This talk was recorded at NDC Techtown in Kongsberg, Norway. #ndctechtown  #ndcconferences  #cplusplus  #testing  #softwaredeveloper    

Attend the next NDC conference near you: 
https://ndcconferences.com
https://ndctechtown.com/

Subscribe to our YouTube channel and learn every day:   
/@NDC 


The failure policy for automatic end-to-end tests differs from that of unit tests. Failing tests, or even flaky tests, are not necessarily a problem. The ambition or dream of a perfect night's test run may in fact be harmful. Having that goal can make you consciously or unconsciously hesitate to write tests that target weaknesses in the system under test. Tests that fail regularly provide an always up to date status of the problems that still need fixing. Intermittent end-to-end test failures tracked over time provide valuable input to development teams to determine the priority for fixing known issues.

However, dealing with broken tests is a challenge. So this talk will focus on a wide range of tried and tested strategies, tools and practices that ensure the meaningful signal from your test results doesn't get lost amid the noise of test failures. This talk will also cover a variety of effective recovery strategies to ensure you still run as many tests as possible on each build--in spite of failures. It will also cover concrete strategies to reduce the number of false positives which quickly drain the value out of your end-to-end testing investment.