"Exploring DevOps Like a Tester" by Parveen Khan

We had a great online meetup with Parveen with lots of questions at the end. We ran out of time, so we’ve added them here!

Can a build/release pipeline become too bulky? Can there ever be too many left-shifted tests that are run as part of the pipeline? - David Clarke

Shouldn’t post pipeline be covered in the monitoring? - Brian McRoberts

Can you tell us about your testing tour? - Simon Prior

Was there any resistance from anyone in implementing the pipeline and how did you deal with it? - Ben Fellows


Saw the session coming up, socialized via twitter I think? Missed it due to time pressure. Assume an intro like https://www.meetup.com/TAQfull/events/266415550/ .

Keen to hear more.

The video recording has just gone live, so you can catch up here https://www.ministryoftesting.com/dojo/lessons/exploring-devops-like-a-tester-by-parveen-khan

@lgibbs also did another excellent set of sketch notes

1 Like

The pipeline definitely can become bulky when we keep adding all the test to the pipeline without retrospecting the goal of adding those tests or time taken to run those tests on the pipeline.
We as a team took step by step approach to add each test and monitor the failures and feedback loops. And it also depends on the size of changes being made, minimizing the size of the change is also very important.
Shifting left should help and empower the team with faster feedbacks, allowing testers to focus more on exploratory testing and helping the team to shift right.

I understand this question in two different contexts.

  1. If you are talking about the context of pre-deploy pipeline tests that I mentioned during my talk then my answer would be

    • As we were new to this concept we were experimenting and learning so we decided to have some
      tests on pre-deploy pipeline and some tests(static code analysis and unit tests) on the post-deploy
      pipeline(integration, API, consumer-driven contract tests, end to end test)
  2. If you are talking about post pipeline tests then yea there’s still lot more testing needs to be done like exploratory and monitoring.

I hope I didn’t make assumptions here for what you were trying to ask :slightly_smiling_face:

Thanks for this question Simon.
I started my testing tour after attending tesbash brighton 2019 Lisi Hocke’s talk where she shared her journey around Testing Tour. Here’s her talk recording - https://www.ministryoftesting.com/dojo/series/testbash-brighton-2019/lessons/cross-team-pair-testing-lessons-of-a-testing-traveler-elisabeth-hocke
I paired with different testers and developers to learn and share on different topics . Here’s more details of different sessions I had on my testing tour - https://www.parveenkhans.com/p/testingtour.html
and here’s Lisi’s blog about her testing tour - https://www.lisihocke.com/p/testing-tour.html

Hope that helps :slightly_smiling_face:

1 Like

Lucky enough there was no resistance in implementing the pipeline and adapting the new way of working from any of the team members because we all were interested in this. But there was more resistance for making testers involved in decision making, deciding on process,visibility of goal and collaboration.