oss-sec mailing list archives

Re: Problems in automatic crash analysis frameworks


From: Tavis Ormandy <taviso () google com>
Date: Tue, 5 May 2015 12:56:21 -0700

On Tue, May 5, 2015 at 12:37 PM, Florian Weimer <fweimer () redhat com> wrote:
On 05/05/2015 09:01 PM, Tavis Ormandy wrote:
On Tue, May 5, 2015 at 5:17 AM, Florian Weimer <fweimer () redhat com> wrote:
On 04/23/2015 09:10 PM, Florian Weimer wrote:
On 04/17/2015 09:16 PM, Florian Weimer wrote:
A quick update on the abrt situation.

Another update.  We now have a public tracking bug listing the issues:

  <https://bugzilla.redhat.com/show_bug.cgi?id=1214172>


There is a public build (against EPEL7) of the consolidated fixes,
available as a Copr repository:

  <http://copr.fedoraproject.org/coprs/jfilak/abrt-hardened/>

This also includes the consolidated fixes.

At this stage, we'd appreciate additional comments/reviews.

Thanks Florian, this looks great. I'm just looking at the new ccpp, Is
it intentional that os-release and so on are still copied from the
process root?

Thanks for taking a look.  Have you downloaded the Copr builds?

In 0092-ccpp-do-not-read-data-from-root-directories.patch for abrt, the
root directory argument for dd_create_basic_files is changed to NULL,
which means that the copying inside dd_create_basic_files will not
happen (although the code is still around).  As far as I can tell, this
patch is also applied in the build.  Is there some other copying
operation I'm missing?

No, you're correct, I was just reading the code and didn't see rootdir
was always NULL, oops.

I'll take a look at the other changes.

Tavis.


Current thread: