On Dec 6, 2013, at 10:31 AM, Clyde E. Kunkel <clydekunkel7...@verizon.net> wrote: > > https://bugzilla.redhat.com/show_bug.cgi?id=1038885
Closed the first time because it's not an F20 bug. As mentioned in bug 864198 it's intended behavior. Also closed because the summary about it not installing the boot loader on btrfs doesn't make sense because that's not the problem nor how it would work if it did work, and the description has nothing to do with the actual problem or getting it fixed. I'd always intended to create a clearly described and explicit rawhide RFE tracking bug for the issue, and it was impossible to clean up and change your bug into that without requiring the reader to read 7 comments that have nothing to do with actually progressing the real problems. So that's why I closed it a 2nd time, after creating the tracker bug, for which there will eventually be other bugs it depends on, not just the grubby bug. There are issues with grub2 and os-prober that also need to be addressed for certain use cases. The rationale for the change is thoroughly discussed in several blocker reviews. It's not some arbitrary change, it was made based on several release criteria being violated. Yes it would have been better to fix that old grubby bug that was also an F20 blocker for a month. But it's worse to throw razor blades at hapless users, excusing it with the suggestion they don't really matter when those with secret decoder ring hacks to work around the problem can easily do so. You can still use kickstart to make the layout as you wish. The real issue is that grub-mkconfig isn't confused about /boot being located on btrfs subvolumes. Why? It seems a possible answer for grubby's confusion is located in grub. Chris Murphy -- test mailing list test@lists.fedoraproject.org To unsubscribe: https://admin.fedoraproject.org/mailman/listinfo/test
