Abuja Electricity Distribution Plc’s Post

A common misconception in QA is confusing tools with processes. Tools are just one part of a much larger picture. Remember, they are only as effective as the process that guides them. You can watch the full State of QA webinar here: https://lnkd.in/e9Js5-aU #SoftwareTesting #QualityAssurance #QAProcess

Creating some software is theoretically simple - all you need to do is: 1. have an idea - preferably in an area where there is no or little competition, why fight with the "big guys" - here the idea came a few years ago - let's make backup software - after all, it's such a niche on the market 😂 2. hire programmers who will translate the vision into working software 3. test 4. release to the market 5. count the $ you get in yourself And then comes the steamroller... 1. idea - it's great, but we hit a niche, open source virtualization is omitted by the "big guys" and is developing very dynamically 2. hire... it's a shame that the times of slavery are a thing of the past - a joke 😎 3. test - trivial, until the software has a few screens and few options. The problem begins when there are so many screens and options that making a nice, energetic online demo takes 30 minutes and it turns out that we are able to show maybe 10% of the options and possibilities. We can release a "test" version to the market - after all, the trendy CD/CI will defend us, some manual QA will test it, it works ... and the interested person takes our software in his hands, wants to touch it, see for themselves if it works as he saw on the video and from marketing... and ... he can't get through the installation, because... (anything). What could be better than a "stupid" machine, which always works the same, always does the same, always expects the same and is unable to react to something "new" - read error. Therefore, putting a few pieces of the puzzle together (#Jenkins, #Groovy, #Ansible, #Bash, #virtualization, and old #PXE, #NFS protocols) I made an auto tester - from downloading our software packages from the repository, through preparing a fully offline Linux image with our software preinstalled, building an ISO, then mounting this ISO to the server, running the installer, preconfiguration the software ... and finally verifying whether I can log in according to the information received and checking whether I can see the created backup destination in the UI. If the flow ends with SUCCESS, it means that in 95% of people who download the ISO it should end the same way - they will receive working backup software, and they can test it on their own. Thanks to full automation, we are certain that the software delivered to you in ISO form works. Of course, this is just the tip of the iceberg, there are many more automated tests that are already eating into the heart of our program, and there will be even more, which will translate into you receiving better and better tested and warmed-up software in a short time. As Paweł Mączka announced a few days ago - version 7.0.0 of Storware is coming ... winter also 😱 Finally my laziness won 😁 🛀