DIY install debian on Oracle Cloud Infrastructure ( Free Tier )

reinstall_VPS_from_inside.sh

# should works on any cloud-init enabled hypervisor (openstack.. )

# start from a ubuntu minimal install
# we need to shrink down the used space to move it in a tmpfs of 700MB

# make sure we are on the highest kernel, so we can delete all the others ...
sudo apt update && sudo apt upgrade -y && reboot

# ... reconnect
sudo apt install lsof
sudo snap remove --purge  oracle-cloud-agent && sudo snap remove --purge core18 && sudo snap remove --purge snapd && sudo sudo apt -y purge snapd && sudo rm -rf ~/snap /snap /var/cache/snapd
sudo apt purge $(dpkg-query -Wf '${Package}\n' | grep header)  $(apt list --installed | grep -oP "^linux.*\d\d\d\d-oracle" | grep -v "$(uname -r)") linux-modules-extra-$(uname -r) lxc* lxd* vim* && sudo apt -y autoremove && sudo apt -y autoclean && sudo apt -y clean  
sudo rm -rf /var/log/* /var/lib/apt/lists/*


# ok, the disk usage of / should be less than 600MB now

cd /
mount -t tmpfs -o size=700m tmpfs mnt
tar --one-file-system -c . | tar -C /mnt -x
mount --make-private -o remount,rw /
mount --move dev mnt/dev
mount --move proc mnt/proc
mount --move run mnt/run
mount --move sys mnt/sys
sed -i '/^[^#]/d;' mnt/etc/fstab
echo 'tmpfs / tmpfs defaults 0 0' >> mnt/etc/fstab
cd mnt
mkdir old_root
mount --make-private /
unshare -m
pivot_root . old_root

# now the root storage is the RAM 

# lets restart all runnig process

service ssh restart 

# now switch to a new ssh connexion to let the old sshd terminate
 
service systemd-udevd restart
service systemd-journald restart
service systemd-networkd restart
service systemd-timesyncd restart
service systemd-resolved restart
service systemd-logind restart
service networkd-dispatcher restart
pkill agetty
pkill dbus-daemon
pkill atd
pkill iscsid
pkill rpcbind
pkill unattended-upgrades
kill 1 

# check with "lsof /old_root" that there is no remaining process 

umount -l /dev/sda1

# check :

df -h 
lsblk 

# the disk should be unmount 
# now, just copy the debian cloud image on the disk.

# curl https://cdimage.debian.org/cdimage/openstack/current/debian-10-openstack-amd64.raw | dd of=/dev/sda bs=1M

# update : 
curl -L  https://cdimage.debian.org/cdimage/openstack/testing/debian-testing-openstack-amd64.raw | dd of=/dev/sda bs=1M

sync

reboot

I would like to say many thanks for the author for all his effort to contribute with this tutorial ..

But I had to try a few images before getting to the right one .. so I'll leave here to those who wants to try: Canonical-Ubuntu-18.04-Minimal-2023.05.10-0

But I reached a problem and I can't get pass it:

root@debian-amd:/# mount --move run mnt/run
mount: /mnt/run: filesystem was mounted, but failed to update userspace mount table.

It gives me this error, but it allows me to continue until...

root@debian-amd:/mnt# unshare -m
mesg: ttyname failed: No such file or directory

When I restart ssh .. it restarts just fine, but when I try to connect, it accepts my login credentials, it stucks and doesn't bring the command line ability .. it just gives me the intro message and stops...

But I can continue on my old terminal plus I opened the console from Oracle and both still works .. let's continue

but unfortunately sudo doeesn't work on the command anymore and none of the following instructions also works ..

service systemd-udevd restart
service systemd-journald restart

I've been trying for hours to install debian on Oracle .. I also don't have the image converter function anymore, it only works on the first 30 days then it's gone ... and I don't want to convert my Free account into a PAYG and I'll never be able to downgrade ..

Thanks for the feedback and sorry to hear it didn't work for you. I see your image name is "debian-amd". Could you please cross-check whether you are using the right arch image (i.e. arm64 on arm64 instance and amd64 on arm64)? Also, please confirm that cloud-init is enabled on the instance before the procedure.

If you are looking for arm64, you can try out my gist which seems to work on debian 12 as well: https://gist.github.com/4abhinavjain/893ec13c651bee08088c8f4661998952

I was thinking .. maybe instead of getting all / into ram .. why not just attach a new storage? I tried to attach and it creates a /dev/sdb .. but I have no knowledge to go further =/

Another possible way (which I haven't tried yet) is to boot the instance in recovery mode, mount the partition and use dd command to overwrite the raw image onto the partition.