Just log it???
While doing testing I often found some bugs which gives me some clue about why it occurred in the first place. Being a tester do we just need to find an issue/bug and report it or should not we do some initial investigation on the bug (homework) before logging it. As per my understanding bugs are a result of misunderstood functionality, edge scenarios which was not thought of during development, improper data handling and finally the border issue (integration point with some other story).
So being a tester we should not only find bugs but rather give inputs in such a way that it helps the developer figure out the fix in a faster way. Implementation of logging in the application is one such attempt where we can guide the development team for more informative logging. Exceptions should not be suppressed while logging also it should not be overloaded. In case of Data handling issues, QAs should be proactive enough to put some effort of playing around with more varied set of data so as to find out if there is any pattern in the data handling.
The role of tester should not limit to just finding a bug but also to help in getting it fixed faster(directly or indirectly :) ).
In the ideal world, QA would be writing a lot of acceptance tests and would write a new test that expose each bug that they report.ReplyDelete
Thus the repro steps for a bug would be:
1) Run this acceptance test
2) Observe that it fails
I agree with you Bill ! Once we log the bug, typical scenario which happens is dev will try reproducing it on their box (following the steps) and then go thru the code keeping some debug point.ReplyDelete
But if we log the bugs with few more details like "logs", some more sampling data etc it will actually help developer find the issue very quickly.
We should not only just find bug but also help team fix it faster.