On 19 February 2018 at 16:39, Per Jessen
Richard Brown wrote:
And yet, creating a 3rd flavour of image, would need a significant amount of build time/power, and significant amount of testing required.
Why should it require more testing? I imagine a larger image would be <dvd-image> + <more packages>. Which is no different to <dvd-image> + <more packages from the net>
Maybe that's an alternative - for a larger image, we simply take the DVD image, and add packages in a separate partition to be used as a local repo. The installer should be able to pick that up without too much work.
Just because something works fine when installed from Installation Media #1 doesn't mean it will work fine when installed from Installation Media #2 That is why we have test coverage like https://openqa.opensuse.org/tests/overview?distri=kubic&distri=opensuse&version=Tumbleweed&build=20180218&groupid=1 and https://openqa.opensuse.org/tests/overview?distri=opensuse&version=15.0&build=131.1&groupid=50 which run a significant number of tests against BOTH the DVD media and the NET media currently and quite often find failures on one, but not the other If you're talking about adding a 3rd Media, then I would expect our openQA coverage alone would need to increase by 33% to account for that. Even if that coverage is 100% duplicated from the tests/needles/etc written for other media, that's a lot of hardware and/or machine time, as well as regular (daily) human review/triage. We shouldn't go deciding for something like this and just throw it over the wall as untested nonsense - that wouldn't be the openSUSE way. But doing things properly has real costs, and they should be considered. -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org