Return a content-length on HTTP204 to prevent client failures

It turns out that eventlet has been injecting a
``Transfer-Encoding`` header as of recent into WSGI application
response headers. The result of this ultimately depends on how
the HTTP client which is passing the request to the server is
written to handle data.

Apache, for example, will return that an invalid response was
received. In part because it sees the request end, with an HTTP
204 response code, but also an encoding indicating there is
a multipart body encoding inbound. Which is confusing.

Other C based HTTP clients can have any number of reactions up to
and including disconnecting sessions. Curl, depending on the
headers present either returns success but notes body weirdness
or actually returns return code 18.

Python-Requests kind of has it a little worse, and we see this
with clients. With it, it tries to prepare a respones content
body based upon the presence of the header indicating there is
a body. But it blows up thinking there is more data to read on
the socket when there is not more data to read.

Regardless, all of this is an RFC7230 violation.

Neither Content-Length nor Transfer-Encoding should be on an HTTP
204 response. However, Content-Length is the lesser evil, and we
have a similar endpoing in Ironic which *does* explicitly get
returned with a zero length content-length, and does not
demonstrate such issues.

As such, in the interest of the lesser evils until Eventlet's evil
ways of header injection are remedied, we're explicitly going to
force a Content-Length header to be sent indicating a zero length
response.

For more information, please see: https://github.com/eventlet/eventlet/issues/746

Change-Id: I014cc65c79222f4d4d7c2b6ff11a76e56659340c
(cherry picked from commit 55e47c630b)
This commit is contained in:
Julia Kreger 2022-01-13 11:55:42 -08:00
parent 06c65757c7
commit 8f9707699f
3 changed files with 39 additions and 0 deletions

View File

@ -152,7 +152,21 @@ def _generate_empty_response(code):
# be included for user friendly response bodies.
if code == 204:
response = flask.make_response('', code)
# NOTE(TheJulia): Explicitly set a content length to zero.
# as this is the *lesser* of all evils until
# https://github.com/eventlet/eventlet/issues/746
# is resolved.
response.headers['Content-Length'] = 0
response.mimetype = 'text/plain'
def _return_headers(cls):
# Dynamically re-maps over get_wsgi_headers() method in
# which werkzueg strips the content-length and sets us up
# for eventlet forcing in a Transfer-Encoding: chunked
# header.
return response.headers
response.get_wsgi_headers = _return_headers
else:
# Send an empty dictionary to set a mimetype, and ultimately
# with this being a rest API we can, at some point, choose to

View File

@ -611,6 +611,12 @@ class TestApiRules(BaseAPITest):
delete_mock.assert_called_once_with(self.uuid)
self.assertEqual('text/plain; charset=utf-8',
res.headers['content-type'])
# NOTE(TheJulia): This should be able to be removed once
# https://github.com/eventlet/eventlet/issues/746
# is resolved and available. This ensures we don't
# send out a content length header out which is the
# less evil of the RFC7230 broken-ness.
self.assertEqual('0', res.headers['Content-Length'])
class TestApiMisc(BaseAPITest):

View File

@ -0,0 +1,19 @@
---
issues:
- |
The response headers for empty body HTTP 204 replies, at present, violate
RFC7230. This was not intentional, but underlying libraries also
make inappropriate changes to the headers, which can cause clients to
experience odd failures. This is anticipated to be corrected once an
underlying issue in
`eventlet <https://github.com/eventlet/eventlet/issues/746>`_ is resolved.
fixes:
- |
Fixes HTTP responses so the Eventlet library, which is used to support
the operation of the WSGI application, does not incorrectly inject
a ``Transfer-Encoding`` header into the HTTP response, even on HTTP 204
replies, which is a violation of RFC7230. This header ultimately can
cause varying client reactions which are not expected and can raise
exceptions. For now, this has been remedied via an explicit return of
a ``Content-Length`` header, which is also an RFC7230 violation, but
it appears to be the lesser of known evils at this time.