I have done upgrade in place (no new harddrive unless we needed a larger 
capacity drive) on several unix/bsd derivatives.  Your file system 
comments are very well taken.   However, using "stock" Ubuntu LTS for 
the OS and file system, is your experience contrary to those of others?

On 5/26/20 7:21 PM, Nico Kadel-Garcia wrote:
> On Tue, May 26, 2020 at 3:54 PM Troy Dawson <[log in to unmask]> wrote:
>> Although in the past the official policy of Red Hat was that you needed to do a fresh install going from EL N to N+1, that is starting to change.
>> There is an internal team called "LEAP" whose job it is to make sure you can do that.
>> I believe RHEL 7.8 to RHEL 8.1 was the first that you could officially do that.
>> I don't know all the details.  all I know is "we're working on it."
>> I'm pretty sure that at the current time we aren't as smooth as Debian, they've been doing it much longer.
>> But we're getting better, and for RHEL9, LEAP is being involved from the start.
>> So maybe in a few releases / years people will be able to say our updates are as good, or even easier than debians.
> Been there, done that. It works great until it doesn't. While updating
> the RPM's may be feasible, there have been subtle changes in
> filesystems which man starting with an old filesystem is prone to
> errors, and upgrading in place is.... not a reliable process. *All*
> operating systems are prone to such issues, unless you can basically
> mount the old file system as an image and apply the updates from
> outside. I've done upgrades with approximately 20,000 Red Hat based
> systems, over my career, and others. If you were foolish enough to use
> ReiserFS, for example, you'll *really* need to rebuld your filesystems
> in between OS upgrades..