On Thu, 2016-02-25 at 12:22 -0500, David Lehman wrote: > On Thu, 2016-02-25 at 18:14 +0100, Garrett LeSage wrote: > > > > On 25/02/16 16:06, David Lehman wrote: > > > > > > Right. Just about any time you make an assumption to > > > significantly > > > simplify storage you're going to be wrong. Sad but true. > > Yeah, it's not possible to cover every case all the time and do so > > in > > an > > easily understandable manner for most things... but especially > > something > > as complicated as storage. > > > > > > With regard to Workstation partitioning, however, I'm confident > > that > > we > > can make life nicer for at least 80% - 90% of the people using > > Anaconda. > > That's one of the goals of this effort, at least. > I can appreciate that, provided it retains some basic flexibility. > > > > > > > For the other 10 - 20% of people installing Workstation who really > > do > > want to specify everything, they will still have a > > similar/familiar > > advanced interface to do so. The difference is that we won't drive > > everyone into the advanced partitioner (and force them to even > > understand partitioning or complicate storage things they *could* > > possibly do). > People seem to resist this, but it would be really great IMO if we > could stop calling it "partitioning" and start calling it "storage > configuration" since partitioning is only a fraction of what goes on. > > > > > > > > > There are three main cases for Workstation installs: > > 1. Empty disk(s). It's obvious what to do here... Have a > > predefined > > recommended partition layout and use that. Advanced partitioning > > spoke > > can, of course, still be selected from the Hub screen for fine- > > tuning > > or > > completely different layouts. > > 2. Disk(s) with data, with more than enough room for Fedora. Same > > as > > #1, > > except existing data will not be wiped out. > We did that in F18 or F19 and several users (including me) didn't > like > that you don't even have the option of making more space in this > case. Also I wonder how frequent this case would actually be - how would you actually get a disk that is partially used, but has significant unallocated space ? Completely empty disk I can imagine (fresh from the factory) as well as completely allocated one (pre-installed Windows, some Linux distro install, etc.) but how would a normal user get a disk that is not either empty or fully allocated to something ? I can only think of people creating such a layout themselves to use/try multiple Linux distros or other operating systems. I'm not sure if there would be very many of those & if they would need a simplified partitioning screen given their apparent ability to create custom storage layouts. > > Also, thanks for working on this. > > David > > > > > > 3. Disk(s) with data, but not enough room. This is where the > > simple > > partitioning screen shows up. There's a path through the advanced > > paritioning tool as an action on this screen (and you can also get > > back > > from the Hub screen). > > > > So: 1 & 2 (enough space due to empty disk or pre-partitioned disk) > > will > > be pre-configured by default. 3 will show the resize screen (which > > has > > options to go to the advanced installer or erase and use the > > entire > > disk, in addition to resizing). > > > > It does get a little more complicated when there are multiple (non- > > USB / > > non-SDcard) storage devices on the system; I have some ideas and > > will > > work on mockups to handle that soon. > > > > > > Partitioning on a Server install is a different beast, of course. > > We're > > not there yet; the first step is mainly simplifying Workstation. > > I'm > > not > > sure how much simplification (if any) can happen on a Server > > install. > > > > > > Cheers, > > Garrett > > > > _______________________________________________ > > Anaconda-devel-list mailing list > > Anaconda-devel-list@xxxxxxxxxx > > https://www.redhat.com/mailman/listinfo/anaconda-devel-list > _______________________________________________ > Anaconda-devel-list mailing list > Anaconda-devel-list@xxxxxxxxxx > https://www.redhat.com/mailman/listinfo/anaconda-devel-list _______________________________________________ Anaconda-devel-list mailing list Anaconda-devel-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/anaconda-devel-list
