Currently, the Fedora Media Writer is not in great shape for Fedora 24. It has the following known issues:
Given that we are now in Beta Freeze, I think it's time to seriously consider punting this [https://fedoraproject.org/wiki/Changes/LUCasPrimaryDownloadable Change] to Fedora 25 and sticking with the existing image-writing mechanisms for Fedora 24.
I think it's a bit premature to punt this change to F25:
Ad 1) I don't think postponing would improve this in any way, building it completely from source would require a lot of releng resources which definitely overweight the benefits and nothing will change about it in the F25 cycle.
Ad 2,3,4) Several of the severe bugs discovered during the test have already been fixed and we're hoping to address the rest by the next week. Yes, it's beta freeze, but we're working with kparal on asking for freeze exceptions. Yes, the test day took place just two days ago, the wide test exposure could have been earlier, but that's the way it is now.
There are no good image-writing mechanisms now, at least not on Windows which this tool is targeting primarily. And the old Live USB Creator is in a terrible state. So there is no safe sticking with the good we have. I'm not saying we should release FMW at any cost, but just dismissing it one and half month before the final release when all parties are working hard to finally get it out of the door is IMHO not a good move.
Replying to [comment:2 eischmann]:
I think it's a bit premature to punt this change to F25: Ad 1) I don't think postponing would improve this in any way, building it completely from source would require a lot of releng resources which definitely overweight the benefits and nothing will change about it in the F25 cycle.
That may be true, but from what I have been able to glean, there's no plan in place yet for how to produce it with the pre-built code either.
I'm concerned that these aren't edge-case or easily-missed mistakes but the basic functionality of the tool not working post-Freeze. If this was Alpha, I'd be much more sympathetic, but blocking Beta over a non-functional tool seems unnecessary. I think this rewrite is good and necessary, but I haven't seen anything that makes me believe that this will be anything but a cause of delays in the Beta and Final releases (and the Fedora 24 schedule really has very little room for delay at this point, due to the already short Fedora 25 schedule).
I'm not necessarily saying that we dismiss it. I have no problems whatsoever if we offer it as an ''experimental'' (and therefore non-blocking) option for Fedora 24 and make it the official solution for Fedora 25 Alpha and onwards.
From my perspective, the current state is not optimal but the risk of releasing something subjectively ''worse'' (from an end-user's perspective) is pretty high. We're very late in the schedule and pushing ahead with making this the preferred installation tool runs a very strong likelihood of introducing release slips.
We have agreed with Jiri (eischmann) to postpone the [https://fedoraproject.org/wiki/Changes/LUCasPrimaryDownloadable Fedora Media Writer Change] to F25. The Change will go through FESCo review as F25 Change once it is ready for it.
OK, as I said above, I strongly recommend aiming for there to be usable packages. We won't ''promote'' it as the official method, but we should certainly ask anyone who is willing to try it and report back.
Sorry that we didn't quite make it.
Log in to comment on this ticket.