Data Security: – Database security is the unauthorized modification and disclosure of the database [607765]

Student: [anonimizat]: – Database security is the unauthorized modification and disclosure of the database
concept. Security is the first place in computer -related problems, and it hits it.

A database security system must determine who has access to the data, what types of data are
required, what type of access is required, and which one is authorized for that type of data. The
intent is to ensure that only authorized users have access to par ticular types of data, and if a user
is not authorized to use particular types of data then there should be no possibility for the user to
modify this data.

Database protection consists of a set of human measures and facilities provided by the database
management system (DMS) to ensure the integrity of data, defined as the correctness of data input
and manipulation, and data security, aimed at prohibiting access to data for people who do not
competent in their use. This is particularly important in the co ntext of expanding the use of large
number of configurations and a large amount of data to process.

1. Secure techniques
The task of ensuring database security is shared between the DMS and the operating system. They
can assure each and every one of the security functions or may be complementary to their
achievement. The tasks to ensure the security of the databases are exemplified in the following
table (table 1)

Table 1. Powers to sec ure database security
Function Task
Identification Operating System or in some cases the DMS
Login
Authorization The DMS (security application modules)
Access control The DMS (Transaction Management Model)
Integrity
Consistency
Audit Operating System or in some cases the DMS

Control over attacks
Previously, we saw that sensory data resulted from unsolicited data through various database
queries.
In order to prevent this, the following methods can be applied:
➢ suppressing application s with sensitive results;
➢ approximation of results;

Student: [anonimizat]
➢ limiting the results of a request that discloses sensitive data;
➢ Combining results.

Requests for access to database elements that result in the display of sensitive results are rejected
without any respo nse. Sensitive data will not be displayed. The result of such a query will be
correct but will not be displayed to the user.
In case of such a request, the system will be able to display close to actual results. The accuracy
of the result in a query that c an reveal sensitive data must be small.
Limiting the result of an application that reveals sensitive data can be done if it is 1 (one).

Table 2 . Table attacked
Name Sex Stud Sal Enfo Comp
Popescu M Valentin
Ionescu A Stelian
Grigore A Marcela
Gergescu P Ion
Simion I Janina
Tanase A Loredana
Alexe A Virgil
Gherase I Mihaela
Constantin I Iulia
Ilie G Ioana
Alexandra A Silviu M
M
F
M
F
F
M
F
F
F
M S
P
L
P
S
S
S
P
S
L
S 6000000
5500000
3500000
3900000
5100000
6000000
5500000
6400000
6100000
6400000
5400000 1
0
0
3
1
0
3
2
1
2
0 C
F
P
F
C
P
P
C
P
F
F

If, in the previous example in Table 2 , we will apply function -like attacks in the form shown in
Table 3 and centralize the results as a table (Table 4), I will see that we have dominant results that
must be suppressed.

Table 3. Indirect Attack Using COUNT Conditioned
COUNT FOR SEX = ”M” AND COMP = “ C”
COUNT FOR SEX = ”M” AND COMP = “F”
COUNT FOR SEX = ”M” AND COMP = “P”
COUNT FOR SEX = ”M”
COUNT FOR SEX = ”F” AND COMP = “C”
COUNT FOR SEX = ”F” AND COMP = “F”
COUNT FOR SEX = ”F” AND COMP = “P”
COUNT FOR SEX = ”F”
COUNT FOR COMP = “C”
COUNT FOR COMP = “F”
COUNT FOR COMP = “P”
COUNT 1 records
3 records
1 records
5 records
2 records
1 records
3 records
6 records
4 records
4 records
11 records

Student: Beniamin Scrob
Emanuel University of Oradea

Table 4. Centralization of results
C
(commercial) F
(financial) P
(production) Total
M 1 3 1 5
F 2 1 3 6
Total 3 4 4 11

In this case I have to suppress the applications that result in dominant values (1) or not allow the
displaying of the results even if the request is executed in such a way that the results are displayed
in the following form (table 5):

Table 5. Output Displaying
C
(commercial) F
(financial) P
(production) Total
M – 3 – 5
F 2 – 3 6
Total 4 5 5 11

It is noticed that the situation is not resolved, having to change the amounts on lines and columns
to create confusion and can not extract sensitive data.
Combining the results is done by displaying them in a range of values that will not allow accurate
data to be extracted. By making a count on the type of sanctions and grouping them by sex, I will
have dominant situations (Table 6):

Table 6. Metering re sults
Number of sanctions
0 1 2 3
M 2 1 0 2
F 2 2 2 0
Total 4 3 2 0

The solution would be to combine the columns so that no exact data can be extracted. Combine
the columns with "0" with "1" and "2" with "3". The result is as follows (Table 7):

Table 7. Combination of results
Number of sanctions
0 or 1 2 or 3
M 3 2
F 4 2
Total 7 4

Student: Beniamin Scrob
Emanuel University of Oradea
In order to prevent attacks, it is necessary to keep a detailed record for each user, even if this
involves complex activity and involves time.

An analysis of requests that may be ill -intentioned should also be made.

Security of multi -level databases
Three basic features of database security are distinguished:

➢ The security of a single element may be different from the security of another item in the
same record or the value of the same attribute. This involves implementing securi ty for
each element.

➢ Several security perimeters are required, which will be areas of access to certain data that
sometimes overlap.

➢ The security of an entire may be different from the security of an individual element. This
may be higher or lower.

To ensure the security of the databases, the following methods can be applied:

➢ Partitioning the database
The database is divided into separate databases, each with its own security
level. The operation is also called atomization of the database. As a side
effect, this operation will destroy the main advantage of the database but
improves accuracy.

➢ Encryption
If sensitive data is encrypted, a user who accidentally finds sensitive data
will not be able to interpret and use it. However, this encryption is
vulne rable to clear -text attacks or when the attacker substitutes the
encryption form with another. In order to prevent this, the following can be
done:

a. using different encryption for the same record and different keys
for each field;

b. Encryption of recording fields using block chaining (CBC, CFB,
etc.).
➢ blocking integrity
It is a way used both to block integrity and to limit access to the database.
The method also bears the name "spray paint" because each element is
colored depending on its sensitivi ty. The color is maintained with the
element it characterizes and not in a separate database.

Student: Beniamin Scrob
Emanuel University of Oradea
Each date will contain three elements:

Classification The amount of control

Date

Project Uranus
Strict Secret (SS) 1027

Data will be stored in clear text to increase efficiency.
With regard to classification, it must be:
❖ non-verifiable – a rash user will not be able to create a new
sensitive date for an item;
❖ unique – the perverse user will not be able to copy a sensitivi ty
level from another element;
❖ secret – the cursing user will not be able to determine the
sensitivity for any object.

The amount of cryptographic control, in order to be unique, must contain
data about:
❖ registration;
❖ camp;
❖ element data (Figure 1).

Figure 1. The amount of cryptographic control

Encryption Key K

No. registration 23

Attribute transfer

Date value project

Classification strict secret

Encryption mechanism
1027

Student: Beniamin Scrob
Emanuel University of Oradea

➢ blocking sensitivity
Sensitivity blocking is a combination of two elements:
❖ the existence of a unique identifier (registration number);
❖ security level.

It should not be possible to find two elements that have the same level of
security just by looking into the security portion of the integrity lock. As a
result of encryption, the blocking content, especially the security level, is
hidden.

➢ front -end security
Front -end security (also known as guard) is provided by a monitor
mechanism.

The sequence of interacti ons between the user and the front -end mechanism
is the following:

❖ the user identifies the front end;
❖ the user sends a request to the front -end mechanism;
❖ the front -end mechanism checks the user's authorization to access
the data;

❖ the front -end mechanism sends a request to the database
management system ( DMS );
❖ the database management system ( DMS ) performs an I / O access
operation;
❖ the database management system ( DMS ) sends the result of the
query to the front -end mechanism;
❖ the front -end mechani sm checks the validity of the extracted data
with the checksums and checks whether the data can be made
available to the user according to the user's access level;
❖ the front -end mechanism will form the data for the user;
❖ data is transmitted to the user.

➢ commutative filter
The commutative filter interacts with both the user and the database
management system (DMS).

The commutative filter will reformulate requests as follows:

Student: Beniamin Scrob
Emanuel University of Oradea
❖ The database management system (DMS) will perform as many
possible tasks as possib le by rejecting as many unacceptable
requests as revealing sensitive data;
❖ selecting the data the user can access.

The commutative filter can be used on both the recordings and the attributes
or elements.

At the record level, the filter requires the desi red data plus the amount of
cryptographic control; if they check the accuracy and accessibility of the
data, then they can be sent to the user.

At the attribute level, the filter checks if all the attributes in the user request
are accessible to the user, and if so, sends the request to the database
manager. Upon return, it will delete all requests that the user can not access.

At element level, the system will request the requested data and the
cryptographic check amount. When returned, they check security
membership for each item.

➢ views of the database
Views are a subset of the database the user can access. The views can be a
subset of the data base for a single user, in which case the requests of the
other users will access the same data type. Data that is accessible to a user
is obtained by filtering the contents of the original database. The user is not
aware of the presence of the tuples miss ing from that view. A view can be
defined from multiple tables for which the user has the proper privilege to
use, but not the use of the base tables. The use of views in this case is more
restrictive than simply holding the privileges granted to the user on the base
tables. The DMS stores the view definition in the database. When the DMS
encounters a reference to a view, it searches for this definition and converts
that request into an equivalent request to the tables that constitute the source
of the view, then executes the request.

The use of views to ensure database security is also s een in the literature as
discretionary security and is characteristic of SQL -based systems.

The use of database views has the effect of creating facilities in operation,
but it can also bring disadvantages (table 8).

Student: Beniamin Scrob
Emanuel University of Oradea

Table 8. Advantages and disadva ntages of using views
Advantages
Disadvantages
Simplification of applications

Multiple queries applied to
multiple databases can only be
applied to a single view. Performances

Queries on views can be converted
into queries on the base tables,
which will be reflected in
performance.

Structural simplicity

Views can create a personal view of
the database to interpret the results. Restrictions on updates

Many of the views can be read –
only; in this case, no updates can be
made.

Security

Restrict access to data for users.

2. Architectures to ensure security databases

At present, there are two major approaches to securing database security. These consider the trust
that can be given to the two elements that will interact with the databas es, namely the database
management system ( DMS ) and the operating system (OS) .

Taking these into account, we will have the following two approaches:
❖ Architecture with secure subjects;
❖ Architecture with uncertain subjects.

Architecture with secure subjects starts from the assumption that both the DMS and SO that
interact with databases are safe (reliable). This approach is common to most the DMS (Sybase,
Informix, Ingres, Oracle, DEC, Rubix).

The architecture with uncertain subjects starts from the assumption that OS is safe, but the DMS
is uncertain.

This architecture is implemented in three variants:
❖ Architecture with integrity blocking;
❖ integrated architecture;
❖ replicated / distrib uted architecture.

Student: Beniamin Scrob
Emanuel University of Oradea
This type of architecture is implemented in the TRUEDATA and Oracle the DMS , as well as in
lesser -known prototypes, Miter and SeaView.

In the case of integrated architecture, access is through two distinct types of the DMS . The
operati ng system, which is considered safe, will play an important role in providing access to data
that is disposed on the disk. In terms of data access, a similarity to the replicated / distributed
architecture is observed .

The replicated / distributed archite cture uses a secure replication mechanism that will channel
requests. The full user (superuser) will also have access to restricted user data (if specified in access
rights) through this mechanism.

Student: Beniamin Scrob
Emanuel University of Oradea

References
http://www.qreferat.com/referate/informatica/PROTECTIA -BAZEI -DE-DATE527.php
http://andrei.clubcisco.ro/cursuri/f/fsym/3bd/12.%20Securitatea%20si%20protectia%20datelor.
pdf
https://ro.wikipedia.org/wiki/Sistem_de_gestiune_a_bazelo r_de_date
http://php.net/manual/ro/security.database.php
https://en.wikipedia.org/wiki/Database_security
http://www.scritub.com/stiinta/informatica/sql/Securitate -si-integritate -in-b1831022218.php
http://www.rasfoiesc.com/educatie/informatica/baze -de-date/Securizarea -bazelor -de-date47.php
http://www.scrigroup.com/ca lculatoare/baze -de-date/PROTECTIA -BAZELOR -DE-DATE –
Inte23282.php
https://www.techopedia.com/definition/29841/database -security
https://www.esecurityplanet.com/network -security/6 -database -security -best-practices.html