Transcript DAC - Faculty of Computer Science & Engineering

DISCRETIONARY ACCESS CONTROLS
Truong Quynh Chi
Faculty of Computer Science & Engineering
HCMC University of Technology
[email protected]
OUTLINE
Introduction to Discretionary Access Control
 Proposed Models for DAC
 SQL for Data Control
 DAC & Information Flow Controls

2
INTRODUCTION TO DAC

Discretionary Access Control (DAC):
User can protect what they own.
 The owner is given all privileges on their own data.
 The owner can define the type of access
(read/write/execute/…) and grant access to others.
 The typical method of enforcing DAC in a database
system is based on the granting and revoking
privileges

3
INTRODUCTION TO DAC

Types of Discretionary Privileges:
The account/system level: The administrator
specifies the particular privileges that each account
holds independently of the objects in the database
system.
 The object level: The administrator can control the
privilege to access each individual object in the
database system.

4
OUTLINE
Introduction to Discretionary Access Control
 Proposed Models for DAC
 SQL for Data Control
 DAC & Information Flow Controls

5
PROPOSED MODELS FOR DAC
Access matrix model
 Take-Grant model

6
ACCESS MATRIX MODEL



Authorization state:
Q=(S,O,A)
For DBs, A[s,o] also
includes conditions
that must be satisfied
in order for s to
exercise the access
modes
Possible conditions:
data-dependent
(sal<1000), timedependent (8:00am5:00pm), contextdependent (“namesalary” pair is
prohibited), historydependent, …
O1
S1
…
Oi
…
Om
A[s1,o1]
A[s1,oi]
A[s1,om]
A[si,o1]
A[si,oi]
A[si,om]
A[sn,o1]
A[sn,oi]
A[sn,om]
…
Si
…
Sn
ACCESS MATRIX MODEL
Role 1
Role 2
Asset 1
Asset 2
file
device
read,
write,
execute,
own
execute
read
write
read
read,
write,
execute,
own
ACCESS MATRIX MODEL

Model implementation:



S {(O,A)}: capability list
O{(S,A)}: ACL (access control list)
Each entry in the list specifies a subject and operation(s): for
example, the entry (Alice, delete) on the ACL for file X gives
Alice permission to delete file X
Advantages & disadvantages of the two above & the
model? [2]
Capability list: compute a set of subjects granted access on a
given object  all lists must be gone through
 ACL: find all objects a subject can access

TAKE-GRANT MODEL

Authorization state: G=(S,O,E)
V=S U O is the set of vertexes, S ∩ O = Ф
 E is the set of labelled arcs


s
take(d,s,x,y): the subject s takes the right d on
the object/subject y from the object/subject x
t
t
s
x
d
y
take(d,s,x,y)
x
d
d
y
10
TAKE-GRANT MODEL
• grant(d,s,x,y): the subject s grants the right d on the
object/subject y to the object/subject x
g
g
s
s
x
x
grant(d,s,x,y)
d
y
d
d
y
11
TAKE-GRANT MODEL
Access modes: read, write, take, grant
 Read, write: inert rights
 Take, grant: transport rights
 A subject S may lose control on the subsequent
transfers of the rights it granted to an object O
 This model is classifiable as an access matrix
model
 Disadvantages?

12
TAKE-GRANT MODEL

Disadvantages:
Non-selectivity of administrative rights: all
authorizations of S owning a ‘GRANT’ authorization
can be transferred, and all authorizations of O/S on
which a ‘TAKE’ right is held can be taken
 No control on propagation of authorizations
 Non locality: S owning the grant privilege on O can
give any of its privileges to O, thus augmenting the
domain of O (the set of authorizations associated to
O)
 Reversibility of the privileges transport flow

13
OTHER MODELS
Acten (Action-Entity) model
 Wood et al. model
 See [2] + Internet

14
OUTLINE
Introduction to Discretionary Access Control
 Proposed Models for DAC
 SQL for Data Control
 DAC & Information Flow Controls

15
SQL FOR DATA CONTROL


Commands:
•
GRANT
•
REVOKE
Based on three central objects:
•
Users
•
Database objects
•
Privileges: select, modify (insert, update,
delete), reference
16
SQL FOR DATA CONTROL
GRANT: pass privileges on their own database
objects to other users
GRANT <privilege list>
ON
<database objects>
TO
<user list>
 REVOKE: take back (cancel) privileges on
their own database objects from other users
REVOKE <privilege list>
ON
<database objects>
FROM
<user list>

17
SQL FOR DATA CONTROL

Propagation of Privileges using the GRANT
OPTION
Whenever the owner A of a relation R grants a
privilege on R to another account B, privilege can be
given to B with or without the GRANT OPTION.
 If the GRANT OPTION is given, this means that B
can also grant that privilege on R to other accounts.

18
SQL FOR DATA CONTROL

DAC with views (virtual relations)
19
If the owner A of a relation R wants another
account B to be able to retrieve only some fields of
R, then A can create a view V of R that includes
only those attributes and then grant SELECT on V
to B.
 The same applies to limiting B to retrieving only
certain tuples of R; a view V’ can be created by
defining the view by means of a query that selects
only those tuples from R that A wants to allow B to
access.

AN EXAMPLE

Suppose that the DBA creates four accounts


A1, A2, A3, A4
and wants only A1 to be able to create base
relations. Then the DBA must issue the following
GRANT command in SQL
GRANT CREATETAB TO A1;

In SQL2 the same effect can be accomplished by
having the DBA issue a CREATE SCHEMA
command as follows:
CREATE SCHEMA EXAMPLE AUTHORIZATION
A1;
20
AN EXAMPLE(2)
User account A1 can create tables under the
schema called EXAMPLE.
 Suppose that A1 creates the two tables
EMPLOYEE and DEPARTMENT



A1 is then owner of these two relations and hence
all the relation privileges on each of them.
Suppose that A1 wants to grant A2 the privilege
to insert and delete tuples in both of these
relations, but A1 does not want A2 to be able to
propagate these privileges to additional accounts:
GRANT INSERT, DELETE ON
EMPLOYEE, DEPARTMENT TO A2;
21
AN EXAMPLE(3)
22
AN EXAMPLE(4)
Suppose that A1 wants to allow A3 to retrieve
information from either of the two tables and also to
be able to propagate the SELECT privilege to other
accounts.
 A1 can issue the command:

GRANT SELECT ON EMPLOYEE, DEPARTMENT
TO A3 WITH GRANT OPTION;

A3 can grant the SELECT privilege on the
EMPLOYEE relation to A4 by issuing:
GRANT SELECT ON EMPLOYEE TO A4;

Notice that A4 can’t propagate the SELECT privilege
because GRANT OPTION was not given to A4
23
AN EXAMPLE(5)

Suppose that A1 decides to revoke the SELECT
privilege on the EMPLOYEE relation from A3;
A1 can issue:
REVOKE SELECT ON EMPLOYEE FROM A3;

The DBMS must now automatically revoke the
SELECT privilege on EMPLOYEE from A4, too,
because A3 granted that privilege to A4 and A3
does not have the privilege any more.
24
AN EXAMPLE(6)


Suppose that A1 wants to give back to A3 a limited
capability to SELECT from the EMPLOYEE relation and
wants to allow A3 to be able to propagate the privilege.
 The limitation is to retrieve only the NAME, BDATE,
and ADDRESS attributes and only for the tuples with
DNO=5.
A1 then create the view:
CREATE VIEW A3EMPLOYEE AS
SELECT NAME, BDATE, ADDRESS
FROM EMPLOYEE
WHERE DNO = 5;

After the view is created, A1 can grant SELECT on the
view A3EMPLOYEE to A3 as follows:
GRANT SELECT ON A3EMPLOYEE TO A3
WITH GRANT OPTION;
25
AN EXAMPLE(7)
Finally, suppose that A1 wants to allow A4 to
update only the SALARY attribute of
EMPLOYEE;
 A1 can issue:

GRANT UPDATE ON EMPLOYEE (SALARY)
TO A4;
The UPDATE or INSERT privilege can specify
particular attributes that may be updated or
inserted in a relation.
 Other privileges (SELECT, DELETE) are not
attribute specific.

26
OUTLINE
Introduction to Discretionary Access Control
 Proposed Models for DAC
 SQL for Data Control
 DAC & Information Flow Controls

27
DAC & INFORMATION FLOW CONTROLS

Inherent weakness of DAC: Unrestricted DAC allows
information from an object which can be read by a
subject to be written to any other object


Bob is denied access to file A, so he asks cohort Alice to
copy A to B that he can access
Suppose our users are trusted not to do this
deliberately. It is still possible for Trojan Horses to
copy information from one object to another.
28
TROJAN HORSE EXAMPLE
29
TROJAN HORSE EXAMPLE
30
TROJAN HORSE EXAMPLE
31
32