In F9, we used /dev/mapper/luks-<devicename> mappings in anaconda, /etc/fstab, and /etc/crypttab. In F10, we use /dev/mapper/luks-<uuid>. Obviously, this can cause problems on upgrade for anything referenced in /etc/fstab and /etc/crypttab, as things won't initialize or mount properly if they don't find the /dev/mapper/XXXX device they're looking for, and we've had bugs reported on this. See bug 468955 for one example. Yesterday, I started to investigate what would be required to migrate fstab and crypttab on upgrade... only to find out we already do so. Somewhere between selecting a partition to upgrade (in the GUI, not via commandline/ preupgrade) and the 'upgrade bootloader' screen, anaconda writes a completely new fstab and crypttab file. So, the questions are: - Is this intentional? Are we actually falling down 'install' codepaths? - Do we risk blowing away other user changes by doing this? - If not, is this all we need? Bill _______________________________________________ Anaconda-devel-list mailing list Anaconda-devel-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/anaconda-devel-list
