Re: Prioritizing different possible optimization work for composes

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Mon, Oct 2, 2017 at 7:57 PM, Matthew Miller <mattdm@xxxxxxxxxxxxxxxxx> wrote:
Hey, here's a question: if you had a choice to improve how the compose
process goes leading up to releases (beta or final), as a Fedora QA
person would you prefer:

1. Vastly increased speed of compose so we can do more than one per
   day, or

2. Making composes self-service, so QA can initiate them with a button
   (possibly in the blockerbugs app)?

Which of these would be more helpful? I mean, ideally, we have both,
but which would you rather worked on first?

If the compose speed also affects the regular updates+updates-testing compose speed, I'd say 1) is more useful in general. Since it's me and Peter who request the composes now (or Adam when he's not on PTO), the timezones work reasonably well together with releng timezones. We usually request composes in the afternoon, after some testing, which works well for us.

One of the issues is that the process is not fully automated. Once the compose finishes, a manual work is still required to copy that compose to stage/ area. I'm not sure if this would be covered in 2). Also, would 2) include the ability to abort the current compose and re-request it again?

Overall, it seems to me that a few hours are saved in each scenario, so 1) has larger benefits overall. Now, if there was another releng person permitted to create RCs in a different timezone, that would also improve things quite a bit (until we get 2)) :-)

Thanks for pushing this,
Kamil


_______________________________________________
test mailing list -- test@xxxxxxxxxxxxxxxxxxxxxxx
To unsubscribe send an email to test-leave@xxxxxxxxxxxxxxxxxxxxxxx

[Index of Archives]     [Fedora Desktop]     [Fedora SELinux]     [Photo Sharing]     [Yosemite Forum]     [KDE Users]

  Powered by Linux