Hi, I have a 3 disk RAID 5 where one disk broke (/dev/sdb). /dev/sda1 has /boot, /dev/sda2
is a RAID partition, /dev/sdb1 is a RAID partition, /dev/sdc1 is a RAID partition. In the
process of moving disks around to recover two unfortunate things occurred. The first is
that my system stopped recognizing the second disk (the one that had no /boot on it) which
I have subsequently added back into the array, and the second is that my system in
emergency mode doesn't have cryptsetup for opening the luks container and will not
recognize my USB ports. How do I get the system to open my disk up as before?
Attempting to boot gives, via journalctl:
"systemd-udevd[434]: Process '/sbin/partx -d --nr 1-1024 /dev/sdc' failed
with exit code 1." and
"systemd-udevd[426]: Process '/sbin/partx -d --nr 1-1024 /dev/sdb' failed
with exit code 1." and
"systemd-udevd[434]: inotify_add_watch(8, /dev/sdc1, 10) failed: No such file or
directory" and
"systemd-udevd[426]: Process '/sbin/mdadm -I /dev/sdb' failed with exit code
2." and
"systemd-udevd[434]: inotify_add_watch(8, /dev/sdb1, 10) failed: No such file or
directory"
Apparently the boot sequence is pre-emptively deleting sdb and sdc partitions from /dev.
When I execute:
"partx -a --nr 1-1024 /dev/sdb" and "partx -a --nr 1-1024 /deb/sdc" it
re-creates their first partitions device nodes.
Then I edit mdadm.conf (which is apparently automatically written by anaconda each time I
try to boot) to read:
"DEVICE /dev/sda2 /dev/sdb1 /dev/sdc1
MAILADDR root
HOMEHOST <system>
ARRAY /dev/md127 level=5 devices=/dev/sda2,/dev/sdb1,/dev/sdc1 metadata=1.2
UUID=f6224251:0ba59f55:05d9cdf4:98e79eca"
so I can execute:
"mdadm --stop /dev/md127" to stop the array with 1 disk and
"mdadm --assemble /dev/md127" to start the array with all 3 disks
RAID is recognized but now I'm stuck. What next for opening luks and then lvm?
Thanks.
Show replies by date