Balu

• Jul 10th, 2006

 

In case of Sanity testing, no need to follow test cases. Here we will check the main areas of functionalty, and smoke also very similar to this.

Thanks

raj

• Jul 10th, 2006

 

It is not at all necessary to write test cases for sanity and smoke testing. This depends from org to org and their policies. There are few test cases which just check the flow/functionality of application which are run for sanity/smoke testing.

raj

• Jul 10th, 2006

 

Alpha testing or Acceptance testing- Here the tester tests the functionality of the application. He tests for all critical flows, dependencies, most visible part etc for the application.

shyam

• Jul 10th, 2006

 

No It is not nessary to follow any test case in this kind of test

anjan

• Oct 25th, 2006

 

please give me the brief disscusion on scenario

Nikharchugh16

 Profile Answers by Nikharchugh16 

• Oct 26th, 2006

 

HI All,

First of all, must tell you that sanity Testing and smoke testing is used interchangeably in some org and some not. Actually sanity testing is done ideally at the begining always. This testing is done just to ensure that application whioch we will be testing is present before we begin out detailed system integration testing. If the application is not present, we will not be able to test. e.g., in order to do the functionality testing, usability testing of a LIFT, you need to first ensure that the lift is actually there in order to test.

Smoke testing is basically done when we are moving the project/product in the PROD environment, just to check that the major functionalities of an application works as intended by the user.

WE can write just one or two sanity test cases, if required (or a norm exists in the org), for each module of an application. Smoke test cases are generally written whil performing the smoke testing and it';s test case can be derived from the system integration test cases (functionality test cases) which we had written earlier.

To answer your other part of the question, Alpha Testing is that phase of the testing which is done before the product is shifted to PROD and generally done by the in house developers or a separate testing team at the developers site (customer location) just to be very much sure on the number of bugs in the application before it is finally delivered.

Thanks and Regards,

Nikhar

mithr17

 Profile Answers by mithr17 

• Nov 22nd, 2011

 

I disagree with all the above answer that said test cases are unnecessary.

"ravikaanthe" clearly does not understand the difference between smoke and sanity testing nor do any of the above people who answered the question.

It is true that words - smoke and sanity - are interchangeably used but there are subtle differences.
If you check search for sanity and smoke testing questions in this site, you will know the difference right away.

Smoke testing:
- Performed on all builds except during regression testing.
- Testcases are never exclusively written for "smoke" testing. Rather, mission-critical test cases from the all the test cases QA writes to positively and negatively test the requirements are selected and grouped as "smoke test suite."
Just to avoid another question: Test suite in simple terms is group of individual test cases.
- When a build arrives, smoke tests are performed to check if the application is stable enough to withstand further tests.

Sanity testing:
- Performed excursively on regression builds. After critical and major bugs have been fixed, regression testing phase commences and the builds are termed as 'regression' builds. Its just jargon to bring everyone on the same page!
- Again, test cases from smoke test suite and rest of the test suite are selected to be ran as 'sanity' tests. Therefore, "Sanity test suite" is collection of test cases that are mission critical to the affected feature/modules/core functionality.

QA always writes test cases for all requirements and smoke and sanity test suites consist of test cases selected from the main test suite.

Alpha testing is application to any application. Some users on this site classify applications as services and product based applications. People say that its a services application if there is a client and you as a QA work for services company. Product based application is if you work as an in-house QA who is testing the in-house application. This is a misunderstanding that most of the Indian IT justa has.

Alpha testing is nothing but user acceptance testing. Does QA perform user acceptance testing? If you are nodding, then isn't user acceptance testing a redundant effort of QA? UAT is performed by end-user who will end up using the application. Therefore, QA role is to support the UAT tester (the business/end users). In some companies BA, with the help of QA team, writes UAT scenarios and publishes it to the UAT testers (the business/end users). In some companies, the responsibility rests solely with QA team. Either way, QA does have a support role to play. QA must support UAT testers in every way possible - answer questions, test data, UAT scripts/scenarios, etc.