git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Paul Sokolovsky <paul.sokolovsky@linaro.org>
To: Jeff King <peff@peff.net>
Cc: Junio C Hamano <gitster@pobox.com>, git@vger.kernel.org
Subject: Re: git 2.2.x: Unexpected, overstrict file permissions after "git update-server-info"
Date: Tue, 6 Jan 2015 14:12:11 +0200	[thread overview]
Message-ID: <20150106141211.2ad83df4@x230> (raw)
In-Reply-To: <20150106034702.GA11503@peff.net>

Hello,

On Mon, 5 Jan 2015 22:47:02 -0500
Jeff King <peff@peff.net> wrote:

> On Mon, Jan 05, 2015 at 09:07:24PM +0200, Paul Sokolovsky wrote:
> 
> > So, after the upgrade, users started to report that accessing
> > info/refs file of a repo, as required for HTTP dump protocol, leads
> > to 403 Forbidden HTTP error. We traced that to 0600 filesystem
> > permissions for such files (for objects/info/packs too) (owner is
> > gerrit user, to remind). After resetting permissions to 0644, they
> > get back to 0600 after some time (we have a cronjob in addition to
> > a hook to run "git update-server-info"). umask is permissive when
> > running cronjob (0002).
> > 
> > I traced the issue to:
> > https://github.com/git/git/commit/d38379ece9216735ecc0ffd76c4c4e3da217daec
> 
> Yeah, I didn't consider the mode impact of using mkstemp. That is
> definitely a regression that should be fixed. Though of course if you
> really do want 0644, you should set your umask to 0022. :)

Well, group permissions are ok - we just need it to be world-readable,
and that's not random, but complies with hosting requirements - our
repos are public otherwise.

> > It says: "Let's instead switch to using a unique tempfile via
> > mkstemp." Reading man mkstemp: "The  file  is  created  with
> > permissions 0600". So, that's it. The patch above contains call to
> > adjust_shared_perm(), but apparently it doesn't promote restrictive
> > msktemp permissions to something more accessible.
> 
> If you haven't set core.sharedrepository, then adjust_shared_perm is a
> noop. But you shouldn't have to do that. Git should just respect your
> umask in this case.

My reference to adjust_shared_perm() was because I initially wanted to
write "apparently, it makes sense to do chmod after mkstemp()", but I
spotted that there's adjust_shared_perm() already, which does some
shuffling of permissions.

> > Hope this issue can be addressed.
> 
> Patches to follow. Thanks for the report.
> 
>   [1/2]: t1301: set umask in reflog sharedrepository=group test
>   [2/2]: update-server-info: create info/* with mode 0666

Thanks much for the prompt reply and patches!

> 
> -Peff



-- 
Best Regards,
Paul

Linaro.org | Open source software for ARM SoCs
Follow Linaro: http://www.facebook.com/pages/Linaro
http://twitter.com/#!/linaroorg - http://www.linaro.org/linaro-blog

      parent reply	other threads:[~2015-01-06 12:12 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-01-05 19:07 git 2.2.x: Unexpected, overstrict file permissions after "git update-server-info" Paul Sokolovsky
2015-01-05 22:23 ` Torsten Bögershausen
2015-01-06  3:47 ` Jeff King
2015-01-06  3:49   ` [PATCH 1/2] t1301: set umask in reflog sharedrepository=group test Jeff King
2015-01-06  3:50   ` [PATCH 2/2] update-server-info: create info/* with mode 0666 Jeff King
2015-01-06 18:47     ` Junio C Hamano
2015-01-06 19:39       ` Jeff King
2015-01-06 21:43         ` Junio C Hamano
2015-01-06 21:47           ` Jeff King
2015-01-06 10:08   ` git 2.2.x: Unexpected, overstrict file permissions after "git update-server-info" Junio C Hamano
2015-01-06 12:43     ` Paul Sokolovsky
2015-01-06 18:44     ` Junio C Hamano
2015-01-06 19:37     ` Jeff King
2015-01-06 12:12   ` Paul Sokolovsky [this message]

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

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

  git send-email \
    --in-reply-to=20150106141211.2ad83df4@x230 \
    --to=paul.sokolovsky@linaro.org \
    --cc=git@vger.kernel.org \
    --cc=gitster@pobox.com \
    --cc=peff@peff.net \
    /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.
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 NNTP newsgroup(s).