Database Management System Chapter-6

Chapter 6:  Integrity and Security

 

n Domain Constraints

n Referential Integrity

n Assertions

n Triggers

n Security

n Authorization

n Authorization in SQL

 

Domain Constraints

 

n Integrity constraints guard against accidental damage to the database, by ensuring that authorized changes to the database do not result in a loss of data consistency.

n Domain constraints are the most elementary form of integrity constraint.

n They test values inserted in the database, and test queries to ensure that the comparisons make sense.

n New domains can be created from existing data types

H E.g.   create domain Dollars numeric(12, 2)
         create domain Pounds numeric(12,2)

n We cannot assign or compare a value of type Dollars to a value of type Pounds. 

H However, we can convert type as below
         (cast r.A as Pounds)
(Should also multiply by the dollar-to-pound conversion-rate)

 

 

n The check clause in SQL-92 permits domains to be restricted:

H Use check clause to ensure that an hourly-wage domain allows only values greater than a specified value.

     create domain hourly-wage numeric(5,2)
           constraint value-test check(value > = 4.00)

H The domain has a constraint that ensures that the hourly-wage is greater than 4.00

H The clause constraint value-test is optional; useful to indicate which constraint an update violated.

n Can have complex conditions in domain check

H create domain AccountType char(10)
    constraint account-type-test
          check (value in (‘Checking’, ‘Saving’))

H check (branch-name in (select branch-name from branch))

 

Referential Integrity

 

n Ensures that a value that appears in one relation for a given set of attributes also appears for a certain set of attributes in another relation.

H Example:  If “Perryridge” is a branch name appearing in one of the tuples in the account relation, then there exists a tuple in the branch relation for branch “Perryridge”.

n Formal Definition

H Let r₁(R₁) and r₂(R₂) be relations with primary keys K₁ and K₂ respectively.

H The subset a of R₂ is a foreign key referencing K₁ in relation r₁, if for every t₂ in r₂ there must be a tuple t₁ in r₁ such that t₁[K₁] = t₂[a].

_H   Referential integrity constraint also called subset dependency since its can be written as
     Õ_a (r₂) Í Õ_K1 (r₁)

Referential Integrity in the E-R Model

 

n Consider relationship set R between entity sets E₁ and E₂.  The relational schema for R includes the primary keys K₁ of E₁ and K₂ of E₂.
Then K₁ and K₂ form foreign keys on the relational schemas for E₁ and E₂ respectively.

 

n Weak entity sets are also a source of referential integrity constraints.

H For the relation schema for a weak entity set must include the primary key attributes of the entity set on which it depends

 

 

 

 

 

 

Checking Referential Integrity on Database Modification

 

n The following tests must be made in order to preserve the following referential integrity constraint:

          Õ_a (r₂) Í Õ_K (r₁)

n Insert.  If a tuple t₂ is inserted into r₂, the system must ensure that there is a tuple t₁ in r₁ such that t₁[K] = t₂[a].  That is

                 t₂ [a] Î Õ_K (r₁)

n Delete.  If a tuple, t₁ is deleted from r₁, the system must compute the set of tuples in r₂ that reference t₁:

                s_{a = t1[K]} (r₂)

    If this set is not empty

H either the delete command is rejected as an error, or

_H   the tuples that reference t₁ must themselves be deleted
(cascading deletions are possible).

 

Database Modification (Cont.)

 

n  Update.  There are two cases:

H  If a tuple t₂ is updated in relation r₂ and the update modifies values for foreign key a, then a test similar to the insert case is made:

4  Let t₂’ denote the new value of tuple t₂.  The system must ensure that

           t₂’[a] Î Õ_K(r₁)

H  If a tuple t₁ is updated in r₁, and the update modifies values for the primary key (K), then a test similar to the delete case is made:

4 The system must compute
     s_{a = t1[K]} (r₂)
 using the old value of t₁ (the value before the update is applied).

4 If this set is not empty

–     the update may be rejected as an error, or

–    the update may be cascaded to the tuples in the set, or

–    the tuples in the set may be deleted.

 

Referential Integrity in SQL

 

n  Primary and candidate keys and foreign keys can be specified as part of the SQL create table statement:

H The primary key clause lists attributes that comprise the primary key.

H The unique key clause lists attributes that comprise a candidate key.

H The foreign key clause lists the attributes that comprise the foreign key and the name of the relation referenced by the foreign key.

n  By default, a foreign key references the primary key attributes of the referenced table

     foreign key (account-number) references account

n  Short form for specifying a single column as foreign key

   account-number char (10) references account

n  Reference columns in the referenced table can be explicitly specified

H but must be declared as primary/candidate keys

   foreign key (account-number) references account(account-number)

 

Referential Integrity in SQL – Example

 

create table customer
(customer-name char(20),
customer-street  char(30),
customer-city     char(30),
primary key (customer-name))

create table branch
(branch-name    char(15),
branch-city        char(30),
assets       integer,
primary key (branch-name))

 

Referential Integrity in SQL – Example (Cont.)

 

create table account
(account-number        char(10),
branch-name     char(15),
balance   integer,
primary key (account-number),
foreign key (branch-name) references branch)

create table depositor
(customer-name char(20),
account-number          char(10),
primary key (customer-name, account-number),
foreign key (account-number) references account,
foreign key (customer-name) references customer)

 

Cascading Actions in SQL

 

create table account

          . . .
      foreign key(branch-name)       references branch
                on delete cascade
                on update cascade
      . . . )

n Due to the on delete cascade clauses, if a delete of a tuple in branch results in referential-integrity constraint violation, the delete “cascades” to the account relation, deleting the tuple that refers to the branch that was deleted.

 

n Cascading updates are similar.

 

n If there is a chain of foreign-key dependencies across multiple relations, with on delete cascade specified for each dependency, a deletion or update at one end of the chain can propagate across the entire chain.

n If a cascading update to delete causes a constraint violation that cannot be handled by a further cascading operation, the system aborts the transaction.

H  As a result, all the changes caused by the transaction and its cascading actions are undone.

n Referential integrity is only checked at the end of a transaction

H Intermediate steps are allowed to violate referential integrity provided later steps remove the violation

H Otherwise it would be impossible to create some database states, e.g. insert two tuples whose foreign keys point to each other

4 E.g. spouse attribute of relation
           marriedperson(name, address, spouse)

 

Referential Integrity in SQL (Cont.)

 

n Alternative to cascading:

H on delete set null

H on delete set default

n Null values in foreign key attributes complicate SQL referential integrity semantics, and are best prevented using not null

H if any attribute of a foreign key is null, the tuple is defined to satisfy the foreign key constraint!

 

Assertions

 

n An assertion is a predicate expressing a condition that we wish the database always to satisfy.

n An assertion in SQL takes the form

          create assertion <assertion-name> check <predicate>

n When an assertion is made, the system tests it for validity, and tests it again on every update that may violate the assertion

H This testing may introduce a significant amount of overhead; hence assertions should be used with great care.

n Asserting
      for all X, P(X)
is achieved in a round-about fashion using  
      not exists X such that not P(X)

 

Assertion Example

 

n The sum of all loan amounts for each branch must be less than the sum of all account balances at the branch.

     create assertion sum-constraint check
     (not exists (select * from branch
                           where (select sum(amount) from loan
                                             where loan.branch-name =
                                                 branch.branch-name)
                                  >= (select sum(amount) from account
                                             where loan.branch-name =
                                                 branch.branch-name)))

 

Assertion Example

 

n Every loan has at least one borrower who maintains an account with a minimum balance or $1000.00

    create assertion balance-constraint check
    (not exists (
         select * from loan
             where not exists (
             select *
                         from borrower, depositor, account
                         where loan.loan-number = borrower.loan-number
                                    and borrower.customer-name = depositor.customer-name
                                    and depositor.account-number = account.account-number
                                    and account.balance >= 1000)))

 

 

Triggers

 

n A trigger is a statement that is executed automatically by the system as a side effect of a modification to the database.

n To design a trigger mechanism, we must:

H Specify the conditions under which the trigger is to be executed.

H Specify the actions to be taken when the trigger executes.

n Triggers introduced to SQL standard in SQL:1999, but supported even earlier using non-standard syntax by most databases.

n                         

Trigger Example

 

n Suppose that instead of allowing negative account balances, the bank deals with overdrafts by

H setting the account balance to zero

H creating a loan in the amount of the overdraft

H giving this loan a loan number identical to the account number of the overdrawn account

n The condition for executing the trigger is an update to the account relation that results in a negative balance value.

 

Trigger Example in SQL:1999

 

    create trigger overdraft-trigger after update on account
referencing new row as nrow                                                                                  for each row
when nrow.balance < 0
begin atomic
      insert into borrower
                (select customer-name, account-number
                 from depositor
                where nrow.account-number =
                          depositor.account-number);
        insert into loan values
                (n.row.account-number, nrow.branch-name,
                                                                – nrow.balance);
        update account set balance = 0
      where account.account-number = nrow.account-number
end         

 

Triggering Events and Actions in SQL

 

n Triggering event can be insert, delete or update

n Triggers on update can be restricted to specific attributes

H E.g.  create trigger overdraft-trigger after update of balance on account

n Values of attributes before and after an update can be referenced

H referencing old row as   : for deletes and updates

H referencing new row as  : for inserts and updates

n Triggers can be activated before an event, which can serve as extra constraints.  E.g. convert blanks to null.

          create trigger setnull-trigger before update on r
      referencing new row as nrow
      for each row
          when nrow.phone-number = ‘ ‘
          set nrow.phone-number = null

 

 

Statement Level Triggers

 

n Instead of executing a separate action for each affected row, a single action can be executed for all rows affected by a transaction

H Use     for each statement      instead of    for each row

H Use     referencing old table   or   referencing new table   to refer to temporary tables  (called transition tables) containing the affected rows

H Can be more efficient when dealing with SQL statements that update a large number of rows

 

External World Actions

 

n We sometimes require external world actions to be triggered on a database update

H E.g. re-ordering an item whose quantity in a warehouse has become small, or turning on an alarm light,

n Triggers cannot be used to directly implement external-world actions, BUT

H Triggers can be used to record actions-to-be-taken in a separate table

H Have an external process that repeatedly scans the table, carries out external-world actions and deletes action from table

n E.g.  Suppose a warehouse has the following tables

H inventory(item, level):  How much of each item is in the warehouse

H minlevel(item, level) :   What is the minimum desired level of each item

H reorder(item, amount):  What quantity should we re-order at a time

H orders(item, amount)  :  Orders to be placed (read by external process)

 

 

 

create trigger reorder-trigger after update of amount on inventory

referencing old row as orow, new row as nrow

for each row

         when nrow.level < = (select level

                                     from minlevel

                                     where minlevel.item = orow.item)

                    and orow.level > (select level

                                          from minlevel

                                              where minlevel.item = orow.item)

   begin

          insert into orders

                  (select item, amount

                    from reorder

                    where reorder.item = orow.item)

   end

 

 

Triggers in MS-SQLServer Syntax

 

     create trigger overdraft-trigger on account
for update
as
if  inserted.balance < 0
begin
    insert into borrower
       (select customer-name,account-number
        from depositor, inserted
        where inserted.account-number =
                             depositor.account-number)
    insert into loan values
       (inserted.account-number, inserted.branch-name,
                                – inserted.balance)
    update account set balance = 0
      from account, inserted
      where account.account-number = inserted.account-number
end

 

When Not To Use Triggers

 

n Triggers were used earlier for tasks such as

H maintaining summary data (e.g. total salary of each department)

H Replicating databases by recording changes to special relations (called change or delta relations) and having a separate process that applies the changes over to a replica

n There are better ways of doing these now:

H Databases today provide built in  materialized view  facilities to maintain summary data

H Databases provide built-in support for replication

n Encapsulation facilities can be used instead of triggers in many cases

H Define methods to update fields

H Carry out actions as part of the update methods instead of
through a trigger

 

Security

 

n Security - protection from malicious attempts to steal or modify data.

H Database system level

4 Authentication and authorization mechanisms to allow specific users access only to required data

4 We concentrate on authorization in the rest of this chapter

H Operating system level

4 Operating system super-users can do anything they want to the database!   Good operating system level security is required.

H Network level:  must use encryption to prevent

4 Eavesdropping (unauthorized reading of messages)

4 Masquerading  (pretending to be an authorized user or sending messages supposedly from authorized users)

H Physical level

4 Physical access to computers allows destruction of data by intruders;  traditional lock-and-key security is needed

4 Computers must also be protected from floods, fire, etc.

–  More in Chapter 17 (Recovery)

H Human level

4 Users must be screened to ensure that an authorized users do not give access to intruders  

4 Users should be trained on password selection and secrecy

 

Authorization

 

 

Forms of authorization on parts of  the database:

n Read authorization - allows reading, but not modification of data.

n Insert authorization - allows insertion of new data, but not modification of existing data.

n Update authorization - allows modification, but not deletion of data.

n Delete authorization - allows deletion of data

 

Forms of authorization to modify  the database schema:

n Index authorization - allows creation and deletion of indices.

n Resources authorization - allows creation of new relations.

n Alteration authorization - allows addition or deletion of attributes in a relation.

n Drop authorization - allows deletion of relations.

 

Authorization and Views

 

n Users can be given authorization on views, without being given any authorization on the relations used in the view definition

n Ability of views to hide data serves both to simplify usage of the system and to enhance security by allowing users access only to data they need for their job

n A  combination or relational-level security and view-level security can be used to limit a user’s access to precisely  the data that user needs.

 

 

View Example

 

n Suppose a  bank clerk needs to know the names of the customers of each branch, but is not authorized to see specific loan information.

H Approach: Deny direct access to the loan relation, but grant access to the view cust-loan, which consists only of  the names of customers and the branches at which they have a loan.

H The cust-loan view is defined in SQL as follows:

          create view cust-loan as
          select branchname, customer-name
          from   borrower, loan
          where borrower.loan-number = loan.loan-number

 

n The clerk is authorized to see the result of the query:

   select *
from cust-loan

n When the query  processor translates the result into a query on the actual relations in the database, we obtain a query on borrower and loan.

n Authorization must be checked on the clerk’s query  before query processing replaces a view by the definition of the view.

 

Authorization on Views

 

n Creation of view does not require resources authorization since no real relation is being created

n The creator of a view gets only those privileges that provide no additional authorization beyond that he already  had.

n E.g. if creator of view cust-loan had only read authorization on borrower and loan, he gets only read authorization on cust-loan

 

 

Granting of Privileges

 

n The passage of authorization from one user to another may be represented by an authorization graph.

n The nodes of this graph are the users.

n The root of the graph is the database administrator.

n Consider graph for update authorization on loan.

_n   An edge U_i ®U_j indicates that user U_i has granted update authorization on loan to U_j.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Authorization Grant Graph

 

n Requirement: All edges in an authorization graph must be part of some path originating with the database administrator

n If DBA revokes grant from U₁:

H Grant must be revoked from U₄ since U₁ no longer has authorization

H Grant must not be revoked from U₅ since U₅ has another authorization path from DBA through U₂

n Must prevent cycles of grants with no path from the root:

H DBA grants authorization to U₇

H U7 grants authorization to U₈

H U8 grants authorization to U₇

H DBA revokes authorization from U₇

n Must  revoke grant U₇ to U₈ and from U₈ to U₇ since there is no path from DBA to U₇ or to U₈ anymore.

 

Security Specification in SQL

 

n The grant statement is used to confer authorization

          grant <privilege list>

          on <relation name or view name> to <user list>

n <user list> is:

H a user-id

H public, which allows all valid users the privilege granted

H A role (more on this later)

n Granting a privilege on a view does not imply granting any  privileges on the underlying relations.

n The grantor of the privilege must already hold the privilege on the specified item (or be the database administrator).

 

Privileges in SQL

 

n  select: allows read access to relation,or the ability to query using the view

H Example: grant users U₁, U₂, and U₃ select authorization on the branch relation:

                   grant select on branch to U₁, U₂, U₃

n  insert: the ability to insert tuples

n  update: the ability  to update using the SQL update statement

n  delete: the ability to delete tuples.

n  references: ability to declare foreign keys when creating relations.

n  usage: In SQL-92; authorizes a user to use a specified domain

n  all privileges: used as a short form for all the allowable privileges

 

Privilege  To Grant Privileges

 

 

n with grant option: allows a user who is granted a privilege to pass the privilege on to other users.

H Example:

grant select on branch to U₁ with grant option

gives U₁ the select privileges on branch and allows U₁ to grant this

privilege to others

Roles

 

n Roles permit common privileges for a class of users can be specified just once by creating a corresponding “role”

n Privileges can be granted to or revoked from roles, just like user

n Roles can be assigned to users, and even to other roles

n SQL:1999 supports roles

   create role  teller
create role manager

    grant select on branch to  teller
grant update (balance) on account to teller
grant all privileges on account to manager
grant teller to manager
grant teller to alice, bob
grant  manager  to  avi

 

Revoking Authorization in SQL

 

n The revoke statement is used to revoke authorization.

revoke<privilege list>

on <relation name or view name> from <user list> [restrict|cascade]

n Example:

revoke select on branch  from U₁, U₂, U₃ cascade

n Revocation of a privilege from a user may cause other users also to lose that privilege; referred to as cascading of the revoke.

n We can prevent cascading by specifying restrict:

   revoke select on branch from U₁, U₂, U₃ restrict

    With restrict, the revoke command fails if  cascading revokes are required.

n <privilege-list> may be all to revoke all privileges the revokee may hold.

n If <revokee-list> includes public all users lose the privilege except those granted it explicitly.

n If the same privilege was granted twice to the same user by different grantees, the user  may  retain the privilege after the revocation.

n All privileges that depend on the privilege being revoked are also revoked.

 

Limitations of SQL Authorization

 

n  SQL does not support authorization at a tuple level

H E.g. we cannot restrict students to see only (the tuples storing) their own grades

n  With the growth in Web access to databases, database accesses come primarily from application servers.

H  End users don't have database user ids, they are all mapped to the same database user id

n  All end-users of an application (such as a web application) may be mapped to a single database user

n  The task of authorization in above cases falls on the application program, with no support from SQL

H Benefit: fine grained authorizations, such as to individual tuples, can be implemented by the application.

H Drawback: Authorization must  be done in application code, and may be dispersed all over an application

H Checking for absence of authorization loopholes becomes very difficult since it requires reading large amounts of application code

 

Audit Trails

 

n An audit trail is a log of all changes (inserts/deletes/updates) to the database along with information such as which user performed the change, and when the change was performed.

n Used to track erroneous/fraudulent updates.

n Can be implemented using triggers, but many database systems provide direct support.

Encryption

 

n Data may be encrypted when database authorization provisions do not offer sufficient protection.

n Properties of good encryption technique:

H Relatively simple for authorized users to encrypt and decrypt data.

H Encryption scheme depends not on the secrecy of the algorithm but on the secrecy of a parameter of the algorithm  called the  encryption key.

H Extremely difficult for an intruder to determine the encryption key.

 

n   Data Encryption Standard (DES) substitutes characters and rearranges their order on the basis of an encryption key which is  provided to authorized users via a secure mechanism. Scheme is no more secure than the key transmission mechanism since the key has to be shared.

n  Advanced Encryption Standard (AES) is a new standard replacing DES, and is based on the Rijndael algorithm, but is also dependent on shared secret keys

n   Public-key encryption is based on each user having two keys:

H public key – publicly published key used to encrypt data, but cannot be used to decrypt data

H  private key -- key known only to individual user, and used to decrypt data.
Need not be transmitted to the site doing encryption.

     Encryption scheme is such that it is impossible or extremely hard to decrypt data given only  the public key.

n  The RSA  public-key encryption scheme is based on the hardness of factoring a very large number (100's of digits) into its prime components.

 

Authentication

 

n Password based authentication is widely used, but is susceptible to sniffing on a network

n Challenge-response systems avoid transmission of passwords

H DB sends a (randomly generated) challenge string to user

H User encrypts string and returns result.

H DB verifies identity by decrypting result

H Can use public-key encryption system by DB sending a message encrypted using user’s public key, and user decrypting and sending the message back

n Digital signatures are used to verify authenticity of data

H E.g. use private key (in reverse) to encrypt data, and anyone can verify authenticity by using public key (in reverse) to decrypt data.  Only holder of private key could have created the encrypted data.

H Digital signatures also help ensure nonrepudiation: sender
cannot later claim to have not created the data

 

Digital Certificates

 

n Digital certificates are used to verify authenticity of public keys.

n Problem: when you communicate with a web site, how do you know if you are talking with the genuine web site or an imposter?

H Solution: use the public key of the web site

H Problem: how to verify if the public key itself is genuine?

n Solution:

H Every client (e.g. browser) has public keys of a few root-level certification authorities

H A site can get its name/URL and public key signed by a certification authority: signed document is called a certificate

H Client can use public key of certification authority to verify certificate

H Multiple levels of certification authorities can exist. Each certification authority

4 presents its own public-key certificate signed by a
higher level authority, and

4 Uses its private key to sign the certificate of  other web sites/authorities

End of Chapter