On 2016.05.25 15:25, Chris Murphy wrote: > May 24 00:07:00 redwood.localnet kernel: Command line: > BOOT_IMAGE=/vmlinuz-4.4.9-200.fc22.x86_64 root=/dev/mapper/vg0-lv1 ro > rd.md.uuid=5a1848ce:cff64c1f:a622dc1f:e5f06d55 LANG=en_US.UTF-8 > SYSFONT=True rd.lvm.lv=vg0/lv0 rd.lvm.lv=vg0/lv1 KEYTABLE=us > rd.luks=0 > rd.dm=0 > > > There is an rd.md.uuid= parameter, so that's good. > > > > May 24 00:10:08 redwood.localnet dracut-initqueue[347]: Warning: > Could > not boot. > May 24 00:10:08 redwood.localnet systemd[1]: Received SIGRTMIN+20 > from > PID 349 (plymouthd). > May 24 00:10:08 redwood.localnet dracut-initqueue[347]: Warning: > /dev/mapper/vg0-lv1 does not exist > May 24 00:10:08 redwood.localnet dracut-initqueue[347]: Warning: > /dev/vg0/lv0 does not exist > May 24 00:10:08 redwood.localnet dracut-initqueue[347]: Warning: > /dev/vg0/lv1 does not exist > May 24 00:10:08 redwood.localnet systemd[1]: Starting Dracut > Emergency > Shell... > > OK so yeah, it's basically not finding rootfs. 29 seconds later... > > May 24 00:10:37 redwood.localnet kernel: md: md0 stopped. > May 24 00:10:37 redwood.localnet kernel: md: bind<sda1> > May 24 00:10:37 redwood.localnet kernel: md: bind<sdb1> > May 24 00:10:37 redwood.localnet kernel: md: raid1 personality > registered for level 1 > May 24 00:10:37 redwood.localnet kernel: md/raid1:md0: active with 2 > out of 2 mirrors > May 24 00:10:37 redwood.localnet kernel: md0: detected capacity > change > from 0 to 262078464 > May 24 00:10:37 redwood.localnet kernel: md: md1 stopped. > May 24 00:10:37 redwood.localnet kernel: md: bind<sda2> > May 24 00:10:37 redwood.localnet kernel: md: bind<sdb2> > May 24 00:10:37 redwood.localnet kernel: md/raid1:md1: active with 2 > out of 2 mirrors > May 24 00:10:37 redwood.localnet kernel: created bitmap (15 pages) > for > device md1 > May 24 00:10:37 redwood.localnet kernel: md1: bitmap initialized from > disk: read 1 pages, set 0 of 29804 bits > May 24 00:10:37 redwood.localnet kernel: md1: detected capacity > change > from 0 to 2000068632576 > May 24 00:10:54 redwood.localnet systemd[1]: Found device > /dev/mapper/vg0-lv1. > May 24 00:10:54 redwood.localnet systemd[1]: Starting File System > Check on /dev/mapper/vg0-lv1... > May 24 00:10:54 redwood.localnet systemd-fsck[2685]: /: clean, > 529693/1048576 files, 3263599/4194304 blocks > May 24 00:10:54 redwood.localnet systemd[1]: Started File System > Check > on /dev/mapper/vg0-lv1. > > > So now it's found, I'm gonna guess this is your manual attempt at > starting up the array? > > What do you get for cat /proc/mdstat at the dracut prompt before you > do anything else? And also the results from > > mdstat -E /dev/sda1 > mdstate -E /dev/sdb1 > blkid > > These can be redirected to fpaste or a file. If you have network > access this early it's easiest to redirect to fpaste but I'm not sure > if it's in the initramfs. > > >From your mdadm.conf, you have UUIDs that do not match what's on the > command line. > > cmdline: > rd.md.uuid=5a1848ce:cff64c1f:a622dc1f:e5f06d55 > > mdadm.conf > UUID=c45384af:aa8fc7ca:7eebaeb8:f0ca1523 > UUID=524079c6:521ca76f:126164a9:3232e116 > > Looks like the mdadm.conf is stale and should be recreated, and > recreate the initramfs. Thing is, the rd.md.uuid ought to be enough > for dracut to startup that array, the mdadm.conf is really just about > assigning a particular md array to a particular /dev/mdX assignment, > which by the way has changed recently so there might be confusion > going on there. > > You might just try renaming the current mdadm.conf, or commenting out > all of the lines, except for one: > > ARRAY /dev/md0 UUID=5a1848ce:cff64c1f:a622dc1f:e5f06d55 > > And after saving it, use dracut -f. > > Chris Murphy > Okay. I have figured this out after a lot of digging through Google. I finally found this page and a post by eileon (Comment: 10) https://bugzilla.redhat.com/show_bug.cgi?id=1085922 The problem turned on differences between a file: /etc/default/grub and the output of: dracut --print-cmdline As eileon stated what is given by the output above by dracut, needed to be reflected in the /etc/default/grub file before you did ran: grub2-mkconfig -o /boot/grub2/grub.cfg and then: dracut -f The problem I had was in the GRUB_CMDLINE_LINUX line in the /etc/default/grub file. The first rd.md.uuid was supposed to be the uuid for the Linux RAID member disks that make up the mdraid devices from which you want to boot. The second rd.md.uuid that pointed to the RAID member disks used for the root filesystems and LVM was missing! Correcting both rd.md.uuid entries to reflect the correct devices in that file and running dracut -f before rebooting resolved my problem. Thank you Chris for your assistance and pointing me in the right direction. Cheers! Bill -- users mailing list users@xxxxxxxxxxxxxxxxxxxxxxx To unsubscribe or change subscription options: https://lists.fedoraproject.org/admin/lists/users@xxxxxxxxxxxxxxxxxxxxxxx Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
