#6739 F26 Alpha image paths and CHECKSUM files changed partially from underscore to dash
Closed: Fixed 5 years ago Opened 7 years ago by robyduck.

I am not sure why you changed the paths again, but there is really no consistency in it. We have so many images and checksum files, websites cannot check them all at every release, just because someone likes the dash more, and another one wants to have the underscore.
What I mean is the change from 26_Alpha to 26-Alpha, not in all images, and not in all checksum files. For example all alternative images have the dash (ppc, aarch...), and also the netinstall images, even for workstation. Cloud is half half and that makes it almost impossible for us to make a good work.

Time ago I agreed with @ausil on having an underscore for all image paths and checksums. That's then how we set up our scripts and variables, and it worked just great!

Now, as I had to fix them all manually, I looked into your pungi config files and found this:
https://pagure.io/pungi-fedora/blob/f26/f/fedora-alpha.conf#_228
vs.
https://pagure.io/pungi-fedora/blob/f25/f/fedora-beta.conf#_231

You can understand, this makes it for us very hard. Please tell us what we should do and then be consistent in your configs. Don't forget there are other teams relying on changes like that, and you can potentially break many things in Fedora world.

Apologies if it comes over a bit aggressive, but I lost a lot of time for just a dash in an image path :(


@mohanboddu I dug into this - the commit where fedora-alpha.conf was added simply copies fedora-final.conf from F25/fedora-final.conf to fedora-alpha.conf

It looks like F25/fedora-final.conf was changed to a dash from the underscore in beta.

@robyduck - do you recall whether F25's final compose for GA broke the site scripts in the same way as this F26 Alpha broke it?

@kellin AFAICT nothing broke our scripts and variables for the last release cycles. I recall I spoke to @ausil during F22 or F23 or so about the changes (yes there were already changes from dash to underscore) and he told me releng will keep the underscore for the next releases.
I am used to ask releng folks (normally Dennis) before each release if there are significant path modifications, but for F24 and F25 all went fine.
That's also the reason why I was upset this time.

Thank you for looking into this. Please let me know what are you going to do for Beta, as the scripts and variables now are mixed up (for prerelease pages).

Metadata Update from @ausil:
- Issue tagged with: meeting

7 years ago

the issue is simply that because of how the data is pieces together it has to be a _ for Alpha and Beta composes and has to be a - for final composes

Discussed this today in our Fedora Release engineering meeting.

Going to update the compose SOP with this information.

Metadata Update from @kellin:
- Issue assigned to kellin

6 years ago

@syeghiay this is fixed in F26 through F28.

Our process here is slightly fragile, opened taiga card 1012 so we can work toward a more durable solution that has less potential for human error.

Metadata Update from @kellin:
- Issue untagged with: meeting
- Issue close_status updated to: Fixed
- Issue status updated to: Closed (was: Open)

5 years ago

Login to comment on this ticket.

Metadata