Testertools Forum

1. johndaerson
   Member

   Reputation:63

    

   Our product is planned to release at the end of May, so in these days, we are doing regression testing on a stable build. The issue is, we just finished a full regression testing (It took the entire team 2 weeks to finish the task) on RC2 and we were told to perform another round full regression testing on RC3(another build). I feel that this is no need, because there is no new features check-in and only bug fixing. I am wondering what is an efficient way to do testing at the end of a release. What's your opinion? How do you do in your team? Is there any better way? How does your development team control code change at the end of the release?
   Please suggest, thanks.

   Posted 1 year ago #

2. Cyril
   Member

   Reputation:372

    

   It's easier (and probably safer) to simply say "test it all" each time, than to think it through.

   We analyze every Release Candidate build, and make our decision based on our findings. Sometimes it is necessary to execute a full round of regression tests. Sometimes it is not.

   Your decision should be based on your understanding of the risks involved, the history of recent Release Candidate builds, the confidence that the most recent build includes "only bug fixing", the criticality of the schedule, the time available, etc, etc

   Posted 1 year ago #

3. Cyril
   Member

   Reputation:372

    

     Quote:
       How does your development team control code change at the end of the release?

   One we have a Release Candidate, ALL code changes must be in response to a show-stopper bug, and must be approved prior to check in. No other changes are permitted.

   Posted 1 year ago #

4. johndaerson
   Member

   Reputation:63

    

   I am new to this area, but from feedback above, it seems that I need to learn how to plan for a regression testing by considering the risks that we may take.

   Posted 1 year ago #

RSS feed for this topic

Reply

You must log in to post.