معمل 18.1: Break and repair the boot

هذا الجزء لم يُترجم بعد. أنت ترى النص الإنجليزي.

الهدف

Practice the three repairs that matter most: a lost root password, a broken fstab, and a damaged initramfs.

Take a VM snapshot of server1 first. You need console access (not SSH).
You need the file system /data from lab 14.1 (or any extra fstab line you can break).

Reset the root password to redhat without knowing the old one.
Add a wrong line to /etc/fstab (for example a mount of a UUID that does not exist, without nofail) and reboot. Repair the system from emergency mode so it boots normally.
Boot into rescue mode once with a kernel argument, check which units run, and continue to the normal target.
Make a backup of the initramfs of the running kernel, delete the original, and rebuild it with dracut. Check that the system still boots.

#su - root
#findmnt --verify
#lsinitrd | head -5
#ls -l /boot/initramfs-$(uname -r).img

You can log in as root with redhat (and SELinux is enforcing: getenforce).
The system boots to the default target without errors; findmnt --verify is clean.
The initramfs exists again and the system boots.

اعرض التلميحات

اعرض الحل

Root password (at the GRUB menu: e, add rd.break to the linux line, Ctrl+X).

#mount -o remount,rw /sysroot
#chroot /sysroot
#echo redhat | passwd --stdin root
#touch /.autorelabel
#exit
#exit

Break and fix fstab.

#echo "UUID=00000000-1111-2222-3333-444444444444 /broken xfs defaults 0 0" >> /etc/fstab
#systemctl reboot

In emergency mode (log in as root):

#mount -o remount,rw /
#sed -i '/\/broken/d' /etc/fstab
#systemctl daemon-reload
#mount -a
#systemctl reboot

Rescue mode (at GRUB add systemd.unit=rescue.target).

#systemctl list-units -t service --state=running
#systemctl default

Rebuild the initramfs.

#cp /boot/initramfs-$(uname -r).img /root/initramfs.bak
#rm /boot/initramfs-$(uname -r).img
#dracut -f
#lsinitrd | head -5
#systemctl reboot

لا توجد ملاحظات بعد.