Software Installation/Uninstallation Testing
Have you performed software installation testing? How was the experience? Well, Installation testing (Implementation Testing) is quite interesting part of software testing life cycle.
Installation testing is like introducing a guest in your home. The new guest should be properly introduced to all the family members in order to feel him comfortable. Installation of new software is also quite like above example.
If your installation is successful on the new system then customer will be definitely happy but what if things are completely opposite. If installation fails then our program will not work on that system not only this but can leave user’s system badly damaged. User might require to reinstall the full operating system.
In above case will you make any impression on user? Definitely not! Your first impression to make a loyal customer is ruined due to incomplete installation testing. What you need to do for a good first impression? Test the installer appropriately with combination of both manual and automated processes on different machines with different configuration. Major concerned of installation testing is Time! It requires lot of time to even execute a single test case. If you are going to test a big application installer then think about time required to perform such a many test cases on different configurations.
We will see different methods to perform manual installer testing and some basic guideline for automating the installation process.
To start installation testing first decide on how many different system configurations you want to test the installation. Prepare one basic hard disk drive. Format this HDD with most common or default file system, install most common operating system (Windows) on this HDD. Install some basic required components on this HDD. Each time create images of this base HDD and you can create other configurations on this base drive. Make one set of each configuration like Operating system and file format to be used for further testing.
How we can use automation in this process? Well make some systems dedicated for creating basic images (use software’s like Norton Ghost for creating exact images of operating system quickly) of base configuration. This will save your tremendous time in each test case. For example if time to install one OS with basic configuration is say 1 hour then for each test case on fresh OS you will require 1+ hour. But creating image of OS will hardly require 5 to 10 minutes and you will save approximately 40 to 50 minutes!
You can use one operating system with multiple attempts of installation of installer. Each time uninstalling the application and preparing the base state for next test case. Be careful here that your uninstallation program should be tested before and should be working fine.
Installation testing tips with some broad test cases:
1) Use flow diagrams to perform installation testing. Flow diagrams simplify our task. See example flow diagram for basic installation testing test case.
Add some more test cases on this basic flow chart Such as if our application is not the first release then try to add different logical installation paths.
2) If you have previously installed compact basic version of application then in next test case install the full application version on the same path as used for compact version.
3) If you are using flow diagram to test different files to be written on disk while installation then use the same flow diagram in reverse order to test uninstallation of all the installed files on disk.
4) Use flow diagrams toautomate the testing efforts. It will be very easy to convert diagrams into automated scripts.
5) Test the installer scripts used for checking the required disk space. If installer is prompting required disk space 1MB, then make sure exactly 1MB is used or whether more disk space utilized during installation. If yes flag this as error.
6) Test disk space requirement on different file system format. Like FAT16 will require more space than efficient NTFS or FAT32 file systems.
7) If possible set a dedicated system for only creating disk images. As said above this will save your testing time.
8 ) Use distributed testing environment in order to carry out installation testing. Distributed environment simply save your time and you can effectively manage all the different test cases from a single machine. The good approach for this is to create a master machine, which will drive different slave machines on network. You can start installation simultaneously on different machine from the master system.
9) Try to automate the routine to test the number of files to be written on disk. You can maintain this file list to be written on disk in and excel sheet and can give this list as a input to automated script that will check each and every path to verify the correct installation.
10) Use software’s available freely in market to verify registry changes on successful installation. Verify the registry changes with your expected change list after installation.
11) Forcefully break the installation process in between. See the behavior of system and whether system recovers to its original state without any issues. You can test this “break of installation” on every installation step.
12) Disk space checking: This is the crucial checking in the installation-testing scenario. You can choose different manual and automated methods to do this checking. In manual methods you can check free disk space available on drive before installation and disk space reported by installer script to check whether installer is calculating and reporting disk space accurately. Check the disk space after the installation to verify accurate usage of installation disk space. Run various combination of disk space availability by using some tools to automatically making disk space full while installation. Check system behavior on low disk space conditions while installation.
13) As you check installation you can test for uninstallation also. Before each new iteration of installation make sure that all the files written to disk are removed after uninstallation. Some times uninstallation routine removes files from only last upgraded installation keeping the old version files untouched. Also check for rebooting option after uninstallation manually and forcefully not to reboot.
I have addressed many areas of manual as well as automated installation testing procedure. Still there are many areas you need to focus on depending on the complexity of your software under installation. These not addressed important tasks includesinstallation over the network, online installation, patch installation, Database checking on Installation, Shared DLL installation and uninstallation etc.
Hope this article will be a basic guideline to those having trouble to start with software installation testing both manually or in automation.
Regression Testing with Regression Testing Tools and methods
What is Regression Software Testing?
Regression means retesting the unchanged parts of the application. Test cases are re-executed in order to check whether previous functionality of application is working fine and new changes have not introduced any new bugs.
Regression means retesting the unchanged parts of the application. Test cases are re-executed in order to check whether previous functionality of application is working fine and new changes have not introduced any new bugs.
This is the method of verification. Verifying that the bugs are fixed and the newly added feature have not created in problem in previous working version of software.
Why regression Testing?
Regression testing is initiated when programmer fix any bug or add new code for new functionality to the system. It is a quality measure to check that new code complies with old code and unmodified code is not getting affected.
Most of the time testing team has task to check the last minute changes in the system. In such situation testing only affected application area in necessary to complete the testing process in time with covering all major system aspects.
Regression testing is initiated when programmer fix any bug or add new code for new functionality to the system. It is a quality measure to check that new code complies with old code and unmodified code is not getting affected.
Most of the time testing team has task to check the last minute changes in the system. In such situation testing only affected application area in necessary to complete the testing process in time with covering all major system aspects.
How much regression testing?
This depends on the scope of new added feature. If the scope of the fix or feature is large then the application area getting affected is quite large and testing should be thoroughly including all the application test cases. But this can be effectively decided when tester gets input from developer about the scope, nature and amount of change.
This depends on the scope of new added feature. If the scope of the fix or feature is large then the application area getting affected is quite large and testing should be thoroughly including all the application test cases. But this can be effectively decided when tester gets input from developer about the scope, nature and amount of change.
What we do in regression testing?
- Rerunning the previously conducted tests
- Comparing current results with previously executed test results.
Regression Testing Tools:
Automated Regression testing is the testing area where we can automate most of the testing efforts. We run all the previously executed test cases this means we have test case set available and running these test cases manually is time consuming. We know the expected results so automating these test cases is time saving and efficient regression testing method. Extent of automation depends on the number of test cases that are going to remain applicable over the time. If test cases are varying time to time as application scope goes on increasing then automation of regression procedure will be the waste of time.
Automated Regression testing is the testing area where we can automate most of the testing efforts. We run all the previously executed test cases this means we have test case set available and running these test cases manually is time consuming. We know the expected results so automating these test cases is time saving and efficient regression testing method. Extent of automation depends on the number of test cases that are going to remain applicable over the time. If test cases are varying time to time as application scope goes on increasing then automation of regression procedure will be the waste of time.
Most of the regression testing tools are record and playback type. Means you will record the test cases by navigating through the AUT and verify whether expected results are coming or not.
Example regression testing tools are:
Example regression testing tools are:
- Winrunner
- QTP
- AdventNet QEngine
- Regression Tester
- vTest
- Watir
- Selenium
- actiWate
- Rational Functional Tester
- SilkTest
Most of the tools are both Functional as well as regression testing tools.
Regression Testing Of GUI application:
It is difficult to perform GUI(Graphical User Interface) regression testing when GUI structure is modified. The test cases written on old GUI either becomes obsolete or need to reuse. Reusing the regression testing test cases means GUI test cases are modified according to new GUI. But this task becomes cumbersome if you have large set of GUI test cases.
It is difficult to perform GUI(Graphical User Interface) regression testing when GUI structure is modified. The test cases written on old GUI either becomes obsolete or need to reuse. Reusing the regression testing test cases means GUI test cases are modified according to new GUI. But this task becomes cumbersome if you have large set of GUI test cases.
Black Box Testing: Types and techniques of BBT
I have covered what is White box Testing in previous article. Here I will concentrate on Black box testing. BBT advantages, disadvantages and and How Black box testing is performed i.e the black box testing techniques.
Black box testing treats the system as a “black-box”, so it doesn’t explicitly use Knowledge of the internal structure or code. Or in other words the Test engineer need not know the internal working of the “Black box” or application.
Main focus in black box testing is on functionality of the system as a whole. The term ‘behavioral testing’ is also used for black box testing and white box testing is also sometimes called ‘structural testing’. Behavioral test design is slightly different from black-box test design because the use of internal knowledge isn’t strictly forbidden, but it’s still discouraged.
Each testing method has its own advantages and disadvantages. There are some bugs that cannot be found using only black box or only white box. Majority of the applicationa are tested by black box testing method. We need to cover majority of test cases so that most of the bugs will get discovered by blackbox testing.
Black box testing occurs throughout the software development and Testing life cycle i.e in Unit, Integration, System, Acceptance and regression testing stages.
Tools used for Black Box testing:
Black box testing tools are mainly record and playback tools. These tools are used for regression testing that to check whether new build has created any bug in previous working application functionality. These record and playback tools records test cases in the form of some scripts like TSL, VB script, Java script, Perl.
Black box testing tools are mainly record and playback tools. These tools are used for regression testing that to check whether new build has created any bug in previous working application functionality. These record and playback tools records test cases in the form of some scripts like TSL, VB script, Java script, Perl.
Advantages of Black Box Testing
- Tester can be non-technical.
- Used to verify contradictions in actual system and the specifications.
- Test cases can be designed as soon as the functional specifications are complete
- Tester can be non-technical.
- Used to verify contradictions in actual system and the specifications.
- Test cases can be designed as soon as the functional specifications are complete
Disadvantages of Black Box Testing
- The test inputs needs to be from large sample space.
- It is difficult to identify all possible inputs in limited testing time. So writing test cases is slow and difficult
- Chances of having unidentified paths during this testing
- The test inputs needs to be from large sample space.
- It is difficult to identify all possible inputs in limited testing time. So writing test cases is slow and difficult
- Chances of having unidentified paths during this testing
Methods of Black box Testing:
Graph Based Testing Methods:
Each and every application is build up of some objects. All such objects are identified and graph is prepared. From this object graph each object relationship is identified and test cases written accordingly to discover the errors.
Each and every application is build up of some objects. All such objects are identified and graph is prepared. From this object graph each object relationship is identified and test cases written accordingly to discover the errors.
Error Guessing:
This is purely based on previous experience and judgment of tester. Error Guessing is the art of guessing where errors can be hidden. For this technique there are no specific tools, writing the test cases that cover all the application paths.
This is purely based on previous experience and judgment of tester. Error Guessing is the art of guessing where errors can be hidden. For this technique there are no specific tools, writing the test cases that cover all the application paths.
Boundary Value Analysis:
Many systems have tendency to fail on boundary. So testing boundry values of application is important. Boundary Value Analysis (BVA) is a test Functional Testing technique where the extreme boundary values are chosen. Boundary values include maximum, minimum, just inside/outside boundaries, typical values, and error values.
Many systems have tendency to fail on boundary. So testing boundry values of application is important. Boundary Value Analysis (BVA) is a test Functional Testing technique where the extreme boundary values are chosen. Boundary values include maximum, minimum, just inside/outside boundaries, typical values, and error values.
Extends equivalence partitioning
Test both sides of each boundary
Look at output boundaries for test cases too
Test min, min-1, max, max+1, typical values
Test both sides of each boundary
Look at output boundaries for test cases too
Test min, min-1, max, max+1, typical values
BVA techniques:
1. Number of variables
For n variables: BVA yields 4n + 1 test cases.
2. Kinds of ranges
Generalizing ranges depends on the nature or type of variables
Advantages of Boundary Value Analysis
1. Robustness Testing – Boundary Value Analysis plus values that go beyond the limits
2. Min – 1, Min, Min +1, Nom, Max -1, Max, Max +1
3. Forces attention to exception handling
1. Number of variables
For n variables: BVA yields 4n + 1 test cases.
2. Kinds of ranges
Generalizing ranges depends on the nature or type of variables
Advantages of Boundary Value Analysis
1. Robustness Testing – Boundary Value Analysis plus values that go beyond the limits
2. Min – 1, Min, Min +1, Nom, Max -1, Max, Max +1
3. Forces attention to exception handling
Limitations of Boundary Value Analysis
Boundary value testing is efficient only for variables of fixed values i.e boundary.
Boundary value testing is efficient only for variables of fixed values i.e boundary.
Equivalence Partitioning:
Equivalence partitioning is a black box testing method that divides the input domain of a program into classes of data from which test cases can be derived.
Equivalence partitioning is a black box testing method that divides the input domain of a program into classes of data from which test cases can be derived.
How is this partitioning performed while testing:
1. If an input condition specifies a range, one valid and one two invalid classes are defined.
2. If an input condition requires a specific value, one valid and two invalid equivalence classes are defined.
3. If an input condition specifies a member of a set, one valid and one invalid equivalence class is defined.
4. If an input condition is Boolean, one valid and one invalid class is defined.
1. If an input condition specifies a range, one valid and one two invalid classes are defined.
2. If an input condition requires a specific value, one valid and two invalid equivalence classes are defined.
3. If an input condition specifies a member of a set, one valid and one invalid equivalence class is defined.
4. If an input condition is Boolean, one valid and one invalid class is defined.
Comparison Testing:
Different independent versions of same software are used to compare to each other for testing in this method.
Different independent versions of same software are used to compare to each other for testing in this method.
Reference- http://www.softrel.org/stgb.html
No comments:
Post a Comment