------------------------------------
Bug #9927: webrick does not unset content-length when responding to HEAD re=
quests
https://bugs.ruby-lang.org/issues/9927#change-47153

* Author: Adrien Thebo
* Status: Rejected
* Priority: Normal
* Assignee:=20
* Category:=20
* Target version:=20
* ruby -v: ruby 1.9.3p484 (2013-11-22 revision 43786) [x86_64-linux]
* Backport: 2.0.0: UNKNOWN, 2.1: UNKNOWN
----------------------------------------
When Webrick responds to HEAD requests it omits the body (per RFC2616 -- 4.=
4 Message Length). However when setting up the response headers the content=
-length field is set to the length of the body, which means that the result=
ing response will have a content length that doesn't match the actual respo=
nse. This means that some HTTP clients may hang when reading the response.

This is reproducible with the following:

```ruby
require 'webrick'

server =3D WEBrick::HTTPServer.new :Port =3D> 8080, :BindAddress =3D> '127.=
0.0.1'
server.mount_proc("/") do |req, res|
  res.body =3D "This will be ignored!\r\n"
end
trap('INT') do
  server.shutdown
end
server.start
```

Running this with curl results in the following:

~~~
=E2=94=94> ruby webrick-head.rb=20=20=20=20=20=20=20=20=20=20=20=20=20=20=
=20=20=20
[2014-06-10 12:07:28] INFO  WEBrick 1.3.1
[2014-06-10 12:07:28] INFO  ruby 1.9.3 (2013-11-22) [x86_64-linux]
[2014-06-10 12:07:28] INFO  WEBrick::HTTPServer#start: pid=3D24798 port=3D8=
080
localhost - - [10/Jun/2014:12:07:30 PDT] "HEAD / HTTP/1.1" 200 0
- -> /
^C[2014-06-10 12:07:36] INFO  going to shutdown ...
[2014-06-10 12:07:36] INFO  WEBrick::HTTPServer#start done.
~~~

~~~
=E2=94=94> curl -v -X HEAD http://localhost:8080
* Rebuilt URL to: http://localhost:8080/
* Hostname was NOT found in DNS cache
*   Trying 127.0.0.1...
  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Cur=
rent
                                 Dload  Upload   Total   Spent    Left  Spe=
ed

  0     0    0     0    0     0      0      0 --:--:-- --:--:-- --:--:--   =
  0* Connected to localhost (127.0.0.1) port 8080 (#0)
> HEAD / HTTP/1.1
> User-Agent: curl/7.37.0
> Host: localhost:8080
> Accept: */*
>=20
< HTTP/1.1 200 OK=20
* Server WEBrick/1.3.1 (Ruby/1.9.3/2013-11-22) is not blacklisted
< Server: WEBrick/1.3.1 (Ruby/1.9.3/2013-11-22)
< Date: Tue, 10 Jun 2014 19:07:30 GMT
< Content-Length: 23
< Connection: Keep-Alive
<=20

  0    23    0     0    0     0      0      0 --:--:--  0:00:01 --:--:--   =
  0
  0    23    0     0    0     0      0      0 --:--:--  0:00:02 --:--:--   =
  0
  0    23    0     0    0     0      0      0 --:--:--  0:00:03 --:--:--   =
  0
  0    23    0     0    0     0      0      0 --:--:--  0:00:04 --:--:--   =
  0
  0    23    0     0    0     0      0      0 --:--:--  0:00:05 --:--:--   =
  0
[Ctrl-C sent to server]
  0    23    0     0    0     0      0      0 --:--:--  0:00:06 --:--:--   =
  0{ [data not shown]
* transfer closed with 23 bytes remaining to read

  0    23    0     0    0     0      0      0 --:--:--  0:00:06 --:--:--   =
  0
* Closing connection 0
curl: (18) transfer closed with 23 bytes remaining to read
~~~

This is reasonably straightforward to fix; when the headers are being creat=
ed and the code is checking to see if the body should be ignored for HTTP 2=
04 and 304, we can check to see if we're responding to a HEAD request and b=
ehave accordingly. I've attached patches to this effect

---Files--------------------------------
0001-lib-webrick-httpresponse.rb-unset-content-length-hea.patch (1.54 KB)


--=20
https://bugs.ruby-lang.org/