On Fri, Feb 01, 2019 at 03:07:27PM -0500, James Taylor wrote:
Perhaps I missed a comment early on in this discussion, but would it not be reasonable to let someone performing an update know during the initial phase update when it looks to the existing software to let them know their file system will not load afterward, and point to the remedy, should they need it? I am in total agreement with the comments previously that expressed concerned about someone updating a perfectly functional system and having it fail without explanation upon reboot.
I'm quite sure I haven't seen anyone in this discussion questioning that. Everyone seems to agree that we should do as much as possible to avoid a risk of someone's system breaking without a warning. In some parts of this thread, people are discussing which way would be best.
I have run into this enough times to feel comfortable with it being part of a standard update process.
That's what I have been warning people since the "rolling" distributions came into fashion. Not distinguishing between updates and upgrades seems cool for some time but one day you will come to the point where you need to do some big and incompatible change which is going to break things without manual intervention and cannot be divided into a series of small and safe steps. I'm not sure all users of "rolling" distributions realize the very idea of smooth and safe "rolling" is an illusion. In this case, there seems to be multiple ways to work around it and minimize the risk. But what if one day upstream decides to drop support for some of these systems completely? Michal Kubecek -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org