ubuntu 16.04 18.04 ximport raid-lvm

BOOT_DEGRADED has no effect when using LVM

It seems that using LVM over RAID will always try to boot, no matter how BOOT_DEGRADED is configured. Removing a disk from the running system will put the array in degraded state, and a reboot at this point will work fine, even if this is somehow risky to boot on a degraded array.

Disk cold-remove

Removing the disk when the machine is not powered will prevent boot, at least on 18.04, dumping to the shell in the initramfs.

Array inspection (with mdadm --detail /dev/mdX) will show the degraded array, as expected.

Rebooting at this point will dump to the initramfs shell again. For forcing the boot on the degraded array: run mdadm --readwrite /dev/mdX then reboot. To fix the array, plug a new disk, add it with mdadm --manage /dev/mdX --add /dev/sdXy, then wait the sync to finish, then reboot.

A new disk may require a partition table, and this partition table can be copied from another active member of the array: dd if=/dev/used_disk of=/dev/new_disk count=4096

~~~

Question, remark, bug? Don't hesitate to contact me or report a bug.