*heavy sigh*
Something broke on my Debian install. My research indicated it was just one of those things, but it made things unworkable for me.
So I tested Linux Mint but it does not like this hardware. So I ended up back on CentOS because I found out how to solve the problem that first drove me off of it. Back where I started.
I’m praying sincerely that this wrestling match is over and I can get back to work.
Update: It was not to be. However, this time the crash-n-burn of CentOS was far more instructive. I was worried some portion the hardware was failing, but the diagnostics from CentOS indicated something else.
So I went into the BIOS. A week or so ago I had to install Windows briefly so I could update the BIOS, which was waaaaay out of date. It’s one of the unsolved mysteries of computer tech support that the manufacturer manages to change BIOS settings when updating, and does not even give you the defaults in the process. Several things were out of whack. Once I restored them, CentOS was already broken.
However, it’s no secret that I vastly prefer Debian in the first place. So I’m updating this post from my netbook while I reinstall Debian on the workstation. And it’s going smoothly.
Thank you for your prayers.
Sorry you had to deal with this pain point. I’m guessing you had sufficient backups that you could get back to being productive without much delay once the final OS was completed.
Which leads into my next thought. I’m guessing a lot of people aren’t as prepared as you to move off their old hard drive if they had to, temporarily or otherwise. I know I’m not.
Do you back things up? Or do you just store the first copy externally to begin with? Is there any special software you use to keep the backups current?
There was a time when it was my job to test many different OSes for a magazine and write up reviews of my experience. I learned very early in the game the habit of picking out what was worth keeping. At first it was burning them to CD, but when jump drives got cheap enough, I used those. Now I use a rather large external drive designed for laptops. Mostly I copy how directories as-is. In Linux, a lot of user configuration information is saved in files under their home directory and I simply copy the whole collections of that, too. Thus, I am not starting from scratch with reconfiguring all of my favorite applications. I think it’s pretty sad that those same applications under Windows do not have such a simple structure for configuration options. It’s highly fragmented and you’d have to really know your stuff to do that with Windows.
Finally, each Linux distribution has their own packaging conventions and ways of breaking stuff up for installation. I keep a written list of what I use by package name from the three or four Linux distros that I can tolerate. Adding back the stuff I commonly used is easier that way.
For those packages you like and keep a written list of, do the packages have a certain version number you like that you go back to, or do you just install the latest release? (E.g web browser or photo editor, or word processor, etc)
And (changing subjects slightly) for the Windows systems you support, how would you advise people to deal with updates from MS? Do none of the updates and just rely on virus and malware software? Or do some or all of the updates based on some criteria you’ve developed?
I keep a list of the Linux packages by name because different versions of Linux use different naming conventions. Going back to a previous version number is pretty rare for me when it comes to those packages, and not very easy to do most of the time. I’m more likely to do that with something like MS Word, where upgrades are not really any improvement for most common uses, and the older version still works just fine if you know how to do it. In Linux land, there isn’t much like that. On the other hand, different versions of Linux fail to include some things that I end up building from source for myself. Example: Debian has no Bibletime package for the current release. It’s not that hard to build if you follow the instructions. If I really have to have something older, I can usually build it to suit me, and it’s often not available otherwise.
The advice I give Windows users about updates varies with the user’s level of expertise and what matters to them. If you really know what you are doing, I recommend you don’t use Windows Update at all, but that you track the security updates only, using something like Autopatcher. Other “recommended” updates are a matter of researching what they do and waiting to see if they actually help anything. There are experts who track such things and they post their results online in forums, like with the Autopatcher forum. There was a time I followed all of that myself, but I gave up when Linux became more useful to me than Windows. You’d be surprised how many large commercial businesses have a tight IT policy about Windows updates. The reasons vary widely. Sometimes it’s because MS tends to break things with their own sloppy coding. Sometimes those large commercial operations tend to create massive in-house software projects that aren’t well done, and maybe in part because Windows stuff is clunky, or simply because the feel obliged to use Windows while having zero trust in MS.
Then again, I recommend you always run Windows from a less-privileged account. The vast majority of malware damage comes from running the primary user account. Go ahead and install with your “master” user, then create an unprivileged account (non-administrator) later and use that for most things. That’s how we do it with Linux.