Forewarning- I tried to ask ChatGPT to help me with this post but it didnt understand and kept misbehaving so I had to forego that for this post. Also I dont normally write about techy stuff so I hope I got it all in the right order.

I use a VPS I got on a black friday sale from Rack Nerd and it has been solid for uptime and everything. I run my a few servers up there including my atuin, ntfy, a budgeting app, and some other random one off stuff. I run the nextcloud and other larger storage things at home thanks to Google Fiber/WebPass.

Most of the time I use it as a jump server to connect over to SDF or to wherever else online I’m going and so I dont see my MOTD. A few weekends back I was setting up an Actual Server aka the budgeting app and I saw that I was out of date and Ubuntu 24.01 was avaialble. Because I’m so good at planning I started up the upgrade and then saw I needed to get to the movies. I hit control c and closed the window. Don’t do that!

So I left everything running for a month or so and decided today I would get it going. First I tried the simple sudo apt update && sudo apt upgrade Update was happy but upgrade was mad. It suggested maybe I should run sudo apt --fix-broken install which then also told me to try sudo apt-get clean from there I was off to startpage to see if I could find something. I tried sudo dpkg --configure -a and I tried sudo apt upgrade --fix-missing and do apt-get -f install at this point I was starting to get a little desperate. How desperate? Messing with some system level files desperate!

I found a Stack Overflow question and the answers suggested moving the dat files from /var/cache/debconf to a backup directory and then trying to upgrade again. That failed but told me that everything failed because it depended on systemd-sysv. Oooh progress!

At this point when I ran the apt update and then apt upgrade I got a warning about the shadow group cdrom already existing. so I edited the /etc/gshadow file to remove it. Then it was another group. I noticed that they were all groups that had ubuntuf in the last field. I put the cdrom entry back, duplicated the file, and then deleted all the ubuntufs in there. Then I ran sudo dpkg --configure -a Hey it finished!

Do you want to continue? [Y/n] y
(Reading database ... 147882 files and directories currently installed.)
Preparing to unpack .../systemd-sysv_255.4-1ubuntu8.4_amd64.deb ...
Unpacking systemd-sysv (255.4-1ubuntu8.4) over (249.11-0ubuntu3.12) ...
Setting up systemd-sysv (255.4-1ubuntu8.4) ...
Processing triggers for man-db (2.10.2-1) ...
needrestart is being skipped since dpkg has failed

I switched over to root to run these because I was done typing sudo sudo sudo so much

Let’s try the apt update .. so far so good

Lets try upgrade .. no not quite there yet

Ok how about apt --fix-broken install

Wooo it was installing like crazy.

It wasn’t quite done though. I ran apt update and we still had quite a few packages to work on.

One more apt upgrade and then it was time to restart.

After reboot we were on 24.01. Now I need to upgrade postgresql eventually…