Reason for changed software keeps on failing (reporting errors).

Sonal's picture

Hi,

 Software is changed by project team. The project team calls this as ‘minor enhancement’. However the changed software keeps on failing (reporting errors). What could be the reason?

What would be the best approach for gaining points in csqa exam? Fish bone or steps to steps to be written?

Points can be( from my point of view):-

1.Root cause  Analysis of defects reported by the customer against the Enhancement.

2. To check the evidence of Impact analysis done before making the changes .

3. Has the design specification and code walkthrough been done?

4. Who had develop and tested the enhancement( against the approved requirement) to trace the responsible and accountable person.

5.By answering all above points , reason can be easily find it out and shared .

 Pls suggest

Thanks

 

 

 

csqa's picture
The question is asking for

The question is asking for Reasons for failure. So you may have to make some assumptions or guesses.

Assumption: Software was working before the change was made without any errors 

Possible Reasons for Failure:

( With above assumption, Change has caused the software to fail )

  • Change was not inspected well, by doing Reviews or walkthroughs hence some of the defects were not detected.
  • The change introduced a new technology , which the current platform or environment does not support. Possibly a patch or something required which was overlooked.
  • Regression testing was not done. The change itself worked fine and it passed the test cases, however it broke some other portion of the software. This could have been caught by regression testing.
  • The change may have conflicted with some other change done to the system at the same time. Proper change management was not done by the team.
  • Skill gap - The developer did not code it well due to skill gap.
  • There were environment differences, between dev, test and production. Hence the code worked perfectly fine in dev and test, but failed in production.
  • The data used for testing was not real time data. Due to different data and volume of data in production caused the software to break.
  • Load Testing was not done on the software, hence it failed under load.

And the list goes on ... :)

Submitted by csqa on November 16, 2007 - 3:41pm.