![]() If all three matches, this probably also means that the partition that's mounted is the second one, the first partition is not mounted but used as a boot partition by UEFI subsystem. Multiple UUID's: ~]# blkid | grep 999C-7B59 Multiple LABEL's: ~]# blkid | grep EFI_SYSPART System has multiple partitions with the same LABEL or UUID LABEL=EFI_SYSPART /boot/efi vfat defaults 0 1 File /etc/fstab contains /boot/efi partition which is mounted by LABEL or UUID, examples as follows: The server has soft-raid 1 with /boot/efi partition and UEFI subsystem is used for boot.Ģ. After the conversion to a Cloudlinux issue stays the same.ġ. We have found that's there's an issue on some Centos 7/8 installations with different hosting companies that prevents updating correct /boot/efi/EFI/centos/grub.cfg file and then boot into a newer kernel. If you are running DELL server, you probably have THIS issue. But when it gets down the part part it just goes ass up.Note. The kickstart file does run the commands in the %pre script. Then it tells me that sdc3 given in part command does not exist. Raid swap -level=RAID1 -device=md3 -fstype=swap raid.03 raid.06īut if I add the commands into a %pre heading in the kickstart file: Raid / -level=RAID1 -device=md1 -fstype=ext4 raid.02 raid.05 Raid /boot -level=RAID1 -device=md0 -fstype=ext4 raid.01 raid.04 Part /boot/efi2 -size 250 -asprimary -onpart=sdc1 -fsoptions="umask=0077,shortname=winnt" -fstype="vfat" -noformat Part /boot/efi -size 250 -asprimary -onpart=sdb1 -fsoptions="umask=0077,shortname=winnt" -fstype="efi" Is there a way to force CentOS to do a partitioning scheme so that it all lines up?īootloader -append=" crashkernel=auto" -location=mbr -boot-drive=sdbĬlearpart -none -initlabel -drives=sdb,sdc While technically this partitioning scheme will work for what I want to do, it's just not really desirable because the partitions don't line up across the different disks. dev/sdc4 - 4GiB RAID1 Mirror for Swap Space dev/sdc2 - 500MiB RAID1 Mirror for /boot dev/sdb4 - 4GiB RAID1 Mirror for Swap Space dev/sdb3 - 500MiB RAID1 Mirror for /boot This way the partitioning scheme for /dev/sdb and /dev/sdc are identical.īut instead the CentOS Install Manual Partitioning wants to do: dev/sdc4 - 4GiB RAID1 mirror for Swap Space dev/sdc2 - 500MiB RAID1 mirror for /boot ![]() dev/sdb4 - 4GiB RAID1 mirror for Swap Space dev/sdb2 - 500MiB RAID1 mirror for /boot What I'd like to do, in a software RAID1 system: And then it wants to take control of the partition order for the remaining partitions.īut I cannot for the life of me figure out how to make just an empty /dev/sdc1 partition, so that I can deal with that after the install. The CentOS installer wants to make /boot/efi on /dev/sdb1 (/dev/sda is my USB install drive). This can be some what remedied by keeping a copy of /boot/efi some where on the second (or 3rd, or 4th.) disk.īut the problem I'm running into, the CentOS Install Manual Partitioning won't allow granular control of partition order. What happens if that drive fails and has to be replaced? The system won't boot because there's no alternative ESP to boot off of. ![]() The ESP can only be installed on one partition on system. I'm curious as to how people set up their partitions in an EFI system with software RAID.
0 Comments
Leave a Reply. |