Supplemental Groups

Klaus Weidner klaus at atsec.com
Wed Feb 23 16:55:19 UTC 2005


On Tue, Feb 22, 2005 at 10:02:55AM -0800, Casey Schaufler wrote:
> --- Chris Wright <chrisw at osdl.org> wrote:
> > I don't know, I don't think it's explicitly required by CAPP (unless
> > you interpret subject identity to include suplemental group IDs).
> 
> If the suplemental group information is used to make access control
> decisions it needs to be in the audit record. Same with the
> capabilities.

Sorry to disagree once again, but auditing the reasons for access control
decisions isn't required by CAPP anywhere, see section 5.1.1.2 and table
1 for the exhaustive list of audit record requirements. It just needs to
record the success/failure result of the operation.

LSPP adds a requirement that the "sensitivity labels of subjects,
objects, or information involved" be included in each audit record
(that's the MAC information), this does not include any DAC information
such as group membership or ACLs.

I agree that having this type of information in the audit records would
be useful, but doing it right would need many changes to the OS and goes
beyond what CAPP/LSPP require.

Capabilities are different, this could be considered to be covered by the
"use of the rights of a role" requirement for FMT_SMR.1. The existing
Linux security targets had not made any fine-grained role distinctions
based on capabilities, they have used the simplified "administrator ==
UID 0 == all capabilities" model which makes it sufficient to record the
effective UID. If you have an ST that identifies distinct roles based on
individual capability bits, you would need to audit the current
capabilities of the process. Note that the audit requirement for CAPP and
LSPP is that "the role and the origin of the request" be included in the
record.

-Klaus




More information about the Linux-audit mailing list