Rebooting an OVH VPS in rescue mode to repair it

1. Go to the OVHcloud control panel and to the screen that reboots the VPS in rescue mode

After a few minutes, an e-mail arrives with the login (root) and the password of the rescue system (unless you registered your SSH key somewhere in the manager):

2. Connecting over SSH

Careful, the login in rescue mode is root!

2.1 First attempt from the console of your Linux PC (or other exotic system)

bash

ssh root@vps-xxxxxxx.vps.ovh.net

You get:

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
@    WARNING: REMOTE HOST IDENTIFICATION HAS CHANGED!     @
@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
IT IS POSSIBLE THAT SOMEONE IS DOING SOMETHING NASTY!
Someone could be eavesdropping on you right now (man-in-the-middle attack)!
It is also possible that a host key has just been changed.
The fingerprint for the ECDSA key sent by the remote host is
SHA256:aHD6pBYP1zHxxxxxxx7hp4KwI4yl5gr5OVmBJZ/ZBE.
Please contact your system administrator.
Add correct host key in /home/kiwiof09/.ssh/known_hosts to get rid of this message.
Offending ECDSA key in /home/coin/.ssh/known_hosts:75
  remove with:
  ssh-keygen -f "/home/coin/.ssh/known_hosts" -R "vps-xxxxxxx.vps.ovh.net"
ECDSA host key for vps-xxxxxxx.vps.ovh.net has changed and you have requested strict checking.
Host key verification failed.

2.2 On your workstation, remove the host key fingerprints of the server

bash

ssh-keygen -f "/home/coin/.ssh/known_hosts" -R "vps-xxxxxxx.vps.ovh.net"
ssh-keygen -f "/home/coin/.ssh/known_hosts" -R "51.38.yyy.yyy"
ssh-keygen -f "/home/coin/.ssh/known_hosts" -R "2001:xxxx:xxx:xxxx::xxxx"

2.3 New connection attempt

bash

ssh root@vps-xxxxxxx.vps.ovh.net

You get:

The authenticity of host 'vps-xxxxxxx.vps.ovh.net (51.38.xxx.xxx)' can't be established.
ECDSA key fingerprint is SHA256:aHD6pBYP1zxxxxxxxKwI4yl5gr5OVmBJZ/ZBE.
Are you sure you want to continue connecting (yes/no)? yes
Warning: Permanently added 'vps-xxxxxxx.vps.ovh.net,51.38.xxx.xxx' (ECDSA) to the list of known hosts.
root@vps-xxxxxxx.vps.ovh.net's password:
Linux vps-xxxxxxx 4.9.0-11-amd64 #1 SMP Debian 4.9.189-3+deb9u2 (2019-11-11) x86_64

GNU/Linux, to the rescue!

Server: vps-xxxxxxx
Login: root
Password: azezeazeaeazezae

root@vps-xxxxxxx:~#

2.4 List the disks

bash

lsblk

NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINT
sda 8:0 0 2.5G 0 disk
└─sda1 8:1 0 2.5G 0 part /
sdb 8:16 0 20G 0 disk
└─sdb1 8:17 0 20G 0 part

You can see that the rescue disk is sda1 and that it is mounted as the root (/), but the disk of the server is sdb1!

2.5 Change your root directory so you can work inside the server

bash

mkdir -p /mnt/sdb1
mount /dev/sdb1 /mnt/sdb1
mount -t proc proc /mnt/sdb1/proc/
mount -t sysfs sys /mnt/sdb1/sys/
mount -o bind /dev /mnt/sdb1/dev/
mount -t devpts devpts /mnt/sdb1/dev/pts
mkdir -p /run/chroot/lock
mount -o bind /run/chroot /mnt/sdb1/run
chroot /mnt/sdb1 /bin/bash
source /etc/default/locale

2.6 List the disks again

bash

lsblk

NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINT
sda 8:0 0 2.5G 0 disk
└─sda1 8:1 0 2.5G 0 part
sdb 8:16 0 20G 0 disk
└─sdb1 8:17 0 20G 0 part /

2.7 You can now repair the server

If you run into trouble, ask for a quote.

2.8 Leave the environment and go back to the root of the rescue system

bash

exit
umount /mnt/sdb1/run
umount /mnt/sdb1/dev/pts
umount /mnt/sdb1/dev/
umount /mnt/sdb1/sys/
umount /mnt/sdb1/proc/
umount /mnt/sdb1

Go back to the OVH manager to leave rescue mode

And reboot the server in normal mode: