git@vger.kernel.org mailing list mirror (one of many)
 help / color / mirror / code / Atom feed
From: Jeff King <peff@peff.net>
To: Junio C Hamano <gitster@pobox.com>
Cc: git@vger.kernel.org, Daniel Stenberg <daniel@haxx.se>
Subject: Re: [PATCH] http: match headers case-insensitively when redacting
Date: Wed, 22 Sep 2021 16:09:00 -0400	[thread overview]
Message-ID: <YUuNXOb5blV7iN6P@coredump.intra.peff.net> (raw)
In-Reply-To: <xmqq8rzo770h.fsf@gitster.g>

On Wed, Sep 22, 2021 at 12:19:26PM -0700, Junio C Hamano wrote:

> Jeff King <peff@peff.net> writes:
> 
> > 	# Ensure that there is no "Basic" followed by a base64 string, but that
> > 	# the auth details are redacted
> > 	! grep "Authorization: Basic [0-9a-zA-Z+/]" trace &&
> > 	grep "Authorization: Basic <redacted>" trace
> >
> > gets confused. It sees the "<redacted>" one from the pre-upgrade
> > HTTP/1.1 request, but fails to see the unredacted HTTP/2 one, because it
> > does not match the lower-case "authorization".
> 
> Neither pattern of the above two will not match the HTTP/2 one, so
> the first one would report "there is no leakage of Auth with a
> caplital letter"; the second one may see only one pre-upgrade Auth
> with a capital letter, but as long as it does find one, it should be
> happy, no?
> 
> I am a bit puzzled how the test gets confused.

The first one matches nothing, because the HTTP/2 one which fails to
redact has a lower-case "A". The second one _does_ match, because we do
issue an HTTP/1.1 request in addition to the HTTP/2 one. We have to in
order to probe the server to say "this is HTTP/1.1, but by the way, we
support HTTP/2".

I am a little surprised that we get as far as sending auth info via
HTTP/1.1, since the initial probe that results in a 401 (causing us to
send the auth) could in theory let us know the server speaks HTTP/2. But
in practice it doesn't.  It looks like the server does not do the
upgrade for a 401 (perhaps that's true for any non-success code, I don't
know).

You can see it in action if you use the test changes I mentioned earlier
_without_ my patch applied (so neither the "grep -i" fix, nor the actual
code change). And then do:

  ./t5551-http-fetch-smart.sh --run=1-17 --debug
  egrep 'Send header:|Recv header:' trash*/trace

I get (with some extraneous headers omitted):

  => Send header: GET /auth/smart/repo.git/info/refs?service=git-upload-pack HTTP/1.1
  => Send header: Connection: Upgrade, HTTP2-Settings
  => Send header: Upgrade: h2c
  => Send header: HTTP2-Settings: AAMAAABkAAQCAAAAAAIAAAAA

  <= Recv header: HTTP/1.1 401 Unauthorized
  <= Recv header: Date: Wed, 22 Sep 2021 20:03:32 GMT
  <= Recv header: Server: Apache/2.4.49 (Debian)
  <= Recv header: WWW-Authenticate: Basic realm="git-auth"

  => Send header: GET /auth/smart/repo.git/info/refs?service=git-upload-pack HTTP/1.1
  => Send header: Authorization: Basic <redacted>
  => Send header: Connection: Upgrade, HTTP2-Settings
  => Send header: Upgrade: h2c
  => Send header: HTTP2-Settings: AAMAAABkAAQCAAAAAAIAAAAA

  <= Recv header: HTTP/1.1 101 Switching Protocols
  <= Recv header: Upgrade: h2c
  <= Recv header: Connection: Upgrade
  <= Recv header: HTTP/2 200
  <= Recv header: content-type: application/x-git-upload-pack-advertisement

  => Send header: POST /auth/smart/repo.git/git-upload-pack HTTP/2
  => Send header: authorization: Basic dXNlckBob3N0OnBhc3NAaG9zdA==
  => Send header: content-type: application/x-git-upload-pack-request

  <= Recv header: HTTP/2 200
  <= Recv header: content-type: application/x-git-upload-pack-result
  [...and so on...]

So you can see both the redacted and unredacted lines in that output.
I'm happy to include that in the commit message if it helps; I avoided
it earlier because it was already getting quite long. ;)

-Peff

  reply	other threads:[~2021-09-22 20:09 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-09-21 18:41 [PATCH] http: match headers case-insensitively when redacting Jeff King
2021-09-21 18:47 ` Jeff King
2021-09-21 20:14   ` Carlo Arenas
2021-09-21 20:40     ` Jeff King
2021-09-21 22:00   ` Daniel Stenberg
2021-09-22  2:32     ` Jeff King
2021-09-21 19:06 ` Eric Sunshine
2021-09-21 19:14   ` Jeff King
2021-09-22 19:10     ` Junio C Hamano
2021-09-21 21:20 ` Taylor Blau
2021-09-22  2:30   ` Jeff King
2021-09-22  2:32 ` Bagas Sanjaya
2021-09-22 20:11   ` Jeff King
2021-09-23  1:22     ` Ævar Arnfjörð Bjarmason
2021-09-23 21:56       ` Jeff King
2021-09-22 19:19 ` Junio C Hamano
2021-09-22 20:09   ` Jeff King [this message]
2021-09-22 20:51     ` Junio C Hamano
2021-09-22 21:18       ` Jeff King
2021-09-22 21:42         ` Junio C Hamano
2021-09-22 22:11           ` [PATCH v2] " Jeff King
2021-09-22 22:14             ` Jeff King

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

  List information: http://vger.kernel.org/majordomo-info.html

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=YUuNXOb5blV7iN6P@coredump.intra.peff.net \
    --to=peff@peff.net \
    --cc=daniel@haxx.se \
    --cc=git@vger.kernel.org \
    --cc=gitster@pobox.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
Code repositories for project(s) associated with this public inbox

	https://80x24.org/mirrors/git.git

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).