On 2016-08-02 09:58, Steve Grubb wrote:
On Tuesday, August 2, 2016 9:25:44 AM EDT Steve Grubb wrote:
> On Tuesday, August 2, 2016 8:56:35 AM EDT Richard Guy Briggs wrote:
> > On 2016-08-02 08:16, Steve Grubb wrote:
> > > On Tuesday, August 2, 2016 5:38:56 AM EDT Richard Guy Briggs wrote:
> > > > Add support for sessionid, sessionid_set (first two patches) and
> > > > loginuid_set (and auid_set) (third patch) in user filters. The
first
> > > >
> > > > two are directly related to issue "ghak4":
> > > >
https://github.com/linux-audit/audit-kernel/issues/4
> > > >
https://github.com/linux-audit/audit-kernel/wiki/RFE-Session-ID->
>
> > > > > User-Filter
> > > >
> > > > The third is to support a kernel change from 3.10 and 3.19 to avoid
> > > > using in-band values to indicate the loginuid is unset.
> > >
> > > Have the above three patches been tested on old kernels?
> >
> > Not yet. How do you usually add new features to userspace to guard
> > against missing features from old kernels? Time to add a bit to the
> > kenrel audit status feature field?
>
> Yes. Otherwise you get EINVAL which doesn't let you explain what exactly is
> wrong with the rule.
Before you get too far...I just looked at the support being added in the first
three patches. There is no code changing auditctl. Is there something missing?
I am pretty sure I looked through auditctl and in the three cases things
were simple enough that the fallthrough case was sufficient. Now that I
think about it, to avoid the ambiguity of -EINVAL, restricting the first
to a positive value and the 2nd and 3rd to true or false would remove
that ambiguity.
-Steve
- RGB
--
Richard Guy Briggs <rgb(a)redhat.com>
Kernel Security Engineering, Base Operating Systems, Red Hat
Remote, Ottawa, Canada
Voice: +1.647.777.2635, Internal: (81) 32635