Upgrading from Ubuntu 20.04 to 22.04 and get errors..

[swheeler78]

I was on Ubuntu 20.04 and went thru the upgrade to 22.04.
When I finished I ran dpkg --configure -a as instructed in the howto. Now I can't upgrade to 24.04 due to the following errors.

Setting up libnginx-mod-http-dav-ext (1.18.0-6ubuntu14.7) ...
ln: failed to create symbolic link '/etc/nginx/modules-enabled/50-mod-http-dav-ext.conf': No such file or directory
dpkg: error processing package libnginx-mod-http-dav-ext (--configure):
 installed libnginx-mod-http-dav-ext package post-installation script subprocess returned error exit status 1
Setting up libnginx-mod-http-upstream-fair (1.18.0-6ubuntu14.7) ...
ln: failed to create symbolic link '/etc/nginx/modules-enabled/50-mod-http-upstream-fair.conf': No such file or directory
dpkg: error processing package libnginx-mod-http-upstream-fair (--configure):
 installed libnginx-mod-http-upstream-fair package post-installation script subprocess returned error exit status 1
dpkg: dependency problems prevent configuration of nginx-full:
 nginx-full depends on libnginx-mod-http-dav-ext; however:
  Package libnginx-mod-http-dav-ext is not configured yet.
 nginx-full depends on libnginx-mod-http-upstream-fair; however:
  Package libnginx-mod-http-upstream-fair is not configured yet.

dpkg: error processing package nginx-full (--configure):
 dependency problems - leaving unconfigured
Setting up libnginx-mod-http-auth-pam (1.18.0-6ubuntu14.7) ...
ln: failed to create symbolic link '/etc/nginx/modules-enabled/50-mod-http-auth-pam.conf': No such file or directory
dpkg: error processing package libnginx-mod-http-auth-pam (--configure):
 installed libnginx-mod-http-auth-pam package post-installation script subprocess returned error exit status 1
Setting up libnginx-mod-http-echo (1.18.0-6ubuntu14.7) ...
ln: failed to create symbolic link '/etc/nginx/modules-enabled/50-mod-http-echo.conf': No such file or directory
dpkg: error processing package libnginx-mod-http-echo (--configure):
root@vultr:~# /home/yellowtent/box/scripts/init-ubuntu.shtallation script subprocess returned error exit status 1ectoryry

Any help would be appreciated.

[SansGuidon]

My 2 cents, I never upgraded the same VM with Cloudron running on it, I just bootstraped a new VM with a fresh upgraded Ubuntu and migrated Cloudron to it ^^ , which allows me to test safely.

Reason: too many topics about problematic Ubuntu upgrades that break things or Cloudron

[james]

Hello @swheeler78
I think I've seen something like this before.
Since on Cloudron servers /etc/nginx/ does not "exist" as a real folder but is a symlink to /etc/nginx -> /home/yellowtent/platformdata/nginx I've seen it happen that for some reason the symlink gets removed and a real /etc/nginx/ folder gets created.
You can simply start the box.service once and stop it again, then /etc/nginx should be a symlink again and the dpkg --configure -a should work.

[jdaviescoates]

@SansGuidon said in Upgrading from Ubuntu 20.04 to 22.04 and get errors..:

My 2 cents, I never upgraded the same VM with Cloudron running on it, I just bootstraped a new VM with a fresh upgraded Ubuntu and migrated Cloudron to it ^^ , which allows me to test safely.

Reason: too many topics about problematic Ubuntu upgrades that break things or Cloudron

Yeah, that's mostly what I've always done too. Not sure what I'll do next now that I'm on a dedicated server though...

[james]

Hello @SansGuidon and @jdaviescoates
That is indeed the most reliable / safest approach.
I have upgraded many Cloudron Ubuntu's with the provided guides and rarely ran into issues.
But when something comes up, like this case, a somewhat deeper understanding of Cloudron is valuable to resolve it.

---

@jdaviescoates If you upgrade your Ubuntu version on your dedicated server with the provided guides, would you be willing to document your process in a forum topic?
If anything is unclear, questionable or breaks having this documented process of yours would be a great help to further improve the guides.

[jdaviescoates]

@james should that day come, sure!

[swheeler78]

Forgive me for my lack of knowledge but how do I start the box.service. Also, I've already spun up another VM and I'll be migrating to it. I'm assuming it's as simple as installing a fresh copy of ubuntu and then installing Cloudron and once logged it doing a restore.

[james]

Hello @swheeler78
You can restart the box.service with:

systemctl restart box.service

[joseph]

@swheeler78 https://docs.cloudron.io/backups/#restore-cloudron has the workflow .

[swheeler78]

Thank you.

[manuel]

Hello @girish i am also going to migrate from ubuntu 20.04 to 22.04. i am reading your guide and it looks very clear. thanks! there only 2 things i don't understand , it's on the post updgrade section. there are a couple of steps i am not sure i get how to do
are these the full commands ?

*Install missing packages: /home/yellowtent/box/scripts/init-ubuntu.sh

*Fix package configuration: /home/yellowtent/box/setup/start.sh

Ubuntu 22 uses cgroups v2 by default. Recreate all Docker containers to recognize this change: /home/yellowtent/box/scripts/recreate-containers. This script is part of Cloudron 7.3. For earlier versions, download it here.

[james]

Hello @manuel

Yes these are command that you can copy-paste into your terminal.
I will update the guides to make it more clear.

[manuel]

@james thanks!