http://bugzilla.opensuse.org/show_bug.cgi?id=932735 --- Comment #2 from Clifford Kennedy --- What is the estimated time frame for dealing with this issue? -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=932735 --- Comment #4 from Neil Brown --- This looks like bug 832501 I sent patches upstream to fix the issue a while back. http://comments.gmane.org/gmane.linux.kernel.initramfs/4087 No useful response.... maybe I should nag. PAtch 3/4 is the most important one. -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=932735 Thomas Renninger changed: What |Removed |Added ---------------------------------------------------------------------------- Flags|needinfo?(nfbrown@suse.com) | --- Comment #6 from Thomas Renninger --- Wait, it should not matter at all whether: /etc/crypttab or /usr/lib/dracut/modules.d/90crypt/block_uuid.map when doing: set_systemd_timeout_for_dev $_dev so I push your mainline version into 13.2 and SLE12 as it is. Please scream out load if this is nothing for maintenance. -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=932735 --- Comment #9 from Clifford Kennedy --- OK -- the status now shows "RESOLVED FIXED". What do I need to do now? Cliff Kennedy -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=932735 --- Comment #11 from Clifford Kennedy --- Here's a followup on my previous test. I discovered that the system had been left with a single disk raid1 configuration with sda being the only disk in the array. Therefore, the dracut update wasn't installed on sdb. I re-added and rebuilt sdb and my test results are outlined below. Unless I missed a step or somehow installed the update incorrectly (see above) there are still serious fault-tolerance issues when booting from a degraded raid1 array. The following are my steps and the results: o Changed md0 back to a normal raid1 containing both sda & sdb o readded sdb to the array and rebuilt. mdadm /dev/md0 --add /dev/sdb2; mdadm /dev/md0 --grow --raid devices=2 o rebooted as a normal 2 disk raid1 system o Tested Booting From sda with sdb removed: o shut down o disconnected sdb o reboot failed o Came to "A start job is running form dev-md0.device ( n / no limit" and sat there for 10+ minutes o Powered down. o reconnected sdb o reboot successful o Tested Booting From sda with sdb removed and md0 configured as a single disk raid1 using sda only: o removed sdb from the array o mdadm /dev/md0 --fail /dev/sdb2; mdadm /dev/md0 --remove /dev/sdb2; mdadm /dev/md0 --grow --raid devices=1 --force o shut down o disconnected sdb o reboot successful o Tested Booting from sdb with sda removed o shut down o disconnected sda o reboot failed o sat on "Reached target Basic System" for a couple minutes o sat on "Found device /dev/md0" for 5+ minutes. o landed in dracut. o Powered down o Reconnected sda o reboot successful o Tested Booting From sdb with sda removed and md0 configured as a single disk raid1 using sdb only: o removed sda from the array o mdadm /dev/md0 --fail /dev/sda2; mdadm /dev/md0 --remove /dev/sda2; mdadm /dev/md0 --grow --raid devices=1 --force o shut down o disconnected sda o reboot failed o sat on "Reached target Basic System" for a couple minutes o sat on "Found device /dev/md0" for 5+ minutes. o landed in dracut. o powered down o Reconnected sda o reboot (from sdb) failed o Came to "A start job is running form dev-md0.device ( n / no limit" and sat there for 10+ minutes o powered down o rebooted (from previously failed and removed sda) o Came to "A start job is running form dev-md0.device ( n / no limit" and sat there for 10+ minutes o powered down o At this point the system will no longer boot. -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=932735 http://bugzilla.opensuse.org/show_bug.cgi?id=932735#c15 --- Comment #15 from Clifford Kennedy --- Neil, OK. I'm finally through with my build obligations and will have the information you requested later this week. I have to do a re-install since my original work was left in an unbootable state. Thanks, Cliff Kennedy -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=932735 http://bugzilla.opensuse.org/show_bug.cgi?id=932735#c20 --- Comment #20 from Clifford Kennedy --- Neil, I created a mirrored swap as you suggested and the system booted successful from both disks independently. I did observed some oddities in the mdadm behavior but didn't have the time to fully test and document these issues. I hope to do so by the end of the week. Cliff -- You are receiving this mail because: You are on the CC list for the bug.