Hello everyone.
I'm trying to install CentOS 8 with root and swap partitions on
software raid. The plan is:
- create md0 raid level 1 with 2 hard drives: /dev/sda and /dev/sdb,
using Linux Rscue CD,
- install CentOS 8 with Virtual Box on my laptop,
- rsync CentOS 8 root partition on /dev/md0p1,
- chroot in CentOS 8 root partition,
- configure /etc/mdadm.conf, grub.cfg, initramfs, install bootloader on
both sda and sdb drives.
I think I can do first four of the above, but my CentOS installation
acts strange after rebooting the server. It recognizes the raid, but
boots randomly with root on /dev/sda1 (and recognizes raid
with /dev/sdb disk), or with root on /dev/sdb1 (and recognizes raid
with /dev/sda disk). When booting from Linux Rescue CD, the raid with
two disk is recognized.
Creating md0 wth Linux Rescue CD.
# lsblk
NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINT
loop0 7:0 0 819.5M 1 loop /run/archiso/sfs/airootfs
sda 8:0 0 232.9G 0 disk
sdb 8:16 0 232.9G 0 disk
sr0 11:0 1 871M 0 rom
# mdadm --create /dev/md0 --level=1 --raid-devices=2
--metadata=0.9 /dev/sda /dev/sdb
mdadm: array /dev/md0 started.
# mdadm --detail --scan --verbose
ARRAY /dev/md0 level=raid1 num-devices=2 metadata=0.90 UUID=fef2f68b:
1a462c80:c44c77eb:7ee19756 devices=/dev/sda,/dev/sdb
# cat /proc/mdstat
Personalities : [raid1]
md0 : active raid1 sdb[1] sda[0]
244198464 blocks [2/2] [UU]
[>....................] resync = 1.9% (4682944/244198464)
finish=52.2min speed=76356K/sec bitmap: 2/2 pages [8KB], 65536KB chunk
unused devices: <none>
(1 hour later)
# cat /proc/mdstat
Personalities : [raid1]
md0 : active raid1 sdb[1] sda[0]
244198464 blocks [2/2] [UU]
bitmap: 0/2 pages [0KB], 65536KB chunk
unused devices: <none>
# lsblk
NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINT
loop0 7:0 0 819.5M 1 loop /run/archiso/sfs/airootfs
sda 8:0 0 232.9G 0 disk
└─md0 9:0 0 232.9G 0 raid1
sdb 8:16 0 232.9G 0 disk
└─md0 9:0 0 232.9G 0 raid1
sr0 11:0 1 871M 0 rom
---------------------------------
So far so good. Now partitions:
# fdisk /dev/md0
Welcome to fdisk (util-linux 2.33.1).
Changes will remain in memory only, until you decide to write them.
Be careful before using the write command.
Device does not contain a recognized partition table.
Created a new DOS disklabel with disk identifier 0x64998720.
Command (m for help): o
Created a new DOS disklabel with disk identifier 0xfe784f57.
Command (m for help): n
Partition type
p primary (0 primary, 0 extended, 4 free)
e extended (container for logical partitions)
Select (default p): p
Partition number (1-4, default 1):
First sector (2048-488396927, default 2048):
Last sector, +/-sectors or +/-size{K,M,G,T,P} (2048-488396927, default
488396927): +220G
Created a new partition 1 of type 'Linux' and of size 220 GiB.
Command (m for help): n
Partition type
p primary (1 primary, 0 extended, 3 free)
e extended (container for logical partitions)
Select (default p): p
Partition number (2-4, default 2):
First sector (461375488-488396927, default 461375488):
Last sector, +/-sectors or +/-size{K,M,G,T,P} (461375488-488396927,
default 488396927): +4G
Created a new partition 2 of type 'Linux' and of size 4 GiB.
Command (m for help): w
The partition table has been altered.
Calling ioctl() to re-read partition table.
Syncing disks.
# lsblk
NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINT
loop0 7:0 0 819.5M 1 loop /run/archiso/sfs/airootfs
sda 8:0 0 232.9G 0 disk
└─md0 9:0 0 232.9G 0 raid1
├─md0p1 259:0 0 220G 0 part
└─md0p2 259:1 0 4G 0 part
sdb 8:16 0 232.9G 0 disk
└─md0 9:0 0 232.9G 0 raid1
├─md0p1 259:0 0 220G 0 part
└─md0p2 259:1 0 4G 0 part
sr0 11:0 1 871M 0 rom
# mkfs -t ext4 /dev/md0p1
mke2fs 1.44.5 (15-Dec-2018)
Creating filesystem with 57671680 4k blocks and 14417920 inodes
Filesystem UUID: 511d0d07-4d55-4b6c-939e-d23ae2da1ce0
Superblock backups stored on blocks:
32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632,
2654208, 4096000, 7962624, 11239424, 20480000, 23887872
Allocating group tables: done
Writing inode tables: done
Creating journal (262144 blocks): done
Writing superblocks and filesystem accounting information: done
# mkswap /dev/md0p2
Setting up swapspace version 1, size = 4 GiB (4294963200 bytes)
no label, UUID=5667d007-0d3e-4b54-8a72-6ade52609adc
--------------------------------------
Mount would be root partition:
# mkdir /mnt/centos8
# mount -t ext4 /dev/md0p1 /mnt/centos8/
--------------------------------------
OK, I have what I need. Now log into virtualised CentOS 8 in Virtual
Box.
# rpm -q rsync
rsync-3.1.3-7.el8.x86_64
# rpm -q mdadm
mdadm-4.1-13.el8.x86_64
# rsync -vaxAXHSz --delete / 192.168.111.167:/mnt/centos8/
The authenticity of host '192.168.111.167 (192.168.111.167)' can't be
established. ECDSA key fingerprint is
SHA256:HukswAQlarthYqbpVHIHm80JaWHKX12AxgOn8UuKQxk. Are you sure you
want to continue connecting (yes/no/[fingerprint])? yes Warning:
Permanently added '192.168.111.167' (ECDSA) to the list of known hosts.
root@192.168.111.167's password: sending incremental file list
./
bin -> usr/bin
lib -> usr/lib
lib64 -> usr/lib64
sbin -> usr/sbin
[lots and lots of files...]
var/yp/
sent 694,980,592 bytes received 745,802 bytes 1,795,422.95 bytes/sec
total size is 1,558,934,091 speedup is 2.24
--------------------------------------
Chrooting time:
# mount --types proc /proc /mnt/centos8/proc
# mount --rbind /sys /mnt/centos8/sys
# mount --make-rslave /mnt/centos8/sys
# mount --rbind /dev /mnt/centos8/dev
# mount --make-rslave /mnt/centos8/dev
# chroot /mnt/centos8 /usr/bin/bash
# source /etc/profile
Configure raid:
# mdadm --detail --scan --verbose
ARRAY /dev/md0 level=raid1 num-devices=2 metadata=0.90 UUID=fef2f68b:
1a462c80:c44c77eb:7ee19756 devices=/dev/sda,/dev/sdb
# mdadm --detail --scan --verbose > /etc/mdadm.conf
Edit /etc/fstab:
# cat /etc/fstab
#
# /etc/fstab
# Created by anaconda on Sun Nov 8 10:05:31 2020
#
# Accessible filesystems, by reference, are maintained under
# '/dev/disk/'. See man pages fstab(5), findfs(8), mount(8) and/or
# blkid(8) for more info.
#
# After editing this file, run 'systemctl daemon-reload' to update
# systemd units generated from this file.
#
#UUID=36c5a887-79c3-4b6a-b433-0e4d7574b1e4 /
#ext4 defaults 1 1
/dev/md0p1 / ext4 defaults 1 1
Configure and install bootloader:
# cat /etc/default/grub
GRUB_TIMEOUT=8
GRUB_DISTRIBUTOR="$(sed 's, release .*$,,g' /etc/system-release)"
GRUB_DEFAULT=saved
GRUB_DISABLE_SUBMENU=true
GRUB_TERMINAL_OUTPUT="console"
GRUB_CMDLINE_LINUX="rhgb quiet"
GRUB_DISABLE_RECOVERY="true"
GRUB_ENABLE_BLSCFG=true
# grub2-mkconfig -o /boot/grub2/grub.cfg
Generating grub configuration file ...
done
# grub2-install /dev/sda
Installing for i386-pc platform.
Installation finished. No error reported.
# grub2-install /dev/sdb
Installing for i386-pc platform.
Installation finished. No error reported.
Configure initramfs:
# dracut -f /boot/initramfs-4.18.0-193.28.1.el8_2.x86_64.img
# 4.18.0-193.28.1.el8_2.x86_64
--------------------------------
Reboot and login to my new CentOS 8 installation:
# lsblk
NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINT
sda 8:0 0 232.9G 0 disk
sdb 8:16 0 232.9G 0 disk
├─sdb1 8:17 0 220G 0 part /
└─sdb2 8:18 0 4G 0 part
sr0 11:0 1 871M 0 rom
# mdadm --detail --scan --verbose
INACTIVE-ARRAY /dev/md127 num-devices=1 metadata=0.90 UUID=fef2f68b:
1a462c80:c44c77eb:7ee19756 devices=/dev/sda
# cat /proc/mdstat
Personalities :
md127 : inactive sda[0](S)
244198464 blocks
unused devices: <none>
I can live with md127, because it happens constantly on multiple
distributions, but why root is on /dev/sdb1 (and sometimes
it's /dev/sda1)? I'm running out of spells now. :) Do anyone can
suggest what else I forgot to do?
thanks
--
Łukasz Posadowski
_______________________________________________
CentOS mailing list
CentOS@xxxxxxxxxx
https://lists.centos.org/mailman/listinfo/centos
