(In reply to Jiri Srain from comment #12) > (In reply to Fabian Vogt from comment #11) > > (In reply to Jiri Srain from comment #10) > > > (In reply to Fabian Vogt from comment #7) > > > > (In reply to Jiri Srain from comment #5) > > > > > YaST should IMO show an error if snapper could not be run from whatever > > > > > reason. > > > > > > > > > > OTOH: There is no need to make a hard dependency on snapper - not all > > > > > systems want snapshots enabled. > > > > > > > > I agree - but a hard dependency in yast2-installation should be fine as > > > > normally yast2-installation isn't installed. > > > > > > Please, no hard dependency. YaST does not require any file system specific > > > tools because the respective filesystem can be used. > > > > Fair enough, but then YaST should not allow those filesystems to be used. > > If it breaks hard if a tool is not used, it's a hard dependency. > > I believe that this has never been handled properly for live installer; for > standard installation YaST just automatically selects the needed packages > (which is not possible with the live image). Well, the only difference is that on the live media a full openSUSE is used instead of installation-images. > > > (In reply to Fabian Vogt from comment #6) > > > > > OTOH, yast *does* show an error already. So for me this looks fine. Just the KDE Live iso needs fixing. > > > > > > > > It's not fine. > > > > > > > > The error is *after* the installation. > > > > > > What does "after installation" mean? > > > > If you look at the linked openQA screenshot, the installation is at 99%, > > which means except for the snapshot creation it's complete. The system would > > probably boot fine. > > OK, so still within the installation :-) My point was: some weeks ago, there > were too many places when on command execution failure YaST just crashed; > compared to that, current behavior of this particular step is correct - > error reported. The error is not reported. The only message you get is that the snapper config is missing, which is highly misleading as it should've been created at the very beginning of the installation, just after the mkfs.btrfs. There was no handling of that error at all. > Avoiding the otherwise correctly handled situation is a different topic - > and this is what should be addressed. It's not handled correctly in any way.