• Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 707
  • Last Modified:

Test cases that are not so obvious

In my current assignment I need to identify test cases/scenarios that are out of the ordinary. I am testing a professional networking site liked linkedIN but one that is targeted to a specific community. It has features such as profile building, connections, watches, messaging, Network updates, Groups, career news, job posting etc that a social networking/professional networking site will have. We already have about 1500 test cases covering the major functionality of the application. But the cases were written by inexperienced testers and cover only the obvious stuff. Today my boss told me to create cases that are not very obvious and only an experienced hand would be able to point out. He pointed scenarios such as
1)When SMTP server is down
2)SPAM filter filters out the invitation mail sent by the system
3)What happens if the user clicks the invitation link after its validity runs out
4)browser session expiry

I have to create cases for these and many other scenarios that are not so obvious. I have been working in IT for a while but  my experience in identifying/creating test cases is limited. My company however expects me to perform like an expert tester.
These cases i would imagine are common to all applications and are not specific to any one domain or technology.
I need the following
a) Please share with me any such scenarios.
b) Is there a book i can refer to for such scenarios or is there any resource available online that points out such cases.
Thanks for reading
0
PearlJamFanatic
Asked:
PearlJamFanatic
1 Solution
 
cdboshCommented:
What you're looking for in general is sometimes called "negative testing" or "destructive testing", and can be similar to "failure mode and effects analysis" or FMEA, although FMEA is often applied to an operational environment, but it can be used as a structured way to consider the possible ways in which a piece of software might fail, and then you can devise a test to detect whether that failure resulted from the hypothetical cause.

Google the quoted terms above and you'll find a lot of information about them.  (Note that "destructive testing" will also turn up hits from product/materials testing, as this is where the term originates - just ignore those!)

Another form of more complex testing is actually called "scenario testing".  Most simplistic functional testing just looks at relatively atomic requirements and tests that single function in isolation.

Scenario testing is more complex because it attempts to truly model real-world behaviors.  You might also consider this "business simulation" testing, for that reason.

So, you may have a functional test on your professional social-networking site that says select a contact from your address book and send an invitation to join the site.

A scenario, though, might be to import a set of contacts into the address book, add some entries, delete some entries, modify some entries, then create a contact group, then delete some more entries, including some in the contact group, then send a mass invitation to the site using the contact group, then search and find one of the contacts already in the site's member database and make a separate "connection" request, then have that same contact accept the earlier email invitation, ...

This is probably a but of an absurd example, but that is the idea - it is a fairly complex series of events, some user actions, some system actions, some 3rd party actions - and see how they all play together.

However, because of the complexity and high-degree of data setup, scenario testing can be very expensive!  You must do some risk analysis and only spend this kind of effort on the riskiest and most critical/core functions.  This is not the kind of thing where you go for 100% coverage, as many do with simple requirements/functional testing.

As for a book with the "common" negative tests - there may be something like that for the examples you listed - especially since you're talking about web applications, which do have these common "web infrastructure" problems.  Note that some of these will be infrastructure configurations (e.g. session time outs, caching, server routing and failover) that will have to be run in the production environment, versus application problems (e.g. how it handles an expired invitation) that can be worked through in the test environment.

Unfortunately I don't have specific experience with a book to recommend, but if you search for "web testing book" I'm sure you'll get a lot of options - check some of the reviews to pick the ones that meet your interest.  Here's one that might work:

http://search.barnesandnoble.com/booksearch/isbnInquiry.asp?r=1&ISBN=9780321369444&ourl=How-to-Break-Web-Software%2FMike-Andrews

It has a strong web security angle, but that will probably be important to the integrity of your site, especially since you will be holding the personal information for professionals.
0
 
PearlJamFanaticAuthor Commented:
Thank for the reply cdbosh, it is much appreciated.
0

Featured Post

Technology Partners: We Want Your Opinion!

We value your feedback.

Take our survey and automatically be enter to win anyone of the following:
Yeti Cooler, Amazon eGift Card, and Movie eGift Card!

Tackle projects and never again get stuck behind a technical roadblock.
Join Now