All of lore.kernel.org
 help / color / mirror / Atom feed
From: Marcel Holtmann <marcel@holtmann.org>
To: Steve Grubb <sgrubb@redhat.com>
Cc: linux-bluetooth@vger.kernel.org
Subject: Re: [PATCH] Drop Posix Capabilities
Date: Tue, 29 Sep 2009 11:03:46 -0700	[thread overview]
Message-ID: <1254247426.2659.90.camel@localhost.localdomain> (raw)
In-Reply-To: <200909290900.51115.sgrubb@redhat.com>

Hi Steve,

> > Thanks foradding this. Another question that comes to my mind. Where do you
> > have the GIT tree for libcap-ng so we can track the development?
> 
> I've never needed to use git - so no git tree. The project has a home page 
> here:  http://people.redhat.com/sgrubb/libcap-ng/
> and I announce package updates at freshmeat.net. The libcap-ng package is 
> stable and I would not have made a release yesterday if it weren't for needing 
> to a pc file. I don't forsee much development in libcap-ng unless there are 
> updates in the kernel that I need to take into account. IOW, all planned 
> features are complete and I'm not tracking any bugs.

from a security point of it is nice if you have a source code repository
with version tracking. Only thinking about your security team if
something might happen :)
 
> > > > I like to have capability dropping in bluetoothd, but I do wanna do it
> > > > with a proper upstream project.
> > > 
> > > one other thing I thought I would point out. The patch I sent can make it
> > > easy  to run the bluetooth daemon as non-root user. If we switch this
> > > line: 
> > > capng_apply(CAPNG_SELECT_BOTH);
> > > 
> > > to
> > > 
> > > capng_change_id(uid, gid, CAPNG_DROP_SUPP_GRP | CAPNG_CLEAR_BOUNDING);
> > > 
> > > then the job is easier. Of course you would likely need to fixup file 
> > > permissions in places, but in theory a non-root bluetooth daemon is
> > > possible  with a 1 line change in the patch. You would probably want to
> > > add error handling and a way to specify the uid/gid, too.
> > 
> > I am not really sold on the non-root daemon idea and there might be
> > hidden problems where this will not work out. However I don't mind
> > trying at some point, but there are other things to sort out first. We
> > should postpone this for the 5.x series.
> 
> Sure, I just wanted to point out that its a 1 line change in code if you ever 
> wanted to do this.
> 
> 
> > Please re-send the original patch using pkg-config so I can go ahead an
> > apply it. Even if Rawhide is not carrying the updated libcap-ng package.
> 
> OK, as soon as I figure out pkg-config. M4 is easier. :)

But the number of people who can read M4 getting less and less. I can
read it, but then nobody else inside the team will have a real clue.
Just send me a patch without the pkg-config magic and I can add that for
you after I applied the patch.

Regards

Marcel



  reply	other threads:[~2009-09-29 18:03 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-09-25 20:47 [PATCH] Drop Posix Capabilities Steve Grubb
2009-09-25 21:35 ` Marcel Holtmann
2009-09-26 14:29   ` Steve Grubb
2009-09-27 20:31     ` Marcel Holtmann
2009-09-28 21:31       ` Steve Grubb
2009-09-28 23:40         ` Marcel Holtmann
2009-09-29 13:00           ` Steve Grubb
2009-09-29 18:03             ` Marcel Holtmann [this message]
2009-10-02  9:46 ` Marcel Holtmann

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=1254247426.2659.90.camel@localhost.localdomain \
    --to=marcel@holtmann.org \
    --cc=linux-bluetooth@vger.kernel.org \
    --cc=sgrubb@redhat.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.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.