A Validation Model of Data Input for Web Services

A Validation Model of Data Input
for Web Services
Rafael B. Brinhosa, Carla M. Westphall, Carlos B.
Westphall, Daniel R. dos Santos, Fábio Grezele
The Twelfth International Conference on Networks - ICN 2013
January 27 - February 1, 2013 - Seville, Spain
Westphall, Daniel R. dos Santos, Fábio Grezele
Post Graduation Program in Computer Science
Federal University of Santa Catarina
{ brinhosa,carlamw,westphal,danielrs,fgrezele }@inf.ufsc.br
1

Content at a GlanceContent at a Glance
• Introduction and Related Works
• Security Issues in Web Services
• A Validation Model of Data Input for Web
Service (WSIVM)Service (WSIVM)
• Implementation Results
– Development
– Case Study
• Conclusions and Future Works
2

• SOA is based on web services but there are
security related concerns
• The lack of proper input validation is a major
cause of data breaches and Web application
attacks
IntroductionIntroduction
• Application attacks: SQL injection and cross-site
scripting (XSS)
Web Services Input Validation Model: an XML schema,
an XML specification and a module for performing input
validation according to the schema
3

IntroductionIntroduction –– Web ServicesWeb Services
Find Web services which meet
certain requirements
(Universal Description, Discovery and
Integration)
Services describe their own
properties and methodsproperties and methods
(Web Services Description Language)
Format of requests(client) and
responses (server)
(Simple Object Access Protocol)
Message transfer protocol
(Hypertext Transfer Protocol)
4

IntroductionIntroduction –– Web ServicesWeb Services
Figure available from: http://gdp.globus.org/gt4-tutorial 5

Related WorkRelated Work
• Lack of input validation is a major cause of Web
application attacks
– SANS, 2011: The Top Cyber Security Risks
– OWASP 2010: OWASP top 10 Web application
security risks
– [T. Scholte, D. Balzarotti, E. Kirda, 2011] - “Quo– [T. Scholte, D. Balzarotti, E. Kirda, 2011] - “Quo
vadis? A study of the evolution of input
validation vulnerabilities in Web applications”
• Few specific mechanisms for Web Services
• [N. A. Nordbotten, 2009] [L. Sun and Y. Li, 2008]
use XML security technologies (encryption)
6

• WS-Security Wrapper: is an intermediate between
the Web service and the client; is an adapter
program that converts plain XML exchanges to and
from SOAP with WS-Security (XML signature and
encryption). It does not include features such as
validation of predefined data entries
• [J. Lin and J. Chen, 2009]• [J. Lin and J. Chen, 2009]
– Collects web pages (crawler), identify weak
points and test them
– insert the input validation (meta-programs) on
the server side, acting as a web application
firewall
– many false positives with blacklist approach
7

• IAPF (Integrated Application and Protocol Framework)
[N. Sidharth and J. Liu, 2007]:
– Protection in UDDI, WSDL, SOAP (WS-Sec)
• XML firewall:
– [A. Blyth, 2009] is concerned with validation of– [A. Blyth, 2009] is concerned with validation of
the structure of XML content but not the
content itself
– [Y. Loh, W. Yau, C. Wong, and W. Ho, 2006]
mentions protection against SQL injection
through an XML schema and a precompiled
blacklist of SQL commands, an approach which
tends to produce many false positives
8

• WSDL scanning: to reveal sensitive information
about invocation patterns, underlying technology
implementations
• Serious and important data manipulation attacks:
SQL injection and XSS
• Normal firewalls, antivirus and using WS-Security
Security Issues in Web ServicesSecurity Issues in Web Services
• Normal firewalls, antivirus and using WS-Security
standards are not able to protect web services
against SQL injection and XSS attacks
9

Databases
LegacySystems
WebServices
Directories
HumanResrcs
Billing
Custom Code
APPLICATION
ATTACK
ApplicationLayer
Accounts
Finance
Administration
Transactions
Communication
KnowledgeMgmt
E-Commerce
Bus.Functions
HTTP
request
SQL
query
DB TableHTTP
response
"SELECT * FROM
accounts WHERE
acct=‘’ OR 1=1--
’"
1. Application presents a form to
the attacker
2. Attacker sends an attack in the
Account Summary
Acct:5424-6066-2134-4334
Acct:4128-7574-3921-0192
Acct:5424-9383-2039-4029
Acct:4128-0004-1234-0293
Account:
SKU:
Account:
SKU:
SQL InjectionSQL Injection –– IllustratedIllustrated
Firewall
Hardened OS
Web Server
App Server
Firewall
NetworkLayer
2. Attacker sends an attack in the
form data
3. Application forwards attack to
the database in a SQL query
4. Database runs query containing
attack and sends encrypted results
back to application
5. Application decrypts data as
normal and sends results to the user
Available from: https://www.owasp.org/index.php/Category:OWASP_Top_Ten_Project 10

• SQL Injection: by sending SOAP requests with
properly handled parameters, for example, “'"1=1
–” as a parameter for a particular service
ERROR: The query was not accomplished. Description:
1064 - You have an error in your SQL syntax; check the
manual that corresponds to your MySQL server versionmanual that corresponds to your MySQL server version
for the right syntax to use near '1=1'' at line 1
Line 11: Incorrect syntax near '')) or ItemId in (select
ItemId from dbo.GetItemParents('4''. Unclosed
quotation mark before the character string ')) ) ) > 0 ‘
11

Heartland Payment SystemsHeartland Payment Systems
12Available from: http://www.databreaches.net/?p=7691

Heartland Payment SystemsHeartland Payment Systems
Available from: http://www.wired.com/images_blogs/threatlevel/2009/08/gonzalez.pdf 13

CrossCross--Site Scripting IllustratedSite Scripting Illustrated
Application with
stored XSS
vulnerability
Attacker sets the trap – update my profile
Attacker enters a
malicious script into a
web page that stores
the data on the server
1
Administration
Transactions
Communication
Knowledge
Commerce
Bus.Functions
3
2 Victim views page – sees attacker profile
Script silently sends attacker Victim’s session cookie
Script runs inside
victim’s browser with
full access to the DOM
and cookies
Custom Code
Accounts
Finance
Administration
Transactions
Communication
Knowledge
Mgmt
E-Commerce
Bus.Functions
Available from: https://www.owasp.org/index.php/Category:OWASP_Top_Ten_Project 14

• XSS: by presenting unvalidated data directly to
the user, Web services can be attacked. Using, for
example, the command
document.write(xmlhttp.responseText), if the
answer to this AJAX (Asynchronous JavaScript and
XML) call made to a Web service contains HTML
XML) call made to a Web service contains HTML
and JavaScript data, these data will be
interpreted and executed, posing a risk to the
user
15

• Validate input data to provide security for Web
Services. Controls on the lexical and syntactic
aspects, type checking
• Best way to avoid attacks: whitelist input
validation (sanitization - change input into an acceptable format)
WSIVMWSIVM –– Web Services Input Validation ModelWeb Services Input Validation Model
16

WSIVMWSIVM –– Web Services Input Validation ModelWeb Services Input Validation Model
17

WSIVMWSIVM –– DevelopmentDevelopment
• Apache Tomcat Web server
• Apache Axis2 framework for SOAP messages
• To implement the validation module for Apache
Axis2 the Rampart module was used
• It was chosen to intercept the message in the
phase PreDispatchphase PreDispatch
• WSIVMXMLSchema, WSIVMXMLSpecification, and
WSIVM Rampart module
18

OperationName: the name of the operation
SanitizeOperation: defines whether the parameters of
this operation can be reformulated if necessary for the
removal of characters that are not accepted
ParamName: the name of the parameter or field
Allowed: an allowed field type, which is valid (text, html,
html+java-script, email, number, and all)
WSIVMWSIVM –– WSIVMXMLSpecificationWSIVMXMLSpecification
html+java-script, email, number, and all)
Length: specifies the exact size of the field
Maxsize: specifies the maximum field size
Minsize: specifies the minimum field size
Nillable: determines whether or not it is possible that
the field is null (true or false)
regEx: allows a regular expression to be specified for
validation 19

Case StudyCase Study
• Case Study: Client Application + Web Service
• UniversityManager web service:
– searchStudent receives a registration number
(ID) and returns the student record containing
a String with his or her information
– registerStudent operation receives the– registerStudent operation receives the
information, which must not contain HTML or
Javascript code, and registers it on the MySQL
database. In the database a student’s table is
created: ID, name, age, email, comment,
site, and birthday
20

WSIVMXMLSpecification
WSIVMXMLSpecification–UniversityManager
<?xml version="1.0" encoding="UTF-8"?>
<valid_inputs_specification mlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
WebServiceID=“UniversityManager"
xsi:noNamespaceSchemaLocation="valid_inputs_specification.xsd">
<operation name="registerStudent ">
<input name="name" type="String" min="5" max="20" accept="text" sanitize="true"/>
<input name="age" type="Integer" min="0" max="150" accept="number"
sanitize="true"/>
<input name="email" type="String" min="0" max="200" accept="email" sanitize="true"/><input name="email" type="String" min="0" max="200" accept="email" sanitize="true"/>
<input name="comment" type="String" min="0" max="200" accept="text"
sanitize="true"/>
<input name="site" type="String" min="0" max="300" accept="url" sanitize="true"/>
<input name="data" type="String" min="0" max="200" accept="regex" regexpattern=
"(d{4})-(d{2})-(d{2})" sanitize="true"/>
</operation>
<operation name="searchStudent ">
<input name="id" type="Integer" min="0" max="10000" accept="number"
sanitize="true"/>
</operation>
</valid_inputs_specification>
21

Case StudyCase Study -- ResultsResults
• Experiments:
– With and without WSIVM
– 150 users
– The test runs for 300 seconds
– soapUi: direct calls to the web service
<soap:Envelope xmlns:soap="http://www.w3.org/2003/05/soap-envelope"
xmlns:univ="http://university.wsivm.example">xmlns:univ="http://university.wsivm.example">
<soap:Header/><soap:Body>
<univ:registerStudent>
<univ:name>John</univ: name >
<univ:age>12</univ: age >
<univ:email>john@hsj.com</univ:email>
<univ:comment>Passed</univ: comment >
<univ:site>http://www.univ.com</univ:site>
<univ:birthday>1980-09-12</univ: birthday >
</univ: registerStudent >
</soap:Body></soap:Envelope>
22

ComparisonComparison
Min.Min.
TimeTime
Without
WSIVM
35 ms
Max.Max.
TimeTime
27848
ms
Case StudyCase Study -- ResultsResults
AvgAvg..
TimeTime
2494,85
ms
TransferredTransferred
BytesBytes
1974195 B
B/sB/s
6506 B/s
InsertionsInsertions
In DBIn DB
10078
With WSIVM
64 ms
Total 83 %
13346
ms
- 52 %
4541,24
ms
82 %
1236330 B
- 37 %
4012 B/s
- 38 %
5134
- 49 %
23

ConclusionsConclusions
• Reusable and independent mechanism for data
entry validation, regardless of the
implementation of the web service
• Based on the white list approach (reduction in
false positives)
– More reliable than the blacklist. If a blacklist is– More reliable than the blacklist. If a blacklist is
created based on the current version of HTML,
in the case of new versions, this list may no
longer be considered valid
– The number of false positives or false
negatives will depend on the WSIVM XML
Specification defined
24

ConclusionsConclusions
– The framework provides the specification
to be customized according to the Web Service
requirements and needs
• Prevention of data injection attacks in Web
services and the waste of server processing with
invalid messagesinvalid messages
• Reduces the possibility of denial of service using
content of messages
• Negative impact on the performance of the
developed Web service but reduces the possibility
of inserting invalid data
• Solution for legacy applications reducing
development costs 25

• Optimization of the implementation to improve
the performance of the proposed model
• Development of a semi-automatic generator of
security specifications from WSDL
• Verification of SOAP messages and paths in XPath
format
Future WorkFuture Work
format
• Use of artificial intelligence or an anomaly
detection system
• Making a feedback loop filter validation of invalid
entries
26

Some ReferencesSome References
1. OWASP Top Ten - https://www.owasp.org/index.php/Category:OWASP_Top_Ten_Project
2. T. Scholte, D. Balzarotti, and E. Kirda, “Quo vadis? A study of the evolution of input validation
vulnerabilities in Web applications,” in Proc. Int. Conference on Financial Cryptography and Data
Security '11, St. Lucia, 2011.
3. N. A. Nordbotten, “XML and Web services security standards,” Communications Surveys & Tutorials,
IEEE, vol. 11, no. 3, pp. 4–21, 2009.
4. L. Sun and Y. Li, “XML and Web services security,” in Proc. 12th Int. Conf. Computer Supported
Cooperative Work in Design, CSCWD 2008, April 16–18, pp. 765–770.
5. N. Sidharth and J. Liu, “A framework for enhancing Web services security,” in Proc. 31st Ann. Int.
Computer Software and Applications Conf., 2007, COMPSAC 2007, Jul. 24–27, vol. 1, pp. 23–30.
6. WS-Security Wrapper - http://wsswrapper.sourceforge.net/
7. J. Lin and J. Chen, “An automated mechanism for secure input handling,” Journal of Computers, vol.7. J. Lin and J. Chen, “An automated mechanism for secure input handling,” Journal of Computers, vol.
4, no. 9, pp. 837–844, 2009.
8. N. Sidharth and J. Liu, “A framework for enhancing Web services security,” in Proc. 31st Ann. Int.
COMPSAC 2007, Jul. 24–27, vol. 1, pp. 23–30.
9. A. Blyth, “An architecture for an XML enabled firewall,” International Journal of Network Security,
vol. 8, no. 1, pp. 31–36, 2009, ISSN 1816–3548.
10. Y. Loh, W. Yau, C. Wong, and W. Ho, “Design and implementation of an XML firewall,” in Proc. 2006 Int.
Conf. Computational Intelligence and Security, Nov. 3–6, vol. 2, pp. 1147–1150.
11. CWE/SANS TOP 25 Most Dangerous Software Errors -
http://cwe.mitre.org/top25/archive/2011/2011_cwe_sans_top25.pdf
12. R. Wu and M. Hisada, “SOA Web Security and Applications”, Technology, vol. 9, no. 2, p. 163-177, 2010.
13. T. Scholte, W. Robertson, D. Balzarotti, E. Kirda, "Preventing Input Validation Vulnerabilities in Web
Applications through Automated Type Analysis," in Proc. 2012 IEEE 36th Annual COMPSAC 2012 , pp.233-
243, July 2012.
27

Thank you!Thank you!
Rafael B. Brinhosa, Carla M. Westphall, Carlos B. Westphall,
Daniel R. dos Santos, Fábio Grezele
{ brinhosa,carlamw,westphal,danielrs,fgrezele }@inf.ufsc.br
28

A Validation Model of Data Input for Web Services

More Related Content

What's hot (20)

Viewers also liked (9)

Similar to A Validation Model of Data Input for Web Services (20)

Recently uploaded (20)

A Validation Model of Data Input for Web Services