After the better part of the last decade running Arch Linux as my primary distro
of choice on my laptops, I’ve decided to hang it up and move back to Debian.
That’s not to say that there’s anything wrong with Arch Linux. In fact, it’s
been one of my favorite distros and I highly recommend it for anybody that wants
to learn more about Linux (since Arch does make you get dirty) or if you always
want to be on the latest and greatest version of things.
If I like it so damned much, why would I consider leaving? A few reasons:
My needs have changed
What brought me to Arch Linux in the first place, bleeding edge software, just
doesn’t interest me as much as it once did.
Arch, by it’s very nature, comes with a lot of what I like to call “update
fatigue”. I’m running a rolling release Linux distribution, of course I want to
keep it up to date. Having the Arch Update GNOME Shell extension adds to the
FOMO by not updating.
Days would go by and hundreds of package updates would become available. Some
days I’d update multiple times a day. Others, I’d run an update of over 200
packages, just to have another 100 show up within a few hours.
Sure, I could very well stop updating my system. I’ve done this a few times, and
while it’s sufficient enough in the short term, I always feel like I’m probably
missing out on critical security updates in doing so. Also, while usually not
catastrophic, waiting extended periods of time between updates does come with
it’s own share of additional problems that don’t exist when you update
Fortunate, the software that I like running the latest version of, GNOME, has
hit an inflection point in my opinion, in terms of how many huge changes we’re
seeing with each new version.
That’s not to say that the GNOME desktop environment has grown stale. Quite the
opposite, it feels extremely mature, and a lot of the major updates seem to be
more polish than major functionality.
That’s a great thing, because now I can flow my roll a bit, and just hang out
with the current version of GNOME and not worry too much about getting onto the
More on why GNOME matters in a bit.
I’ve never actually fully committed to Arch Linux
This has always been a sore spot for me. While I love Arch Linux, I’ve only ever
committed to using it on my personal computers, and never for servers.
Fake fan, I know.
The reason for this, is that it’s always been in the back of my head, that Arch
would be a nightmare to run on a server, and I’d probably run an update and be
left with a machine I can no longer access.
Sure, that could reasonably happen with any distro, but with years of
conditioning of having to jump through quite a few hoops after updating my Arch
Linux system, and fueling quite a few “how to fix X on Arch” posts on my blog, I
never could get over the line mentally to run Arch on my servers.
Now over the years, I’ve run Ubuntu servers, while I was running Ubuntu as my
daily driver. I then migrated from Ubuntu to Debian on my servers, again, while
I was running Debian as my daily driver. Then when I switched to Arch, I
remained loyal to Debian for my server needs.
Migrating from Arch Linux to Debian Stable
While I have my reasons for switching, I did need a plan. The first thing I did
was boot up a Debian Stable live disk, and kicked the tires a bit.
It was as I remembered it, but something immediately stuck out.
The version of GNOME shipping with Debian Stable was old and tired.
The problem wasn’t GNOME itself, it was the fact that I’ve used the newer
versions of GNOME and they are fantastic enough that the older version had some
noticeable shortcomings. Just a perception problem that I had to work through.
Knowing that Debian Stable’s GNOME was going to be a bit too dated, I went ahead
and repeated the process with a Debian Testing live disk. GNOME was the same, or
close to the same version I had running in Arch Linux. Good to go.
With my suspicions confirmed, I proceeded to get Debian Testing up and running
on my machine. That’s something I’ve done a hundred times, so no big deal there.
What comes next may surprise you.
Instead of pinning
testing, I opted to pin it to
codename of the next major release of Debian.
The logic there is that in the future, when Debian 12 comes out, I want to stop
running Debian Testing and start running Debian Stable, and then staying on
Debian Stable with the roughly every 2 year release cycle.
This allows me to have a smooth transition from Arch Linux to Debian, with a
future consideration of being able to move to Debian Stable from Testing,
without being influenced by any usage of a newer version of GNOME.
So far so good, as I’ve been running Debian Testing for a few weeks now, and
really couldn’t be happier. Even on the Testing branch, the volume of package
updates have been extremely modest.
I did run into a few small issues with installing a few things. While I was
scared of how bad life would be without the Arch User Repository (AUR), it
really didn’t take much to source the couple of packages that I needed that
didn’t exist in the official Debian package list.